Railroads in historical context vol. I

RAILROADS IN HISTORICAL CONTEXT:

vol.I 2011 ANNE MCCANTS EDUARDO BEIRA JOSÉ M. LOPES CORDEIRO PAULO B. LOURENÇO (eds.)

RAILROADS IN HISTORICAL CONTEXT: vol.I 2011

ANNE MCCANTS EDUARDO BEIRA JOSÉ M. LOPES CORDEIRO PAULO LOURENÇO (eds.)

ISBN: 978-989-97134-5-1 Graphic design and layout, and cover design, by Ana Prudente.

Samantha Evaristo contributed with revision and editing of the texts from non english authors.

Edited and printed by Inovatec (Portugal) Lda. (V. N. Gaia, Portugal).

Cover printing and book binding by Minerva – Artes Gråficas, Lda. (Vila do Conde, Portugal).

PREFACE

The book you hold in your hands is an eclectic one. It represents the work of scholars from a broad array of disciplines: from oral history to structural engineering, from economics to museum design, from digital geographic information systems to the history of public policy, corporate finance, firm management, and engineering education. Some might find such breadth daunting, or perhaps tedious, or fear that the intellectual value of any resulting product must be diminished for being so diffuse. But we hope to persuade you otherwise. For this volume was the outcome of a rather extraordinary academic meeting that brought together over thirty scholars from all these fields and yet others, to devote three days to a sustained, collective, and truly interdisciplinary conversation on the multiple aspects of a single phenomenon. Our object of focus was the historical (and indeed on-going contemporary) experience of what was itself an extraordinary engineering project in its time: the construction of the first fully Portuguese-built railroad in difficult mountainous terrain along the banks of the Tua river over the latter decades of the nineteenth century and into the first decades of the twentieth. Our task was to consider the capacity of this engineering (and financial and political) feat to fundamentally reshape a place in every aspect of its collective life. Finally, and perhaps most unusually, the conference took place in situ, that is at the site of the object of shared study at the mouth of the Tua river where it flows into the Douro River in the Tras os Montes region of northern inland Portugal, deep in the heart of Port wine country.

This site at the confluence of a small, but much beloved mountain tributary and a world historical river with major economic significance, is once again the focus of a major engineering initiative, this time one that affects the flow of the river itself rather than primarily the landscape of its banks. Like its nineteenth century counterpart, the construction of a hydroelectric power-generating dam, will bring change to the human and natural ecology of the valley, to the flow of goods and people across space, and to the economic activities and capacities of those people. It will also present engineering and management challenges for which historical experience may suggest some guidance. What better time then to pause and reflect on the history of the last major engineering project at this site in all its many facets, and to place this geographically narrow slice of history into a broader global context of experience and research?

The contents of this volume represent the efforts of the first stage of a three-year endeavor to engage in exactly such a project. To this end we began by considering not

• iii

iv •

just our initial reflections on the history of the Tua Valley and the political and economic underpinnings for its construction (although that is certainly a critical part of this volume as well), but by inviting a range of internationally recognized scholars on the economic and social history of railroads and railroad building around the world. For example, Robert Margo of Boston University and Jeremy Atack of Vanderbilt University contribute a chapter on the impact of railroad expansion and associated land redistribution on increasing economic inequality in the American Midwest in the nineteenth century; Ian Kerr of the University of Manitoba contributes a chapter on the difficulties associated with labor management in the building of the notoriously steep, engineering challenge that was the Bhor Ghat Railway Incline in western India in the same period; and Roe Smith of MIT contributes a chapter on the role of contract work on Russian railroad building projects for the education and professionalization of American rail engineers.

Likewise a number of scholars at the forefront of the most innovative work using Geographical Information Systems (GIS) participated in the conference and have contributed to this volume. Robert Schwartz of Mount Holyoke College uses GIS methodology to consider the impact of railhead access across the rural agricultural landscape in France and Britain; while Marta Felis-Rota of Universidad Autónoma de Madrid and Jordi Martí Henneberg of the Universidad Lleida have generated a time-series mapping of the expansion of the British rail system over two centuries. Richard Healey and Michael Jones, both of the University of Portsmouth demonstrate the range of historical textual and visual sources available for incorporation into GIS projects using the Northeastern United States as their example. And finally, Dominic Fontana also of the University of Portsmouth, offers a remarkable insight into the range of research possibilities inherent in the GIS technology, using the marine archeology exploration of the sinking of the Mary Rose in the sixteenth century, and its subsequent twentieth century excavation and recovery as his case example.

Another interesting, and atypical for most academic proceedings, component of this project was the invitation of a number of prominent scholars working on various understandings of the multiple uses for old railway infrastructure, most especially for tourism and economic revitalization of now marginal areas. Guenter Dinholb, of the ÖBBInfrastruktur AG, Michel Cotte, of the University of Nantes, and Stefano Maggi, of the University of Siena, all contribute different perspectives on current efforts across a range of European countries (Austria, France and Italy respectively) to incorporate old rail lines into regional economic development plans and to promote tourism. From a very different academic perspective, Paulo Lourenco, from the School of Engineering at the University of Minho and one of the co-conveners of the conference, reports

on the initial findings of a team of structural engineers which he leads that are using the existing Tua rail line and infrastructure (bridges and tunnels in particular) to study historical building technologies, and to perform life-cycle analysis on historic materials with an eye toward the development of best practice engineering techniques for the present. Because the hydro-electric power project entails the submersion of a small section of the original track, time was of the essence in carrying out the ‘reverseengineering’ research dependent on the integrity of original objects of construction.

We hope that you will find the diversity of this volume as stimulating to peruse as we found the collective conversation that generated it. It is only on rare occasions that most academic scholars have the opportunity to so fully engage with their colleagues from very different disciplines around a focused topic of common concern. To be able to do so while also incorporating a wealth of international experience and relevant research comparisons is even less common. We all arrived in Foz Tua because of the disciplinary research we had engaged in previously from our many corners of the global scholarly community. But just as we converged on a particular place, we likewise converged on a shared sense of scholarly enquiry. The first fruits of that convergence are represented in this volume.

With tremendous appreciation for the single-minded efforts of Professor Eduardo Beira of the University of Minho and the MIT Portugal Program, whose personal connections to and abiding love for the Tua Valley, when combined with his capacious intellectual interests and boundless energy, gave this project its conception. And with similarly great appreciation to EDP (Energias de Portugal) for their willingness to entertain an historical project of this scope and vision in the context of their hydro-electric power project, and their faithful partnership with us as we have begun to carry it out.

And finally to the people of the Tua Valley and Foz Tua in particular who have welcomed us into their landscape, their history and their lives as we seek to better understand the lessons of a remarkable nineteenth century building project.

Anne McCants

MIT, USA

•v

vi •

ABOUT FOZTUA PROJECT AND THE 1st INTERNATIONAL CONFERENCE IN FOZ TUA

1. About this book

• vii

This book includes the proceedings of the first FOZTUA International Conference, organized by FOZTUA project and sponsored by EDP. This first international interdisciplinary meeting about the memory (and the future!) of Tua Valley and Tua railways was held last 7-8-9 October 2011, in the small village of Foz Tua, in the mouth of the Tua River with the Douro River.

2. About the FOZTUA project

View from the site of thew dam under construction: Tua river. Douro river crosses after the bridge

viii •

FOZTUA Project is a joint interdisciplinary project including MIT (USA) and University of Minho (Portugal), a sponsored by EDP, in order to study, preserve and disseminate the memory of Tua valley and Tua railroad. A new local nucleus about these memories will be built in the village of Foz Tua. Part of FOZTUA project runs with EDAM MIT Portugal program (EDAM stands for “Engineering design and advanced manufacturing”).

EDP, the largest Portuguese utility company, is building a new dam in Tua river, close to the Foz Tua railways station, where the Tua river meets the Douro river. The new dam will submerge a section of the Tua railtrack (around 11 kms), close to the Foz Tua station. Tua line operations has already been closed from 2008, due to safety reasons.

The Foz Tua railways station is a junction between the main Douro and Tua rail lines, located at center and heart of the Porto wine region (Alto Douro – High Douro region).

The Douro line runs from Porto to close the border with Spain, along the Douro valley.

The Tua line used to run from Foz Tua station to Mirandela city, along the Tua valley, and then to Bragança city, and is has been a subsidiary rail line for the main Douro line.

3. About the Tua railway Throughout the first half of the 20th century, the railway line from Tua to Bragança (begun in 1883, completed the first step (Mirandela) in 1887 and the second leg (Bragança) during 1906) was the “highway” for the transportion of people and goods in and

Tua rail track. Tua river runs in the left. Around km 8.

out of the region of the upper Douro River, home to the viticulture that made the world famous Port wine, to the very far, interior and isolated cities of Mirandela and Bragança.

The construction of the first leg of the line, from 1894 to 1897, was feat of advanced engineering in mountain and rocky lands, and the first railroad built by Portuguese engineers and contractors. It allowed easy access from a previously remote region to the city of Porto, and from there to the rest of Portugal, and from there to the world. One hundred years later, this heritage and memory deserves to be retrieved, recorded, analyzed and disseminated. Those interested in economic growth and regional development have a lot to learn from the lessons of the past and must build upon it.

This project intends to challenge the academic community to study the century-long history of the Tua railways and the development of a peripheral region (Trás os Montes) in a peripheral country (Portugal, 19th century), to publicize the memory and the “stories” of the line, and to discuss its role in the region..

The project intends to bring together scholars of various aspects of railroad history in order to share their research on other railroad projects, considering their decisionmaking processes, the management of labor, technical difficulties to overcome, or the economic and social impact of the lines.

4. About FOZ TUA International Conferences These “weekend” conferences do have a “closed research meeting” format and they are held in the Tua region itself, with opportunity both for the discussion of scholar

• ix

research in a comparative perspective, as well as to become familiar with the unique local context and regional flavor of the Tua Valley environment.

Site of the conference: house of Quinta dos Ingleses., in Foz Tua.

x•

Working sessions happened in the main old house of an important local Tua and Douro “quinta” (estate), just close to Foz Tua station.

This is a very old “quinta”. It has been the house of the Cockburn and the Smith families, traditional English families operating in the Port wine business, both in production and distribution. It is locally known as the “Quinta dos Ingleses” (the farm of the English families). Previously, it was one of Dona Antonia’s (1811- 1896) houses, the famous lady that has been perhaps the most important character from Douro region in the history of Port wine.

This “quinta” is now owned by Symington Group, which made the house available for this meeting. We are very much grateful for their kind hospitality.

A fabulous view of the Douro River and Tua River, as well as the river margins with their old and new terraces, can be enjoyed from the balcony of the house. The Tua railtrack crossed just in front of the house. The new Tua dam is being built just less than one kilometer from this very classic “Douro style” house.

Douro and Tua river (right side), as seen from the balcony of the house (Quinta dos Ingleses). The bridge is a railways one, Douro line from Porto to Pocinho.

5. Acknowledgements Thanks very much for the support of EDP (Energias de Portugal SA. www.edp.pt) and all the cooperation by Symington Group (www. symington.com).

Thanks also to the staff that helped us to make this first international meeting to happen and its logistics to work: Marta Meira, Ricardo Fernandes and Ana Prudente must be mentioned.

Support from local counties, C.M. de Mirandela and C.M. de Carrazeda de Ansiães, is also very much acknowledged.

The FOZTUA project coordination team: ANNE MCCANTS MIT Massachusetts Institute of Technology (U.S.A.) EDUARDO BEIRA School of Engineering, University of Minho (Portugal) and MIT Portugal Program JOSÉ M. LOPES CORDEIRO School of Social Sciences, University of Minho (Portugal) PAULO B. LOURENÇO School of Engineering, University of Minho (Portugal)

• xi

xii •

INDEX

PREFACE

___ iii

Anne McCants

ABOUT FOZTUA PROJECT AND THE 1st INTERNATIONAL CONFERENCE IN FOZ TUA

___ vii • xiii

Eduardo Beira

PART 1: TUA VALLEY AND THE PORTUGUESE CONTEXT O VALE DO TUA E O CONTEXTO PORTUGUÊS

___ 1

The Tua line and the portuguese railway (1856-­1906)

___ 3

A linha do Tua e os caminhos de ferro portugueses (1856-1906) Maria Fernanda Alegria

The significance of the Tua Valley in the context of the Portuguese wolfram boom (1st half of the XX Century)

___ 15

Railways in Portugal and Spain: corporate and public policies

___ 43

O papel do Vale do Tua na euforia do volfrâmio português (primeira metade do século XX) Maria Otília Lage

Os caminhos de ferro em Portugal e Espanha: políticas públicas e empresariais Luis Lopes dos Santos

Railways in Trás-os-Montes during the second half of the nineteenth century: projects and achievements

___ 67

The impact of railroad accessibility on the population of Portugal’s Inland North Region (1878-1930). The Tua and the Beira Baixa lines

___ 95

Os caminhos de ferro em Trás os Montes na segunda metade do século XIX: projetos e resultados Hugo Silveira Pereira

O impacto das acessibilidades ao caminho de ferro no interior norte de Portugal: linhas do Tua e da Beira Baixa Luís Espinha da Silveira, Nuno Miguel Lima and Ana Alcântara

Tua Valley: how different is it now? An introduction to the dynamics of population (1864-2011)

___ 119

Narrative(s) on the construction of the landscape: the Tua Valley Memory Center

___ 141

PART 2: ECONOMIC IMPACT OF RAILROADS: METHODS AND INTERNATIONAL COMPARISONS O IMPACTO ECONÓMICO DO CAMINHO DE FERRO: METODOLOGIAS E COMPARAÇÕES INTERNACIONAIS

___ 149

Landownership and the coming of railroad to the American Midwest, 1850-60

___ 151

The evolution of the establishment of the railway network in Britain using GIS

___ 179

O Vale do Tua: quais as diferenças? Uma introdução à dinâmica populacional (1864-2010) Eduardo Beira

Narrativa(s) na construção da paisagem: o Núcleo da Memória do Vale do Tua Maria Manuel Oliveira xiv •

Propriedade da terra e o impacto da chegada do caminho de ferro ao Midwest Americano, 1850-1860 Jeremy Atack and Robert A. Margo

A instalação da rede de caminhos de ferro no Reino Unido, com base em SIG Marta Felis-Rota and Jordi Henneberg

Development of an historical GIS of railroads in the North East USA 1826-1900: phase II

___ 193

Globalization, railways, and agriculture in France and Great Britain, 1850 to 1914

___ 207

The Alto Douro and emigration, 1855-1914

___ 237

PART 3: ENGINEERING, FINANCE AND MANAGEMENT ENGENHARIA, FINANCIAMENTO E GESTÃO

___ 251

More than a brass nameplate on the door: foreign ownership and control in the Companhia Real dos Caminhos de Ferro Portugueses (1860s-1890s)

___ 253

The man behind Tua railways: the chief engineer Dinis da Mota

___ 279

Becoming engineers in early industrial America

___ 301

The building of the Bhor Ghat Railway Incline in western India in the mid-19th century

___ 343

Desenvolvimento de um SIG histórico dos caminhos de ferro no nordeste dos E.U.A.,1826-1900 Richard Healey and Michael Jones

Globalização, caminhos de ferro e agricultura em França e no Reino Unido, 1850 a 1914 Robert Schwartz and Thomas Thevenin

O Alto Douro e a emigração, 1855-1914 Joaquim da Costa Leite

Investimento estrangeiro e controlo de gestão na Companhia Real dos Caminhos de Ferro Portugueses (1860-1890) Álvaro Ferreira da Silva

O homem por trás da Linha do Tua: o engenheiro chefe Dinis da Mota José M. Lopes Cordeiro

A formação de engenheiros nos primórdios industriais da América M. Roe Smith

A construção do grande declive do caminho de ferro em Bhor Ghat, na Índia Ocidental, em meados do século XIX Ian Kerr

• xv

Financing public goods and social overhead capital: some historical lessons

___ 357

PART 4: THE FUTURE OF HISTORICAL RAILROADS O FUTURO DAS LINHAS HISTÓRICAS

___ 371

Railways: industrial and maritime archaeology, geographic information systems, history and culture

___ 373

Dismantling an old railtrack: opportunities in Tua Valley

___ 383

Two case studies in heritage and valorization of ancient mountain railways in France

___ 399

A new role for old railways: tourism

___ 427

Railways heritage: an overview.

___ 445

O financiamento de bens públicos e do capital social: algumas lições históricas Anne McCants

Caminhos de ferro: arqueologia industrial e marítima, sistemas de informação geográfica, história e cultura Dominic Fontana

O desmonte de antigas linhas de caminhos de ferro: oportunidades no Vale do Tua Paulo Lourenço

xvi •

Dois casos de valorização de património ferroviário de antigas linhas de montanha em França Michel Cotte

Um novo uso para antigas linhas de caminho de ferro: turismo Stefano Maggi

Património ferroviário: uma visão global.

Guenter Dinholb

PART 1 Tua Valley and the Portuguese Context O Vale do Tua e o contexto portuguĂŞs

Maria Fernanda Alegria

THE TUA LINE AND THE PORTUGUESE RAILWAY (1856-­1906) A LINHA DO TUA E OS CAMINHOS DE FERRO PORTUGUESES (1856-1906) Maria Fernanda Alegria (Centro de Estudos Geográficos, Lisbon, Portugal) Maria Fernanda Alegria is professor (retired) in the Faculty of Social and Human Sciences in Universidade Nova de Lisboa, and researcher in the Center for Geographic Studies in the University of Lisbon. His book about transportation in Portugal (published in 1990) is a reference.

Maria Fernanda Alegria é professora (reformada) da Faculdade de Ciências Sociais e Humanas da Universidade Nova de Lisboa., e investigadora do Centro de Estudos Geográficos da Universidade de Lisboa. O seu livro sobre organização dos transportes em Portugal (editado em 1990) é uma referência.

Abstract Resumo

My participation in a conference on Portuguese railways at the Tua station in 2011 takes me back to the 1980s, when I was preparing a doctoral thesis on the modernization of transport network in Portugal (1850-1910), but also to the 1950s when I lived with my parents in the Douro valley, close to the Tua river. Therefore, this communication has a dual perspective. A more technical one – the progressive implementation of the rail network in the country – and a more personal one – memories of life in this region during my adolescence.

The first part relates to the construction of two sections of the Tua railway line (Tua – Mirandela e Mirandela – Bragança) as part of the development of the national rail network from its start in 1856 (Lisbon - Carregado), until it reached Bragança in 1906 . Some maps will show the distribution of the railway throughout the territory, the methods of financing its construction and the performance of major railways, including the Tua. The second part recalls aspects of life in this unforgettable landscape, as well as the harsh living conditions and climate, especially at the bottom of valleys such as the Tua where according to popular sayings, there are nine months of winter and three of Hell.

Retorno aos anos de 1980, quando preparava a minha tese de doutoramento sobre

•3

Railroads in historical context: construction, costs and consequences

4•

a modernização da rede de transportes em Portugal (1850-1910), mas também aos nãos de 1950, quando os meus pais viviam no vale do Douro, próximo do rio Tua. Esta comunicação tem, por isso, duas perspetivas. Uma mais técnica – a progressiva implementação da rede de caminhos de ferro em Portugal - e outra mais pessoal – memórias de uma adolescente na região.

A primeira parte relaciona-se com a construção dos dois troços da linha do Tua (Tua a Mirandela e depois Mirandela a Bragança), como parte do desenvolvimento de uma rede nacional dos caminhos de ferro, iniciada em 1856 (Lisboa ao Carregado) até que chegou a Bragança, em 1906. Alguns mapas mostra a distribuição dos caminhos de ferro através do território, os métodos de financiamento da sua construção, e a performance das principais linhas, incluindo a do Tua. A segunda parte recorda aspetos inesquecíveis da paisagem, mas também as difíceis condições de viver e do clima, especialmente no fundo de vales como o do Tua, onde, de acordo com alguns ditos populares, há nove meses de inverno e três de inferno.

Maria Fernanda Alegria

The Tua line and the portuguese railway (1856-­1906) Maria Fernanda Alegria

I. THE PROGRESSION OF THE RAILWAY (1856 - 1906) •5

There are several explanations for the late start and slow progression of the Portuguese railway system that are usually related to the country’s economic backwardness - which is a rather vague explanation. There are more direct reasons, such as the following three: (i) There was never a clear policy concerning the plans for the railway system, i.e., a clear specification of objectives, geographic areas and populations to serve, despite the multitude of proposals from various sources. The formulation of these plans shows two general trends: firstly, the argument for lines that would connect Portugal to the rest of Europe, thus contributing to increased international trade and secondly the argument for lines that would favour the development of the internal market. (ii) The role of the State in the construction of the railway system was never clearly defined: should the State be directly involved in the construction, or should it limit itself to examining the quality and characteristics of the proposals that were being made? The two options were adopted but there were distinct stages in this process, as will be seen. (iii) These two reasons are related to a third one - the unawareness of the real conditions of the territory. There were no topographic surveys: the Direcção Geral dos Serviços Geodésicos (General Directorate for Geodesic Services) was

Railroads in historical context: construction, costs and consequences

not created until 1856. The first survey was carried out between 1860 and 1865; the first map of Portugal at a scale of 1: 500 000, in which the relief is shown by means of contour lines, dates from 1875. The country’s geology was poorly known: the first geological map covering the country dates back to 1876. Statistical analyses were deficient: the natural resources were poorly inventoried, as were the agricultural production and the correct distribution of the population.

One can even say that in several respects the railways spurred the understanding of the territory, including its mapping.

Despite the uncertainties, much was expected from the railways: the development of agriculture, industry, trade, and civilization itself. As it is well known, in the nineteenth century, terrestrial communications meant railways,rather than roads. Most of the roads were, in fact, dependent on the route of the railway with short sections connecting the stations (M. F. Alegria, 1990).

1.1 Fifty years of railway construction: 1856-1906 Leaving aside the numerous plans for the railway system, most of which never left the paper, we will deal with the lines that were effectively implemented (Fig. 1). Two major routes were completed by 1876 (map A) - the Eastern line 6•

Figure 1 - Railway lines open to the public up to 1876 (A), between 1877 and 1891 (B) and between 1892 and 1910 (C).

Source M. F. Alegria, 1990, p. 240, 275 and 301.

Maria Fernanda Alegria

from Lisbon to Elvas with a connection to Badajoz, which was the first international route to be built (1863), and the Northern Line from Lisbon to Oporto (1864). Sections in the Minho region and Alentejo were under construction as well as two suburban lines near Lisbon (Cascais line, 1889) and Oporto (PortoPóvoa Line, 1875).

Between 1877, when the Portuguese Society of Civil Engineers presented the overall plan for the Portuguese railway system, and 1891, the start of a period of crisis, there was a great expansion of the railway system with the opening of 11 lines (map B).

The main ones were, from north to south: the completion of the Minho line up to Valença (1882) and to the border (1886), the Douro line up to Barca de Alva (1887), the Beira Alta line (1880) and the Southern line to Faro (1889). In addition, the exploitation of the Western line up to Figueira da Foz (1888), of the Beira Baixa line to the Covilhã (1891) and of the Caceres line (1880) was started in this period. Amongst the branch lines off the main lines, such as Viseu, we should point out the section of Tua to Mirandela that opened in September 29, 1987.

From 1892 to 1910, few railway lines were opened to the public (map C).

There were mostly extensions of existing lines, or extensions of branch lines off the main lines. Amongst the first, are the section of Bragança to Mirandela that started operations on December 31, 1906, and some extensions of the lines of Minho, such as Guimarães to Fafe (1907) and in the South, the branches of Mora (1908), Moura (1902), the links from Faro to Vila Real de Santo António (1906) and Tunes to Portimão (1903). Some new sections that joined the main axis were built, e.g., the Corgo line, from Régua to Chaves (1907), the Tâmega line from Livração to Amarante (1909) and the Vouga line (1909). Highlighting the area under study, it is worth noting that the first section of the Tua line to Mirandela, 55 km long, opened in 1887 but the 80 km connection of Mirandela to Bragança, built in an area with less vigorous relief than the Tua-Mirandela section was only completed 19 years later!

1.2 .The financing of the railway system Figure 2 shows the railway lines built by the State and by private companies and the type of funding, if any, the companies received. It further shows three distinct zones in the country: (i) Most lines north of the Douro River were built by the State. The exceptions are the Porto-Póvoa-Famalicão line, and the line between Bougado-Guimarães and Fafe that were built without subvention, and the line that currently occupies: the Tua. The latter received a guarantee that covered part of the interest to the capital invested.

•7

Railroads in historical context: construction, costs and consequences

Figure 2 - Railway lines built by the State and by private companies and types of subvention.

Source: M. F. Alegria, 1990, p. 317.

8•

(ii) Private companies built most of the railway lines in the centre of the country, between the Douro and the Tagus rivers; the companies were given different forms of financial support. Generally, the construction of first lines received an allowance per kilometre, while the ones built later were given security of the interest. In the central part of the country, the 16 km link between Figueira da Foz and Alfarelos (station of the Northern Line) was built without any financial help from the state. The same happened to the suburban lines of Cascais and Sintra and the section between Lisbon and Torres Vedras. (iii) The State assured a subsidy per kilometre to the companies that began construction in the south of the country, between Barreiro and Beja, including the branch of Évora. However, the State finished the construction of the Southern line and its branches in order to speed up the construction, with the exception of the link between Vendas Novas and Setil.

Figure 3 gives a quick insight into the chronology of the types of State funding to the companies. With the exception of the first 36 km of the Northern Line, between Lisbon and Carregado, and the lines of Cascais, Sintra and Torres Vedras, all lines built after 1882 received a guarantee, from the State, which covered some interest of the investment.

Only the Tua line, in the North of the country, was financed in a similar way to the lines that were being built in the Centre and in the South of the country at the time.

Maria Fernanda Alegria

A. Lopes Vieira (1983) made an important study on the financing of the railway system in Portugal. Among his data, it is worth noting the total price per kilometre of the lines built between 1874 and 1890 by the State and various companies. The State built 608 km at a total cost of £5,303,696, with an average per kilometre of £8,723. Several companies built, over the same period, 1.432 km of railway lines that cost £15,740,977, which comes to an average of £10,992 per kilometre (1983, Table 19, p. 128). From 1882 onwards, the railway investment was even more advantageous for the companies, which meant that the decision to involve foreign companies was a bad financial decision for the Portuguese State.

Figure 3 - Chronology of State subventions to railroads and kilometres of lines built.

Source: M. F. Alegria, 1990, p. 312.

1 – C.C.P (Compania Central Peninsular...); 2 – Companhia «Brasileira»; 3 – Companhia Real; 4 – Companhia do Sueste (John Sutherland Valentine); 5 – Companhia do Sueste; 6 – Companhia Porto-Póvoa-Famalicão; 7 – Companhia da Beira Alta 8 – Companhia de Caminhos-de-ferro do Bougado a Guimarães 9 – Companhia Nacional 10 – Companhia do Vale do Vouga

1.3. The revenue from the lines The three graphs in figure 4 show the lines operated by the State and by private companies. Figure 4 C shows the case of the largest company, the “Companhia Real dos Caminhos de Ferro Portugueses”. For ease of comparison of income, the three graphs have the same scale.

Among the routes operated by Companhia Real, the Northern line between Lisbon and Oporto stands out, due to the higher revenue per kilometre. With the exception of this line and the lines of Lisbon -Torres Vedras and Lisbon - Cascais, which had large revenue fluctuations, almost all the lines showed increasing albeit low revenues, generally under 1.000 thousand reis. The most significant exceptions amongst the lines operated by the State are the lines of the Minho and Douro that had higher average annual revenues; amongst the lines exploited by private companies, the Guimarães line showed higher revenues.

Among those with average revenues (between 500 and one thousand reis)

•9

Railroads in historical context: construction, costs and consequences

the ones with higher revenues are the Beira Alta line - in fact the only profitable international line - the suburban line Porto-Póvoa-Famalicão and the Southern and South-eastern lines. The revenues per kilometre of the remaining companies are generally under 500 thousand reis: Corgo, Tamega, Tua-Mirandela and Mirandela-Bragança, Viseu, Vouga, Setil and Beira Baixa.

Some lines showed revenues close to zero or even negative, from the start of operation: Tâmega, Vouga and Caceres.

Among the routes with low revenues, the Tua to Mirandela line was, during this period, one of the most profitable, even better than the Caceres line (international) and the line of Beira Baixa. Built with the purpose to help develop agriculture, namely cork and olive oil productions, it turned out to be an important incentive to the launching of the industrial complex of Cachão that galvanized the region in the mid-twentieth century.

Figure 4 - Evolution of the net revenue per Kilometre of the railway lines built until 1910.

Source: M. F. Alegria, 1990, p. 316.

10 •

Maria Fernanda Alegria

II -LIVING MEMORIES OF THE TUA VALLEY REGION (1950-1970) Although we recognize that these days there are strong regional differences in Portugal, the situation did not improve until the 1970s. In addition to inequality, there was poverty in certain areas, as was the case in the Tua Valley. I witnessed these difficulties, because I spent my youth moving between the county of Tabuaço, about 20 km south of the Pinhão railway station, and Carrazeda de Ansiaes about 20 km north of the Tua station. I was born in Sendim near Tabuaço, where my grandparents lived, and my parents lived and worked in Carrazeda.

I often travelled the winding roads between Sendim and Carrazeda de Ansiães both by bus, then called “camionetas de carreira” that were by today’s standards rather unsafe, and later by car. My going to school in Lamego in the early years of secondary education, and to Oporto in the later years meant that I often had to use the Douro railway line, departing and arriving at the Tua station. Although these trips were necessary, the pleasure from admiring the magnificent scenery of the valley, albeit under some coal dust, made them unforgettable.

My father, a physician in Carrazeda de Ansiães, had a contract with CP (Caminhos de Ferro Portugueses) to provide clinical assistance to the railway staff and their families. Part of Tuesdays was devoted to this task that took place in a makeshift office in the Tua station building, where he examined the staff and other patients. I went with him on some of those trips, sometimes to visit friends in nearby farms, other times without any particular reason. Although my father did not comment on the clinical condition of his patients, it was easy to see they went to the doctor as a last resort: often, unfortunately, only to request a death certificate because the family had no means (financial, transportation, etc.) to consult the doctor earlier.

Those who lived in the local area did not have an easy life. It was said that CP sent those employees against whom it had had complaints to the Tua station, as a form of retribution: “ou Tua, ou rua” (Tua, or out). The very few people in the area, besides the ones who worked in the railways, lived without the minimum basic conditions such as sewage, piped water or electricity. The wealthy kept themselves to their properties that came to life around the time of grape harvest; then the slopes filled with workers from surrounding land and sometimes with distant migrants.

Traffic was generally low in the stations along the Tua line largely due to lack of access to the villages. For example, to go to the village of Brunheda that had a train station on the Tua line, my father preferred to travel by bad roads from Carrazeda de Ansiães to a trip along the Tua line. It was hard work to walk from

• 11

Railroads in historical context: construction, costs and consequences

the station to the village, which was situated on a steep slope. The low schedule of trains gave no guarantee that the doctor would get to see the patient in time to treat and hopefully save him. As mentioned above, the doctor was only called in desperate cases, sometimes too late.

Nevertheless, the route between the Tua station and Mirandela stayed in memory of those who travelled on it. Its wild beauty sent a shiver down one’s spine when one looked at the steep slopes that rose from the valley on one side and at the bottom of the tight valley on the other. I was very much aware of the risk, though there were no serious accidents that I recall during the decades of the 1950s and 1960s.

The bustle in the Tua station, with the arrival of passengers hurrying to catch the train waiting in the Douro line, brought life for a moment to the quiet station.

Busy people with suitcases, baskets, bundles, 5 litre bottles, sometimes with live animals, all in a hurry, making sure they would not stay behind. Then peace and quiet returned to the stillness of the sultry months of summer, or to the winter silence that heightened the majestic slopes. The train briefly livened up the lives of residents who looked with some envy on those who departed. There were long periods of little activity interrupted by short moments of excitement that the train brought on and that no longer will.

12 •

IN CONCLUSION...

We know that the Tua line will be shut down, like other small lines in the country, which have closed or are awaiting closure (Fig. 5).

We know that the reasons that lead to the closure of railway lines are varied: the construction of dams in some cases, a non cost effective operation in other cases, and possibly other reasons. When any cycle closes, as in this case, we must remember the past either written in the landscape or preserved in people’s memory. As a former resident of this beautiful region, I wished to give a short testimony of my research on the beginning of the Portuguese railways and of my memories of life close to the Tua.

Maria Fernanda Alegria

Figure 5 - Railway lines closed down or due to be closing. Situation in June 2011.

Source: C. Cipriano, 2011, adapted.

• 13

Railroads in historical context: construction, costs and consequences

BIBLIOGRAFIA ALEGRIA, Maria Fernanda (1990) -A organização dos transportes em Portugal (18501910). As vias e o tráfego. Lisboa: Memórias do Centro de Estudos Geográficos (CEG), n.s 12.

CIPRIANO, Carlos (2011) - “Estudo entregue a troika propõe fecho de 800 km de linha ferra”, Público, 26-6-2011, p. 40 -41.

CIPRIANO, Carlos (2011) - “Visão financeira ditou proposta de fecho de linhas férreas que já esta a provocar crfticas”, Público, 28-6-2011, p. 24 -25.

FERREIRA, Luis; CANOTILHO, Luis (2006) - 100 anos da linha do Tua 1906-2006.

Bragança: INATEL. 14 •

PEREIRA, Hugo Silveira [2010] - “Caminhos-de-ferro entre técnica, estrategia, economia e polftica (1845-1892)”. Disponfvel em www.iseg.utl.pt/aphes30/ docs/progdocs/Hugo SilveiraPereira.pdf (acedido em 29-06-2011).

SEIXAS, João (2011) - “Caminhos de ferro. Opinião “. Público, 17-7-2011, p. 41.

VIEIRA, António Lopes (1983) - The role of Britain and France in the finance of Portuguese railways, 1850-1890. A comparative study in speculation, corruption and inefficiency. Thesis for degree of Doctor in Philosophy, University of Leicester (polic).

Maria Otília Lage

THE SIGNIFICANCE OF THE TUA VALLEY IN THE CONTEXT OF THE PORTUGUESE WOLFRAM BOOM (1ST HALF OF THE XX CENTURY) O PAPEL DO VALE DO TUA NA EUFORIA DO VOLFRÂMIO PORTUGUÊS (PRIMEIRA METADE DO SÉCULO XX) Maria Otília Lage (CITCEM, U. Porto, Portugal) Otília Lage is professor in Universidade Lusófona, Porto and researcher in CITCEM (Transdiciplinar Research Center of Culture, Space and Memory, University of Porto). She has a PhD in Contemporary History by University of Minho and she is also professor (retired) in the Polytechnic Institute, Porto. She was born in Carrazeda de Ansiães (in Tua Valley).

Otília Lage é professora da Universidade Lusófona do Porto e investigadora do CITCEM, Universidade do Porto. Doutorada em História Contemporânea pela Universidade do Minho. Professora reformada do Instituto Politécnico do Porto. Natural de Carrazeda de Ansiães.

Abstract Resumo

“I went to Oporto, with ore to sell / went down by trail / up by car / I do not care” This popular quatrain, sung in Carrazeda de Ansiães in the 1940s, is taken as “evidentiary paradigm” of the central argument of this paper whose conceptual and methodological framework is that of micro-history based on oral sources.

Trams/trains and mining/minerals have remained technico-economic and sociohistorically closely related, since the dawn of the current world economy. Were they responsible for the Portuguese trade surplus? Will such historical evidence be found at the Tua Line with the transport of tungsten in Trás-os-Montes, especially during the 1940s when this metal ore, a highly profitable commodity, was then explored in hundreds of mines and alluvial deposits in northern and central Portugal by large segments of the population governed by Portuguese businessmen and hundreds of domestic and foreign dealers?

The Tua Line (135km narrow gauge), an audacious feat of Portuguese engineering, was inaugurated in 1887 (section Foz-Tua to Mirandela, 55km) and in 1906, it

• 15

Railroads in historical context: construction, costs and consequences

16 •

reached Bragança (80km-stretch Mirandela-Bragança) forging onwards to Puebla de Sanabria (Spain) as expected. A lobby of landowners, who opposed the diversion of the main routes to the Beira Alta Line, managed to successfully leverage its connection to the Douro Line.

A few years earlier, the first mining operation of tungsten (also known as wolfram), a metal ore, of the twentieth century, had been registered in Ribeira, in the Bragança District, in 1868/1871, for its many technological and industrial applications, of which Portugal was the main European producer. Tras-os-Montes thus symbolically became, a pioneer in mining this ore and marketing its high strategic potential, the “black gold” intensely mined during World War II, fought over by two warring blocs, the Allies and the Axis. From 1930 to 1940 the geographic area of tungsten was already serviced by railways that served the flow of ore to Oporto and Lisbon, interfaces that exported internationally. In addition, there were other means of transport, such as trucks, vans, cars, animal and craft vehicles. Thus, a significant portion of transmontana mining production would be transported to the port, along the Douro, Tua and Sabor Lines. By that time, when the Portuguese railways had lived through a situation of crisis from lack of fuel and competition from road transportation, the Tua Valley Line, transportation of passengers and growing outlet for transmontano agricultural products became a target of several repairs and its holding passed to the CP (Portuguese Railway Company).

In this historical context, what were the local, regional, economic, and socio-cultural effects of the Tua Line at national and global scale?

“Eu fui ao Porto, levar minério para vender Para baixo fui de comboio Para cima de carro Não quero saber” Esta quadra popular, cantada em Carrazeda de Ansiães nos anos de 1940, é ume evidencia paradigmática do argumento central deste trabalho, cujo quadro conceptual e metodológico é o da micro história baseada em fontes orais. Comboios / caminhos de ferro e minas / minérios sempre foram realidades técnico económicas e sócio-históricas relacionadas. Encontrar-se-á essa evidencia histórica na relação da linha do Tua com o transporte do volfrâmio, especialmente nos anos de 1940, quando este minério foi explorado em centenas de minas e depósitos de aluvião no nordeste e centro de Portugal? Nessa altura a exploração do volfrâmio (tungsténio) ocupou largas massas de uma população empregues por empresários portugueses e por centenas de intermediários domésticos e estrangeiros, e na altura foi mesmo

Maria Otília Lage

responsável por um excesso da balança comercial portuguesa.

A linha do Tua (135 kms de linha estreita) constitui um feito audacioso da engenharia e foi inaugurada em 1887 (troço Foz Tua a Mirandela, 55 kms) e chegou a Bragança em 1906 (troço Mirandela a Bragança, 80 kms), e esperavase que viesse a ligar-se a Pueblo de Sanabria (Espanha). A pressão de um grupo de proprietários, que se opuseram à diversão das linhas principais para a linha da Beira Alta, teve sucesso na sua ligação à linha do Douro.

Antes disso, tinha sido registada em 1868/1871 uma primeira concessão de volfrâmio em Ribeira (distrito de Bragança), minério então com múltiplas aplicações tecnológicas e industriais, e de que Portugal viria a ser um dos principais produtores. Trás os Montes tornou-se então num símbolo pioneira desta exploração mineira com elevado potencial estratégico, o “ouro negro” intensamente explorado e durante a segunda guerra mundial, e procurado pelo dois blocos, os aliados e as forças do eixo. Em 1930-40 a área geográfica de exploração do volfrâmio estava servida por caminhos de ferro para transporte do minério para interfaces para a exportação, para além de serem usados no seu transporte camiões e todo o tipo de veículos de tração mecânica e animal. Uma parte importante da exploração transmontana seguiu para o Porto através das linhas do Douro, Tua e Sabor. Nessa altura as linhas portuguesas de caminho de ferro sofriam uma crise por falta de combustível, e ao mesmo tempo pela concorrência rodoviária, mas a linha do Tua, para transporte de pessoas e de produtos agrícolas transmontanos, foi objeto de melhoramentos e passou a integrar a nova CP. Neste contexto histórico, o valor económico e sócio cultural da linha do Tua teve efeitos locais e regionais, mas também à escala nacional.

• 17

Railroads in historical context: construction, costs and consequences

18 •

Maria Otília Lage

The significance of the Tua Valley in the context of the Portuguese wolfram boom (1st half of the XX Century) Maria Otília Lage

1. INTRODUCTION This paper articulates two central analytical dimensions: the exploration and commercialisation of Portuguese wolfram1, during the largest boom of World War II (1939-1944) and its transportation by railway: topics presented in point 2.

In this general framework, the empirical study carried out based on the exploration of this strategic metallic ore in the region of Trás-os-Montes and its flow along the Tua Line are summarily presented. This hypothesis is scrutinised in topic 4 – presentation and discussion of the findings obtained – where this probable occurrence is analysed. The methodological analysis used is that of microhistory, presented in topic 3, and is based on the cross-reference of quantitative and qualitative data respectively selected and arranged using railway statistics, documental information obtained through literature review and fragments of oral sources – narratives, memories and testimonials from the local populations, who were actors and/or privileged informants of the phenomena in observation.

Lastly, the conclusion reflects on the historical significance of the correlations that were possible to establish between the intense exploration of wolfram/ tungsten and its transportation along the Tua Line and its importance to the local population and economy, confirmation turned possible with this study, whose potential and limitations are outlined.

1 Thematic studied in the scope of historic sociology and social studies of sciences and techniques, during our PhD thesis, University of Minho, 2001 and referenced in the bibliography.

• 19

Railroads in historical context: construction, costs and consequences

Figure 1 - Tua Line In FERREIRA, Luís, CANOTILHO, Luís – ob cit., 2006, p 43.

20 •

Figure 2 - Wolfram Vein In LAGE, M.Otília P.Lage, ob cit., 2003 cover

Focus here is given to aspects that have been omitted in historical studies of wolfram and the Tua Line, which emphasise specific aspects relevant to social and economic dominance at a regional scale in Trás-os-Montes e Alto Douro.

Maria Otília Lage

2. ARGUMENT / PROBLEMATIC: TUNGSTEN/WOLFRAM AND THE TUA VALLEY In 1938, the crisis of the Portuguese railways was accentuated by the general crisis and by “effects of unruly trucking competition, especially regarding the transportation of goods,” as well as the difficulties challenging narrow gauge companies, whose mergers were being prepared.2 The lowering of commodity prices was advocated, whose transportation profits were much higher than those of passenger transportation, an occurrence that would be repeated many times over until the 1950s.3 At the same time, the so-called “rarefaction of transportation” was also observed in the 1940s,4 which contributed to the diminishing importance of the railways. Despite the trends of deceleration of the railway’s role in the general economy, at the beginning of the 1940s many repairs were done by the Companhia Nacional de Caminhos de Ferro (National Railway Company), in all the stations of the Tua Line, not only to the premises of the repair shops and personnel, but above all to passenger buildings,5 which can be seen as an indication of the continued importance of this Line at the time.

2.1.

Such a circumstance may not be alien to the vast dynamic emerging at the time across the district of Bragança where the Tua Line passed through and to the local population evoked by the socio-economic and political phenomenon of the unruly exploration of wolfram at the time in the north and centre of Portugal.

This exploration produced significant amounts of wolfram at European level, with mines such as that of Panasqueira, which made our country one of the largest producers of this ore in the western world.6 Its exportation and commercialisation in Portugal during World War II not only marks the only time that the Portuguese balance of trade was positive, but also marks the revitalisation of the national mining sector and of Salazar’s foreign policy, which during the war was based on the supply of this raw material for weapons to German Nazis who fought the country’s old ally, Britain.

Around that time, the race in search of wolfram reached very elevated prices creating socio-economic effects, such as those described: “mining production 2 Eng. J. Fernando de Souza - O ano de 1939 e os nossos caminhos de ferro. Balanço de 1938.

“Gazeta dos Caminhos de Ferro”, year 51, 1 January, 1939, n. 1225, p. 13-14 3 Eng. Américo Vieira de Castro – Passageiros e Mercadorias “Gazeta dos Caminhos de Ferro”, year 54, 1 Jan. 1954, n. 1297, p. 9-12.

4 Eng. Manitto Torres - A rarefacção do transporte . “Gazeta dos Caminhos de Ferro“, year 51, n.1345, 1 January 1944, p. 20-21.

5 O que se fez em caminhos de ferro no ano 1940 - “Gazeta dos Caminhos de Ferro“, year LIII, n.

1274, January 1941, p.83-89.

6 Portugal still has tungsten resources that can once again be important if Chinese production decreases. Fernando J.A.S. Barriga – resources Minerais em Portugal – situação e perspectivas.

Theoretical classes. Faculty of Sciences of the University of Lisbon.

• 21

Railroads in historical context: construction, costs and consequences

22 •

increased 50% in value in 1941 and wolfram made up for 63% of the total value.

The following year, the position of the balance of trade was inverted, with an excess of 29% in export value and where wolfram represented 20% of the total.

Thus, by way of a human factor the economy of a country is momentarily revolutionised. The whole northern region was sensitive to the euphoria brought on by the ore. Poor country folk acquired a piece of land, villages of thatched houses were covered with tiles, a large part of the capital was invested in construction, in full boom while Europe was experiencing a housing crisis. “A special class of unscrupulous, and oft times without vision, adventuring “wolfram dealers” was created. A lot of money was lost on ruinous eccentricities and luxury products and, a class of well-off country folk never came to be nor a reserve of capital was ever constituted that could insure a safe investment in the industrialisation of the country. It was a cycle of prosperity that quickly came to an end….”7 At the level of foreign trade, “wolfram, went from a modest 18.200.000 escudos before the War to 542.000.000 escudos in 1941, thus occupying by far, first place in our export, followed by tinned fish that went from 239.000.000 escudos in 1939 to 495.000.000 escudos”8 However, the exploration of wolfram that took place in Portugal9, from the end of the 19th century, and with redoubled intensity in war times (World Wars I and II and the Korean War), turned into a strong movement of technological development in rural areas, in a violent and accelerated interference of man in nature, and in a more widespread and generalised action that left profound marks on the Portuguese social and cultural field. Only isolated in laboratory at the end of the 18th century and considered, for its many applications, “the metal of the 20th century”, wolfram was, namely for its use in the military industry, one of the most sought after commodities and generator of sudden and fleeting wealth, an aspect which survives today in strange stories from some areas in Portugal and in the imagination of generations.

At the level of politics the “issue of wolfram” was the object of intense diplomatic negotiations and celebration of many agreements between Portugal, Ger7 RIBEIRO, Orlando – Traços essenciais de economia. In RIBEIRO, Orlando, Hermann Lautensach – Geografia de Portugal. Vol IV. A Vida Económica e Social, Org. comments and updates by Suzanne Daveau. Lisbon: Ed. João Sa da Costa, 1981, p. 1187.

8 Cap. VII do Relatório das Contas Públicas de 1941, In LAGE, Maria Otília Pereira – Wolfram= Volfrâmio. Terra revolvida memória revolta. Para uma análise transversal da sociedade portuguesa (years 1930-1960). Braga: U.M., 2002, p.679.

9 The mineral resources and the national mining sector of the 1st half of the 20 th century, were concentrated in the Districts of Vila Real, Bragança, Beja, Castelo Branco, Guarda, and Aveiro: northern and central Portugal. From 1836 to 1953, there were 2883 licensed mines in mainland Portugal, hundreds of which were integrated into the 51 delimited mining enclosures from 1920.

Even without considering clandestine explorations, mines with expired manifestos and provisional and temporary concessions, from 1942 to 1952, the horizon was of small operations, profitable and active in favourable situations, where about 50% of these mines were of tin and wolfram. In 1918, of the total number of most important concessions, 33% were of wolfram, the mines of this ore occupying about 35% of the total area. Cr Avelãs Nunes and M.Otília P.Lage, obs cit.

Maria Otília Lage

many and Britain, given its importance to the war effort of the blocs in conflict.

According to those successive agreements first celebrated between the Portuguese and German governments in 1942-43, with a duration until 1944, Salazar agreed to control the price of wolfram, which had increased exponentially with the black market (1.500$00/kg), and guaranteed the annual supply of thousands of tonnes of this ore to Germany at 150$00/kg, and would in exchange receive 300 railway box cars of 15 tonnes, drums and machines, thousands of tonnes of ammonium sulphate, as well as iron for railways and steel necessary for civil and naval construction at the prices of 1938. Later on, wolfram was partially paid in foreign currency and in gold. Of the products to be supplied by Germany to Portugal, there was iron for rails, plates and wire, 500 railway freight cars, 1000 wheels for box cars and 200 OPEL-BLITZ trucks. The Portuguese government undertakes to maintain the price of wolfram stable at 150$00/kg, which provides the Germans with major advantages over black market prices (between 500 and 1000$00/kg.).

There were other agreements between Portugal and Britain aimed at controlling the commerce of Portuguese wolfram and the embargo of its supply to Germany that depended on it exclusively since March 1942, due to the cut in supply of wolfram from the East.

Thus, Britain and the United States intensified the economic blockade of the Iberian Peninsula at the beginning of 1944, when the Allies increased the strategic embargo manoeuvres of the supply of raw materials to Germany. Salazar, who had always been resistant to the idea of the ban of tungsten, which he considered to be fatal to Germany, in a skilful policy of procrastination, contended that the embargo represented a loss of 9 to 10 million pounds per year to the Portuguese economy and to the unemployment of 80,000 to 100,000 tungsten mine workers. Pressured by the British, the Portuguese Government eventually declared in mid-1944 the closing of the wolfram mines and the end of its export.

The movement and export of wolfram was subject to strict regulation conditions by the Regulatory Commission of the Commerce of Metals including its translation in the binding legal registration in waybills made to the Axis and Allies, from Oporto and Lisbon, the main commercial interfaces and by “bloctrains” whose implementation went through the framework of the aforementioned agreements, as well as the participation of Spain.

The «bloc-trains» were complete compositions that travelled at great speed, constituting a true bloc. The large quantities of accumulated goods in transit raised complex logistic issues with insufficient storage, immobilisation of box cars in transit, descending (with destination to Lisbon and Oporto) and ascending towards Vilar Formoso. These compositions that disappeared at the end of

• 23

Railroads in historical context: construction, costs and consequences

World War II with the replacement of maritime circuits are part of the solutions of an economy of war, of speculation and the black market, where demand increases well above the supply in a very adverse situation.10 However, beyond the possibility generated by war dealings, the accumulation of gold and money essential to economic development in the post-war era, wolfram constituted a strong driving force of the Portuguese economy.11

24 •

2.1.1. In Trás-os-Montes, a region rich in mineral and geological resources served by the Tua Line, wolfram and tin from Borralha, Montesinho among other places, were of great importance, as has been said before. In various subdivisions of the district of Bragança the wolfram/tin mines of Ribeira (Coelhoso) and Argozelo, the gold mines of França, the wolfram/tin mines of Tuela, Nuzelo de Baixo and Janca, Vilar Seco de Lomba, the tin mines of S. Martino de Angueira in Miranda do Douro and the wolfram mines of Ifanes were in operation.

Their exploration recorded a steep rise in the first half of the 20th century, especially during the two World Wars, as shown in the records and concession of exploration licenses of more than 30 mines between 1912 and 1918, reactivated and expanded during the period of 1939-1944, as evidenced by the maps of mining records, published monthly.

Also, the emergence of mining companies and societies can be observed.

For example, the Empresa Mineira de Trás-os-Montes, registered in 1919 and, in 1943, the Sociedade Mineira de Paradela, in the village of Pombal, subdivision of Carrazeda that during the same time registered the licensing of more than 11 wolfram mines in villages such as Marzagão, Seixo, Castanheiro, Fontelonga, Amedo.12 Alongside this movement, which also included awards and transfers of mining properties and mining concessions, there was also a rampant race to the free exploration and smuggling of wolfram. The licensed ore was smuggled from the production site using all possible means to the buyers and intermediaries who operated in places such as Foz-Tua, in towns like Moncorvo, Macedo, Mirandela or cities such as Bragança, Vila Real and Mirando from where it was shipped via railway to Oporto and Lisbon from whence it was exported.

10 The «Bloc–trains» (1941 – 1946) – Communication by Gilberto Gomes. Consultant of Historical Heritage. CP – Caminhos de Ferro Portugueses, EP. [available at: http://www.neps.ics.uminho.pt/ aphes28/papers/Gilberto%20Gomes.pdf, consulted in June 2011] 11 Amongst other works and academic thesis, see LAGE, Maria Otília Pereira – Wolfram= Volfrâmio.

Terra revolvida memória revolta. Para uma análise transversal da sociedade portuguesa (years 1930-1960). Braga: U.M-ICS-NEPS, 2003.

12 FERNANDES, Hirondino da Paixão – Bibliography of the District of Bragança. Serie Documents.

Documents (Texts) published: 569-1950. Tomo II (1871-1950). Bragança: ISPB, CMB, B. e ADB, 1996.

Maria Otília Lage

2. 2. The “TUA Line” [1884/1906 – 1998/2008] 2.2.1. Until the mid-nineteenth century, the overview of transportation in Portugal had changed very little in comparison to Western Europe, where transportation was still insufficient, slow, and costly. The network of railways, which began in Portugal in 1856 with the train from Lisbon to Carregado contributed greatly to the increase of mobility and accessibility, by comparison, 19 years after France and 31 years after Britain. The State’s intervention was patent in the railway sector, from the start subject to a concession regimen of public service.

Construction and exploitation of fundamental networks of transportation by private companies began to develop at the end of the 19th century.

In the first quarter of the 20th century, with the development of the automobile and road transportation, competition began in the transportation sector, which provided a greater dynamic to economic growth and development, with public authorities trying to safeguard the economic survival of railway transportation that had suffered a progressive decrease of passengers. “(…) the war period amounted to a retraction of the offer provided by railway companies, as a consequence of lack of coal and the difficulty in acquiring material and equipment. However, there was an increase in the routes for the transportation of goods.

In a continent at war, the management of international railway traffic from the Franco-Spanish border was done according to the directives of the Reich Railways, whose peninsular headquarters were located in Barcelona, with a delegate in Lisbon who was not only responsible for the management of traffic between Portugal/Germany, but also for the management of traffic to France, Switzerland, Finland, and Sweden among other countries.

At that time when “everything could be bought, everything could sold and price was of no importance,” the increase in railway traffic reflected just that, especially in the transport of goods, which constituted a “breath of fresh air” for companies after a decade of losses.13 Investment in the Portuguese railway would continue with the Intermediary Plans for Development (from 1953 to 1973) created with the objective of integrating the Portuguese economy into the European economy. The entire twentieth century was marked by a policy of strong regulation in transportation, which remains a deficit public service, under the ambiguous control of the State legislator and owner.14 2.2.2. The history of the “Tua Line” - metric gauge railway link (narrow gauge), connecting the Foz-Tua station to the Bragança station, with a total length of 133.80km, is since the beginning of its construction 7 years ago, a turbulent and epic history in terms of studies, projects, contracts and undertakings, 13 The «Bloc–trains» (1941 – 1946) by Gilberto Gomes, ob cit.

14 PJS Chenrim - As Políticas Públicas no Sector dos Transportes Ferroviários Portugueses gesde 1974. Mestrado em Administração Pública., 2008. [Available at http://repositorium.sdum.uminho. pt/bitstream/1822/8927/2/TESE%20MESTRADO.pdf, consulted in July 2011]

• 25

Railroads in historical context: construction, costs and consequences

and makes up what has been considered “as one of the most notable works of Portuguese engineering of the time”.15 The purpose of the Tua Line was initially to ensure a connection between the Douro and Zamora. The project was considered definitive for the construction of the railway line from Foz-Tua to Mirandela.16 Its future was in Mirandela, central town of the province, capital of the designated hot land, to where the production of Valpaços, Vinhais, Macedo, Franco and where many other people would converge and where fast transportation to anywhere could be found. The construction of the section to Bragança was of great interest because it would attenuate the problem of communication routes in this region of elevated economic importance due to its production in livestock, grains and ores, especially tin, wolfram and gold.

During the 1940s when unification of the exploration of the national railway network took place, and with the approval of CP’s new statutes in 1947, the Tua Line was then managed by this company.

At the beginning of the 1960s, the communication network in Trás-os-Montes (road and railway) was not very efficient in connecting remote areas to larger urban centres. JAE had a road plan for the district of Bragança, but there was also a need for an identical study for railway communications.

26 •

2.2.3. The railways in Trás-os-Montes and Alto Douro, especially the railways of the Douro and Tua, stamped dynamism on the population and economy of the region in an era when there was a lack of roads, when they represented a connection to the outside world besides the many villages and places they served, transporting people, agricultural products and bringing in fertilisers.

On the Tua Line, the Brunheda station, a subdivision of Carrazeda, became a huge added value to the people from the neighbouring villages of Pinhal, Pombal, and Sentrilha. For a long time wine and olive oil were shipped from there.

Traffic began to dwindle only with the construction of the road bridge linking Carrazeda and Murça in the 1950s and ‘60s.

15 FERREIRA, Luís, CANOTILHO, Luís – 100 anos da Linha do Tua (1906-2006). Bragança: INATEL, Delegação de Bragança, 2006, p. 52 16 A span of 54km, with 6 tunnels (with a total of 522m), 2 metallic viaducts and 4 bridges (with platforms totalling 220m). The stations are as follows: Foz Tua, Tralhariz, Amieiro renamed Santa Luzia, a station that has always serviced the people from S. Mamede de Ribatua, Safres and Amieiro from the subdivision of Alijó, who successively crossed the river by boat, suspension bridge of wood and rope and cable car, (nonexistent today), S. Lourenço, Brunheda, Abreiro, Vilarinho das Azenhas (transferred near to the bridge of S. Pedro de Vale do Conde), Cachão, Frechas and Mirandela (ALVES 1975-1990, IX, 225-226), and the following flag stops of the subdivisions of Carrazeda and Mirandela: Tua, Tralhariz, Castanheiro, Santa Luzia, S. Lourenço, Tralhão, Brunheda, Codeçais, Abreiro, Ribeirinha (built later on), Vilarinho, Cachão, Frechas, Latadas and Mirandela.

List of Tunnels: Presas, Tralhariz, Fragas Más I, Fragas Más II, Falcoeira and Frechas. List of viaducts and bridges: at km 1, Presas, Paradela, Cabreira, Vieiro and Carvalha.

Maria Otília Lage

Also, the agro-industrial complex of Cachão experienced its first growth spurt when it was served by the Tua railway line.

Not far from Mirandela is the Carviçais station, located in an area rich in olive oil. It was important to the agricultural development of the region throughout the years, where the School of Agriculture was established, the only one in the district of Bragança that up to the 1970s trained many people specialised in agriculture.

At the station in the town of Macedo de Cavaleiros, the Tua Line was also an event whose importance was recounted in the recent fictional novel “Um tiro na Bruma” by veterinarian Manuel Cardoso who portrays the local society of the first half of the twentieth century.

In Romeu de Jerusalém, where the line arrived in August of 1905, the railway station would also come to be of great importance for the transport of cork and olive oil, as well as for the parish itself whose people saw their social and economic life improved with the railway line through which the workers for the House Meneres arrived and the area of Romeu whose settlement is also closely linked to the Tua Line. The following station of Cortiços, now restored, was for decades, along with the station of Azibo, a point of transport for cork, one of the richest of this region.

Through the station of Salsas, in the place by the same name, the transport of dried fruits and nuts, such as chestnuts and walnuts, was of relevant importance.

Near Bragança, the small flag stop of Chãos serviced the largest livestock fair in the district of Bragança for many years.

In 1955, the Tua Line, where speed was more than 50km/h, recorded major improvement in services and circulation, with the allocation of modern Dutch railcars, consisting of a motor vehicle and its respective trailer, purchased by CP from the construction company Allan. These new railcars gave way to the practice of more convenient schedules to suit the interests of the public.17 Still, in 1964, Mirandela, Cachão and Romeu as well as many other villages in the district of Bragança were better served by the railway system than by the road system regarding their connection to the outside world.

Although the situation with the road network in 1967 was distressing in the district of Bragança, where it was necessary to urgently build hundreds of the kilometres of municipal and national roads,18 road competition did not fail to affect the railway system. However, in 1968, the policy for the restructuring of railway communications that occurred at that time made it possible to resume train connections between Oporto and Salamanca. This was also of great importance to the region of Trás-os-Montes and Alto Douro, which began to record a growing attraction in tourism, and which benefitted from improvements introduced to the roads and used the tourist trains that began to operate in the Douro Line as well as in the Tua Line.

17 CP Bulletin, 1955, p. 9 18 REIS, Rogério - A CP e o progresso de Trás-os-Montes. NM. Mirandela. (3/9/67), pp. and 8.

• 27

Railroads in historical context: construction, costs and consequences

3. METHODOLOGY/ EMPIRICAL WORK

28 •

The study on the composition of railway traffic raises specific methodological problems, which we have not ignored.

The transport sector depends a lot on data and information which, at times, are pseudo-numbers that cannot be mechanically treated. Socio-economic statistical data have a universe of their own and cannot be used as pure numbers. Thus, these data were used as much as possible within their context in order to avoid abusive transpositions that could result in estimation errors.

On the other hand, statistics on railway traffic, whose unit measures are not uniform (at times it is shown in tonnes, in kilograms or in weight), do not always have the same significance, so we opted for the weight category, since it is the most frequently reported.19 Information on the transport of commodities and within these, of ores, is rarely shown segmented in official statistics whose complete series over the years are also difficult to obtain from the myriad of archives and libraries in which they are dispersed. We sought to overcome this difficulty by resorting to information from oral sources reconstituted through memories and testimonies from privileged informants recovered from narratives gathered by interviews.20 These sources give us insight to important facets of historical phenomena (economic, social, political and cultural) that influenced the lives of the local populations and villages, and contributed to the regional development, which had its impact at national scale and, indirectly, even at global scale.

The testimonies collected from some of the people and residents from riverside villages of the Tua Valley, especially from the subdivision of Carrazeda de AnsiĂŁes, constitute relevant empirical information, which following a crossreference methodology of sources complement each other and interact with numerical data extracted from railway and mining statistics, with information published in local, regional and national periodicals, with other secondary sources and bibliography selected from literature review of the speciality of each of the two fields of study.

The base cognitive strategy being followed fits into the microhistory, which emphasises the indirect, evidentiary and conjectural nature of historical knowl19 Weight - Unit of measure that corresponds to the quantity of goods/commodities in tonnes (1000 kilograms). The weight to be considered includes, transported goods, packages and the tare weight of the containers, movable boxes, pallets and road vehicles transported by railway, in the scope of combined transport operations.

20 The 10 semi-structured interviews carried out to date, included people of both sexes, male and female, with ages from 75 to 85 years-old, originally from and/or long-time residents in the following villages of the Tua Valley that simultaneously represent the stations and flag stops of: Sentrilha and Brunheda (2 interviews in August 2011), Pombal (1 interview in August 2011), S. Lourenço, (1 interview in August 2011), Fiolhal and Tua (5 interviews in August/September 2011), Castanheiro /Tralhariz (1 interview in August 2011).

Maria Otília Lage

edge, and mobilises the notion of “evidentiary paradigm,” (Carlo Ginzburg, 1989)21 which is characterised by the capacity of describing a complex reality using apparently irrelevant data that would otherwise not be scientifically experimental.

4. RESULTS AND DISCUSSION: WOLFRAM AND TUA LINE With the statistical data examined from various sources, but from which it was only possible to have access to information from the Companhia Nacional dos Caminhos de Ferro (National Railway Company) and not from CP, company to which the Tua Line was transferred in the 1940s, the following tables illustrate, as a whole, the transport of goods and passengers of the Tua Line (Foz-Tua - Bragança) in the first half of the twentieth century. These tables also provide insight, particularly in certain periods, into the social differentiation of passengers who travelled in 1st, 2nd and 3rd class, and into the main commodities transported and corresponding quantities.

Table 1 – Commodities (*) and passengers transported along the Tua Line: Foz-Tua to Mirandela (years 1914 – 1921) (**) Years

Passengers 1st class

2nd class

• 29

Commodities

3rd class

High speed

Low speed

1914

2,350

7,493

33,360

Total 43,203

977

27,980

1915

2,187

7,084

35,022

44,293

911

27,719

1916

2,396

7,729

39,525

49,650

1,189

28,333

1917

2,932

8,236

46,742

57,910

1,678

25,858 25,225

1918

3,085

7,511

41,960

52,556

2,379

1919

3,889

9,945

47,946

61,780

3,637

17,573

1920

4,077

13,242

52,870

70,180

2,913

21,643

1921

3,448

12,987

52,121

68,556

2,455

26,620

(*) Here, the weight (1000 kg/tonne) was also the unit of measure considered. (**) The unit of measure considered was “weight” given the discrepancy observed in the statistic designations consulted throughout the various years considered in this and other tables. The recorded numbers refer to global quantities, transported by slow route The data were compiled and extracted from the “Quadro resumo do tráfego no continente da República nos anos de 1914 a 1921.” Ministério das Obras Publicas -Direcçao Geral das Estatísticas. Caminhos de Ferro: Statistical data… p. 5 -14.

Observation of the table above shows that during the first half of the twentieth century, there is an annual turnover of more than 100,000 passengers, with a 21 GINZBURG, C. - A micro-história e outros ensaios. Coleção ‘Memória e Sociedade’. – Lisbon: Difel; Rio de Janeiro: Bertrand Brasil, 1989; also from the same author O queijo e os vermes: o cotidiano e as idéias de um moleiro perseguido pela inquisição. São Paulo: Companhia das Letras, 1987.

Railroads in historical context: construction, costs and consequences

Table 2 – Commodities (*) and passengers transported along the Tua Line: Foz-Tua to Bragança (years 1914 – 1921) (**) Years

Passengers 1st class

2nd class

Commodities

3rd class

Total

High speed

Low speed

1914

2,501

7,915

47,542

57,958

689

18,168

1915

2,413

7,962

49,617

59,992

621

19,979

1916

2,369

8,638

58,264

69,271

813

19,819

1917

2,602

8,233

61,499

72,334

1,028

17,567

1918

2,727

7,448

62,954

73,129

1,600

18,333

1919

3,579

9,443

63,854

76,876

2,424

11,930

1920

3,522

11,448

71,657

86,627

1,990

15,230

1921

3,690

13,538

77,743

94,971

1,561

18,842

(*) Here, the weight (1000 kg/tonne) was also the unit of measure considered. (**)Data extracted from the “Quadro resumo do tráfego no continente da República nos anos de 1914 a 1921.”

Ministério das Obras Publicas -Direcçao Geral das Estatísticas. Caminhos de Ferro: Statistical data. .. p. 5 -14.

Table 3 – Mobility of Passengers and Commodities along the Tua Line (1914-1941) Global Table synthesis 30 •

Years

Passengers

Commodities (weight)

Total (1st, 2nd and 3rd classes)

Total (hs + ls)

1914

101,161

47,814

1915

104,285

49,230

1916

118,921

50,234

1917

130,244

49,131

1918

125,685

47,537

1919

138,656

35,564

1920

156,807

40,776

1921

163,527

49,478

1922

185,435

48,043

1923

182,602

52,823

1924

183,009

50,072

1926

186,081

58,505

1927

176,803

60,547

1929

176,611

70,7 12

1937

119,401

47,057

1938

112,599

40,508

1941

148,605

44,793

Maria Otília Lage

steady rise only interrupted by a slight decrease in some years, which was more pronounced in 1937/1938. This break was soon corrected by a rise in 1941. The same behaviour occurs with the transportation of goods.

Table 4 – Commodities (*) and passengers transported along the Tua Line: Foz-Tua to Bragança (years 1914 – 1921) (**) Period

Thousands of tickets sold

Goods expedited (in tonnes).

Total

Goods expedited (in tonnes).

H.spd.

L.spd.

1966

292.3

19,126

1,189

18,937 12,777

1967

293.7

13,089

1,312

1968:Jan. – Set.

214.9

12,1 99

785

11,414

1973

255

11,408

453

10,955

Comparatively observing the whole of the previous table of total passengers and goods transported, it can be concluded that the respective luxuries were reported in the second half of the twentieth century, between 1966 and 1973. Opposite behaviour, where the number of passengers increased greatly, always over 200,000, whereas the freight of goods suffered a very sharp decline in relation to the time period previously considered.

It can be observed in the previous table with disaggregated data of the volume of goods, that the volume of transported ores was never very significant in comparison to other commodities where fertilisers and agricultural products predominated. This was also highlighted by the reconstituted oral sources. The highest volume noticeably corresponds to 1917, year still covered by World War I, and from which the volumes decreased sharply, regaining some significance in 1924. A period of total inexistence of records between 1926 and 1937 follows, then there is another relevant increase from 40 to 70,370, between 1938 and 1941, period which coincides with the beginning of World War II. However, the interpretation of these data should be mediated by the analysis conducted in the previous items on the historical importance and significance of the exploration process of wolfram/tungsten ore, namely from Trás-os-Montes.

Not able to refer here to a series of comparable phenomena, we took this observation as a sign or indication for a broader research study and interpretation that we are allowed by the use of the “evidentiary paradigm” notion (Carlo Ginzburg, 1989) already explained and that we continue to resort to by using fragments of oral sources compiled into 10 interviews carried out in riverside villages of the Tua Valley and Line, subdivision of Carrazeda.

The selected narratives illustrate the significant importance of the Tua Line

• 31

Railroads in historical context: construction, costs and consequences

Table 5 – Commodities transported along the Tua Line (years 1917-1941) + Years

32 •

Fertilisers

Olive oil / edible oils

Dried fruit/ chestnuts

Fabrics (linen, wool) and raw cotton

Wood and timber

Salt

Wines

Ores

1917

1,042,509 1,838,085 1,680,174 1,019,358 5,508,165 3,089,688 3,053,854 440, 482

1921(*)

2,131,061 4,077,650 1,671,263 1,184,298 3,503,622 2,847,043 2,069,410

14, 498

1922 (*)

2,417,593 1,310,737

317,794 1,972,984

3,08,564 1,809,265 3,156,627

21, 316

1923 (**)

5,178,396 1,824,524

856,291

567,126 3,748,184 2,008,224 2,728,182

65, 324

1924 (**)

3,892,396 1,683,835 1,929,136

775,966 3,017,800 1,830,557 3,371,436

79, 508

1926 (***)

3,562,080 1,891,052 2,152,745

935,942 2,908,867 2,431,380 4,108,517 -----------

1927(***)

3,254,870 2,094,490 1,916,824

701,416 3,001,650 2,170,310 4,118,740 -----------

1929 “pv”.

5,415,092 2,028,796 1,774,878

1937

7,155,530

19,576 1,561,980

---------

----------- -----------

990,190 2,155,342

197,449 1,749,242

710,705 2,215,091 ----------

1938 (****) 6,112,169 2,908,265 1,812,681

395,661 1,889,213

421,682 2,230,826

40

1941 (****) 6,118,634 3,232,645 1,420,746

448,379 3,177,803

477,689 1,452,137

70, 370

(+) Data taken from “Relatórios de exercício da Companhia Nacional de Caminhos de Ferro”, years: 1917 to 1941 The sources with data from the years 1917 – 1927 do not discriminate between the transportation at low and high speed, summarising only the global quantities of goods transported at low speed, for this reason this element could not be taken into account for the purpose of the table. In the years 1929-1941, low and high speed is discriminated, which is translated in the table by “ls and “hs”, even though the latter, for the most part is for fresh produce, is not significant by comparison; (*) in these 2 years there is an addition of cork, respective value (1,435,532 / 1416,615) and wheat (3,031,650/ 1,377,621); (**) in these 2 years there is an addition of cork, respective value (1,570,841 / 1104,346), wheat (2,873,616 / 3,539,479) and potatoes (1,202,441 / 583,205); (***) in these years there is an addition of cork (1130,660 / 984,825), wheat (3,695,400 / 3,639,798), potatoes (979,509 / 476,150) and dregs/husk and waste (2,881,302 / 3, 097,744), increasing edibles to 2, 209,728; (****) the quantities of wheat and rye transported in progressive growth, in these two years, reach 6,666,856 / 6,417,159 and 1,051,456/ 872,470, respectively; in 1941 there is a significant increase in the transport of potatoes reaching a value of 4,799,614.

for the people living alongside the river and for the economy of the neighbouring lands. As well as the large flows of people and goods, those of basic necessity subject to rationing, and the exceptional dynamic created by the exploration and commercialisation and smuggling of tungsten ore, partially and cyclically transported from the Tua station where dealers and intermediaries converged and operated, namely during World War II.

Maria Otília Lage

Figure 3 - S. Lourenço Station

• 33

Figure 4 - Cable car connecting the village of Amieiro to the Santa Luzia station

Railroads in historical context: construction, costs and consequences

Figure 5 - Tralhariz

34 •

Figure 6 - Fragas Más Tunnel, curving, 6km from the Tua station*

* Companhia Nacional de Caminhos de Ferro Linha de Foz-Tua a Mirandela. - Porto: Lilh: Emilio Biel & Cª , [s.d.]. - [24] f. : todo il.

Maria Otília Lage

Narrative n. 1 – Brunheda Station: “…the Tua Line…there at Brunheda station, at that time they drilled a very strong trench and a lot of ore spilled out… there was a vein that passed beneath the line …they let the train pass through…and they were in the trench taking the ore …the buyers would come at night for the ore, they would load the bags onto donkeys and would carry it away through whichever trails possible…there were a lot of people involved and many who would buy … during those years of ore nothing was produced …they would only mine, buy and sell the ore …they would go by train to Mirandela and to Tua …the railcar was always full…in Sentrilha there were 12 couples, with more than 7 children each and 7 carts of oxen and goat and sheep…a lot of wine was transported from here… there were a lot of people that owned a hoe, they would dig wells … behind them… in Pombal, in Areais, along the hillside of Pinhal, in Zedes, Amedo …there in Carrazeda the ore was from a German company … the ore could travel through the Tua Line, but you didn’t know what was being transported… the packages were camouflaged…a lot of the ore was smuggled…only the company could sell it, but it had the trucks and vans to transport it…many people from Cachão would come to buy ore… there was a lot of movement at that time …everything was sent to Oporto…olive oil, grains, wine that went straight from the mills in 750 litre canisters (= 30 almudes = 30x25 = 750 litres)…from here to Brunheda station in ox carts through trails because there weren’t any roads yet…up north, they shipped fertiliser… …trains ran on coal…during the war many pine, cork and eucalyptus trees were cut down …to make coal …they would carry those logs on their backs… a line of women would take the firewood and carry it … rail wagons would come to the Line to load the logs and firewood… there were inspectors who would watch the carriages…they were responsible for everything… to look at the end of the line and not see it pass by…or hear its whistle…is nostalgic … they used the train a lot to go to Tua and Mirandela, to go to Régua, to the Casa do Douro and Bragança… the railway was essential to the economy of these lands…the time of ore…they were still around for 10 years…in Codeçais there was this guy, Bernardino, who had some land with a lot of ore and he smoked cigarettes made from bills because there wasn’t any paper…many made it good from ore, they knew how to take advantage…”22 Narrative n. 2 –S. Lourenço Station: “…wolfram? …At Sousa Velha, a lot of ore was extracted …on the hillside towards S. Lourenço…a miner from S. Mamede died there …and near to Penedo Furado …at the entrance to Pombal …there were open pit mines…there were springs on the hill and there they would wash the ore…they would sell it to Mr. 22 Interview n. 1 - sr. J.A. T, 77 years-old, Sentrilha, 14/8/2011

• 35

Railroads in historical context: construction, costs and consequences

36 •

Mário Lima, from Pombal… they were the most well-off people …if it was transported along the Tua Line?.. of course, just like with other things, wine, olive oil, fruit… everything would go to S. Lourenço to be shipped by train…they would travel at dawn…there were many loads…almost all homesteads had ox carts…at that time in Pombal there were 8 or 9 carts of oxen…there wasn’t any road yet, but there was a wide path… …the Tua Line was always very important…from there wine, olive oil, cork was transported… goods would come in by train for the “sótos”23… we would use the train to travel to Oporto, to go to the fair in Mirandela… …in 1945, because of the famine we had nothing to eat…it was rations…there were a few water mills upriver…and Isabel, the baker, would go to Vilarinho das Azenhas and bring back bread from Cachão to sell and eat…a little rye bread cost a lot of money … people from here had a lot of connections with the people down along the Mirandela Line…at that time there were a lot of people…!

There were 3 passenger trains and others for goods, transport of fertilisers and other things …there was the black market…they would travel by train and would bring back 3 or 6 alqueires (measure for grains) of wheat hidden … people from Pinhal, Zedes, Areias, Amedo would come here to pick olives, dig the vineyards, harvest the grapes… herds of men …they all came by train… all the roads around here are only from 1947 … in Pombal, during the war there were 800 people… today there are about 200…”24 Narrative n. 3 – Flag stops of Fiolhal and Tralhariz “… in the subdivision of Carrazeda I have profitable vineyards along the hillsides of the Tua and the Douro…I was able to get EDP to move the portion of construction of the dam so as to not wet my property that has access to the Tua Line… I took photographs of that small estate, which document all the work of the vineyards and wine… …here in Fiolhal a lot of wolfram is explored…in its time ore was extracted from the corner of this garden…this property belonged to my father…brother to Srª Generosa, her husband explored wolfram in the Lousa Valley, in Rosão, and in Cruz do Raio above Vinha da Rosa next to the Lousa Valley, which belonged to a woman who moved to Brazil …in Cortinha do Ribeiro…all of Fiolhal and Tralhariz had ore…Tralhariz stole from Fiolhal and in Fiolhal they also stole because no concession was ever made…one day someone gave my father 50.000 escudos…but he didn’t want it because they would extract ore from land that hadn’t been licensed…that person then put 80.000 escudos in the bank …money from wolfram to start an almond business…just yesterday I spoke of this with Srª Maria who is over 90 years old…

23 Emic word that means small trade settlement 24 Interview n. 2 – srª . F.L, 82 years-old, Pombal, 17/8/2011

Maria Otília Lage

Narrative n. 4 – Foz –Tua Station “…the ore would go to Vila Real where there was someone who would separate it … there were also separators in Amedo…there’s this boy in Fiolhal whose father worked in Amedo and died of silicosis he got from working the mines…

His younger brother died in Tralhariz when he set fire to the fuse… gunpowder… it was crazy…everyone was after it, the ore…when there was little, 1kg or 2, they would take it to Tua where the intermediaries would buy it…when there was a lot, the men would carry it to the flag stops of the Tua line or they would take to the trails at night all the way to Tua, from where it would then go to Vila Real, Oporto and Spain… the chauffer of Mário Assunção, a big buyer here in Tua…bought a property he then sold to the Malvedos… I was also a driver for Mário Assunção who would buy wolfram from small miners…money from wolfram didn’t last because they would spend it badly…you would eat kid with sponge-cake… there was hardly any bread, with rationing during the war …once, a shipment of 500/600kg of ore being transported here by car, was apprehended in Moncorvo…the guard flagged me down, but I sped off and fled and the guard wasn’t able to catch me…I could have killed the guard when I sped off to flee …I then went from Moncorvo to Oporto to take the ore…another time it was a shipment of 1000 kg apprehended by the guard from Fernando of Fiolhal who carries a lot on his back… to Alijó…he would have dinner there and would then return by taxi… when it was all over he moved to Brazil…large part of the ore here in Tua would be shipped to Vila Real by van …by train, it wasn’t possible…only if it was camouflaged…it was the black market…”25 Interview n. 6 – Tua and Santa Luzia Stations “…the main goods along the Tua Line were cereal that would come to Tua from the stations of Bragança, Rossas, Sendas, Macedo de Cavaleiros, Cachão where there was a warehouse for cereal going to Lisbon, Xabregas, and to the Póvoa and Guimarães lines…going up north were fertilisers…there were trains and trains of fertiliser…back then there were 3 or 4 trains with 200 to 300 tonnes each, apart from other detailed goods, in low and high speed trains…generally high speed trains were for fresher goods, but also solids, but it was quicker transport with higher fees ….there was also gold from the mines of Bragança shipped in low speed trains… in wooden barrels weighing 800 kg each…each rail wagon only carried 4 barrels…it was then at the beginning of the 1950s… yes, I remember hearing about wolfram…it would ship out of Tua in trucks and vans to Vila Real, to the separator…yes, from Timpeira…I went there a few times with Mário Assunção and his cousin Manuel Luis…he would buy van-loads and van-loads of ore…on the hill in front of Fiolhal …from Tua to S. Mamede and Alijó the Ameri25 Interview n. 4 - sr A. S.da S., 83 years-old, father of Calça Curta, Foz-Tua, 19 /8/2011

• 37

Railroads in historical context: construction, costs and consequences

cans and British explored wolfram … in 1945 cereal was shipped directly from Tua to Barca d’Alva and then to Spain…and we would eat crumbs… …iron in stone would come from Carvalhal…Moncorvo…trains and wide gauge trains from Pocinho to Leixões…it was for expensive ovens… …yes, they dug up everything looking for wolfram…along the Tua and Douro Lines…they dug wherever they found veins of wolfram…back then …there was a lot of inspection and they would apprehend free ore…even from Tua to Vila Real we would travel at night to evade inspectors…back then, just at the Tua station 50 people worked on the trains…transfers, goods, passengers … …the importance of the Tua Line?..it was of great importance to the people…if they wanted to go to Mirandela they would take the train to Tua …at the week-end, for those who studied in Mirandela …but the Tua Line is a very dangerous line …”26

5. CONCLUSION

38 •

Some conclusions can be drawn from this preliminary study on the historical significance of the exploration of wolfram and its partial and cyclical shipment along the Tua Line, and its importance to the people of Trás-os-Montes, in the first half of the twentieth century.

The first, in general terms, has to do with the relation between mining in which vans loaded with “tout venant” circulated by railcars transporting the ore and the general circulation of goods and people by railway that was being built around the world is a historically proven fact.

The construction of the Tua Line, an outstanding work of engineering history, was no exception to it having been closed down. It had been expressly declared that its creation was aimed at dispersing the ores that were and are rich to the region of Bragança. Thus, it was indeed used that way if we look at the volume of ore transported, for example, during the two World Wars and especially in 1917, which ascended to more than 400 tonnes. Such a correlation between the Tua Line and the dispersion of the designated “ore” is also widely referred to in the oral sources treated regarding the boom period of Portuguese wolfram in World War II and in local events where vicissitudes inherent in an economy of war are evident, for example the “camouflaged” transport of this ore along the Tua. It was also noted that the Foz-Tua station – crossing platform of the Tua and Douro lines and historical nucleus for the circulation of people and goods between Trásos-Montes and Alto Douro and the country, thus, became a small hub of trade for wolfram/tungsten.

26 Interview n. 6 – Sr. A.J.A., 77 years-old, railway worker, originally from Foz-Tua, son and grandson of railway workers of the Tua Line, Vila Flor, 5/9/2011

Maria Otília Lage

These findings have yet to be read in light of the historical conditions of Portuguese exploration of wolfram, the strategic metallic ore of the twentieth century. The exploration of tungsten was largely of the responsibility of companies dominated by German, British, Belgian and French capital with road transportation at their disposal on the one hand, and the dissemination of “ore extraction” by the people who would flock to any opportunity to get their hands on a product that attained exorbitant sales prices in an expansive black market alongside the regulation and setting of quotas in successive agreements with the belligerent blocs: Axis and Allies. This resulted in large flows of traded wolfram passing in large scale hidden from the formal railway transport subject at the time to stronger constraints and more controlled by the authorities.

Nevertheless and as can be seen here, history is full of situations that can trick even the best laid plans by man. Thus is how the volume of ore transported along the Tua Line was not confirmed, which could be foreseen during the peek of wolfram exploration in Portugal, the most important European producer of this ore during World War II, period in which this ore was the only surplus factor of the country’s balance of trade.

From this analysis we can also conclude that it would be incorrect to attribute certain changes/implications to a sole specific factor, which would complement the obtained findings through other means and distinct approaches. To that extent, this study can bring new challenges to the research, contributing to the reinforcement or alteration of classical perspectives regarding the historical interrelation between trains and ores and the controversial relationship of the railway as an engine of economic growth.

• 39

Railroads in historical context: construction, costs and consequences

BIBLIOGRAPHICAL REFERENCES Historical sources - Fundação Museu Nacional Ferroviário Armando Ginestal Machado Centro Nacional de Documentação Ferroviária, Lisbon •

Subfundo da Direcção Fiscal de Exploração dos Caminhos de Ferro - documents relating to the 1880s: traced from the Tua Line, near Foz; General Plan of the Tua Line to Bragança; requirements; lists of land expropriation; transversal profiles; train station projects and guardhouses; drawings of works of art (Bridges, Culverts and Aqueducts).

•

Maps of Expropriation for the construction of the 2nd section of the Tua Line (Mirandela to Bragança)

•

CP Bulletin, news in the year of 1955 on “Improvement of railway services: new railcars for the Tua Line”

•

CP Exercise Reports, years: 1922, 1923, 1924,1925,1927, 1928, 1929, 1937, 1918, 1938, 1941 – 43

•

“Exercise Reports” with the following information:

40 •

 Revenue maps, maps of transported passengers, goods (low and high speed) and general maps of goods  Statistics of passengers by class and departure station on the Tua Line to Bragança;  Detailed maps of operating costs.

Maria Otília Lage

BIBLIOGRAPHY ABREU, Carlos d’; CALVO, Emílio Rivas (2006) «Os projectos para a rede transmontanoduriense de caminho-de-ferro e a linha do Tua», Brigantia, XXVI (1-2-3-4), pp. 345-356.

AGUILAR, José - Carrazeda de Ansiães e seu termo: esboço e subsídios para uma monografia. Carrazeda de Ansiães, Câmara Municipal de Carrazeda de Ansiães, 1980, ALVES, Francisco Manuel Alves (1909-1947) Memórias Arqueológico-Históricas do Distrito de Bragança. Bragança: Câmara Municipal de Bragança / Instituto Português de Museus – Museu do Abade de Baçal. Bragança, 2000. Vols I-XII].

O Caminho de Ferro de Mirandela a Bragança: esclarecimentos prestados aos accionistas....

Lisboa : Typographia de J.F.Pinheiro, 1907.

Companhia Nacional de Caminhos de Ferro Linha de Foz-Tua a Mirandela. - Porto : Lilh: Emilio Biel & Cª , [s.d.]. - [24] f. : todo il.

CASTRO, Francisco - “Um olhar sobre a CP (1945-1980)”, in Gilberto António e Marques Gomes (orgs.), O Caminho de Ferro em Portugal de 1856 a 1996. Lisboa: CP, 1996, 6868.

FERREIRA, Luís, CANOTILHO, Luís – 100 anos da Linha do Tua (1906-2006). Bragança: INATEL, Delegação de Bragança, 2006.

GOMES, Gilberto - Os «Comboios–bloco» (1941–1946).

Disponível em http://www.neps.ics.uminho.pt/aphes28/papers/Gilberto%20Gomes.pdf, consultado em Junho de 2011] GOMES, Rosa e GOMES, Gilberto - Os Caminhos-de-ferro Portugueses 1856-2006. Lisboa: CP – Comboios de Portugal, 2006.

HALBWACHS, Maurice -La mémoire collective. Paris : Albin Michel, 1997, pp. 65-66, 118119 e 131.

LAGE, Maria Otília Pereira Lage -Wolfram = Volfrâmio: Terra revolvida memória revolta.

Para uma análise transversal da sociedade portuguesa (1930-1960). Braga, UM-ICSNEPS, 2003.

LE GOFF, Jacques [et al.] - A Nova História. Lisboa: Edições 70, 1984.

MARTINS, João - A Questão Ferroviária. Estudos ferroviários – Para a história do caminhode-ferro em Portugal. Lisboa: CP – Caminhos-de-ferro Portugueses, EP., 1996.

MARTINS, Oliveira - “Revigorar o Caminho-de-Ferro”, in MOPTC (orgs.) - Reconverter e

• 41

Railroads in historical context: construction, costs and consequences

Modernizar o Caminho-de-Ferro. Lisboa: Gabinete do Ministro, 1987, 1-5.

NUNES, J.P. Avelãs - O Estado Novo e o volfrâmio (1933-1947). Coimbra, IUC, 2010.

RAMALHO, Margarida Magalhães - Comboios com Histórias. Lisboa: Assírio & Alvim, 2000.

REIS, Rogério - A linha do Tua e a estação de Cachão. NM. Mirandela. (31/8/69), p. 1.

REIS, Rogério - Beneficiação da linha do Tua. NM. Mirandela. (30/6/68), p. 1.

REIS, Rogério - Trás-os-Montes e as comunicações ferroviárias. NM. Mirandela. (9/6/68), p. 1.

SERRÃO, Joel - “Breve introdução à história dos caminhos de ferro em Portugal”, in António, Gilberto e Gomes, Marques (org.), O Caminho de Ferro Revisitado - O Caminho de Ferro em Portugal de 1856 a 2006. Lisboa: Caminhos de Ferro Portugueses, 1998, 10-12.

VISEU, Albano Augusto Veiga – Memórias Históricas de um espaço rural:três aldeias de Trás-os-Montes ( Coleja, Cachão e Romeu) ao tempo do estado Novo. Porto: Faculdade de Letras da Univ do Porto, 2007.(tese de doutoramento, policopiado)

42 •

Luís Lopes dos Santos

RAILWAYS IN PORTUGAL AND SPAIN: CORPORATE AND PUBLIC POLICIES OS CAMINHOS DE FERRO EM PORTUGAL E EM ESPANHA: POLÍTICAS PÚBLICAS E EMPRESARIAIS Luís Lopes dos Santos (U. Complutense de Madrid, Spain Espanha) Luís Santos graduated in History in Nova University of Lisbon and completed a PhD (2011) in Contemporary History in the Faculty of History and Geography, Universidade Complutense, in Madrid (Spain).

Luis Santos conclui em 2011 o doutoramento em História Contemporânea na Faculdade de História e Geografia da Universidade Complutense, Madrid (Espanha).

Abstract Resumo

Railways arrived late to the Iberian Peninsula mainly as consequence of the low awareness of the political leaders concerning the reality outside the country and the importance the new mean of transportation was gaining. Generally, the railways network in the peninsula was constructed under two phases more or less simultaneously, interrupted by serious crises resulting from the lack of traffic: one between 1855 and 1865, the other between 1875 and 1891.

However, upon entering the 20th century, the railway traffic that flooded the railway network exceeded the capacity of the transport services supply. This was the period of maximum prosperity for the Iberian railways, with the various administrations having been able to, in a general way, launch investment programs in order to absorb the increase in demand, utilizing only their own resources.

The First World War put an end to this period of prosperity and marked a turning point that changed the way transportation philosophy had been conceived since the introduction of the railway, mid-nineteenth century. Indeed, the conflict lead to a severe increase of railway exploitation costs that lead to loss of profitability since the prices were frozen by the original concession contract. If on one hand, the railways could not maintain the previous profitability levels on the other hand, with the new prices it was no longer possible for the private companies to redeem the investments

• 43

Railroads in historical context: construction, costs and consequences

necessary for a technological update or for the alignment of the supply to the demand.

Faced with the impossibility of interrupting such a vital service, as is the transport service, the State found itself compelled to assume a gradual increase of subsidies granted for a fundamental public service. From this moment on, the railways began an inexorable transfer from from the private sector to the public sector, with the concentration of the national networks more or less directly under the state management near the beginning of the Second World War.

44 •

Os caminhos de ferro chegaram tarde a península Ibérica, em grande medida porque as elites detentores do poder político e financeiro, mantinham escassa consciência da realidade exterior à península, bem como do protagonismo que o caminho de ferro ia adquirindo, em todos os países em que era introduzido. De um modo geral, a rede ferroviária ibérica foi construída em duas vagas mais ou menos simultâneas, interrompidas por graves crises, fundamentalmente resultantes da falta de tráfego. O primeiro desses períodos de máxima intensidade de construções ferroviárias, decorreu entre 1855 e 1865, o segundo entre 1875 e 1891.

Contudo, na transição do século XIX para o XX, pela primeira vez o tráfego aumentou ao ponto de exceder a oferta de serviços de transporte ferroviário.

Este foi o período de máxima prosperidade dos caminhos de ferro ibéricos como negócio, tendo as diversas administrações sido capazes, de uma maneira geral, de lançar programas de investimento de forma a absorver o incremento da procura, recorrendo exclusivamente a recursos próprios. A Grande Guerra pôs cobro a este período de prosperidade do negócio ferroviário, tendo representado um ponto de viragem no conceito que tinha presidido à exploração de caminhos de ferro, desde o momento da sua introdução, em meados do século XIX. O conflito induziu um importante aumento dos custos de exploração, traduzindo-se em perda de rentabilidade, visto que limitadas pelas contratos de concessão, as tarifas não podiam ser aumentadas sem autorização estatal. Se por um lado já não era possível recuperar os índices de rentabilidade anteriores à guerra, por outro lado, à luz dos novos preços, já não era possível amortizar dentro do que restava dos prazos de concessão, os investimentos necessários à atualização tecnológica ou á absorção da procura. Foi neste momento que se começou a falar do “problema ferroviário”.

Perante a impossibilidade de interrupção do vital serviço de transporte ferroviário, o Estado vê-se obrigado a subsidiar em proporção crescente o que se tratava de um fundamental serviço público. A partir deste momento, começa uma inexorável passagem do caminho de ferro da esfera privada para a pública, culminando nas proximidades da Segunda Guerra Mundial, na concentração das redes básicas sob o domínio mais ou menos direto do Estado.

Luís Lopes dos Santos

Railways in Portugal and Spain: Corporate and public policies Luís Lopes dos Santos

INTRODUCTION The arrival of the railways to the Iberian Peninsula was a delayed one when compared to its European counterparts. The inauguration dates for the first lines, October 28th 1848 for Barcelona to Mataró, and 28th October 1856 for Lisbon to Carregado symbolic since it was necessary to wait for mid-’60s of the 19th century to be able to really speak of a significant network in terms of territory coverage.

Generally, the railway network in the peninsula was built in two phases more or less simultaneous, the first starting in 1855 in Spain, and in 1860 in Portugal being finished around 1865 in both countries. The second phase started in 1875 in Spain and in the beginning of 1877 in Portugal, being concluded in 1885 and 1890/1891 respectively.

Even if both countries chose the private management of the railways using the temporary concessions system, the the economic activity of the railways would present a different structure in the neighbouring countries. In Spain, the entire essential network would be built and explored by private sector, even if with some public aid. The direct intervention of the state would only be an exceptional event limited to short side branches dispersed through the territory, which while having strategic importance for the regions they served, had no capacity to remunerate the private capital. On the contrary, in Portugal, the state

• 45

Railroads in historical context: construction, costs and consequences

46 •

would be forced from the very beginning to buy part of the capital of the main private company, Companhia Real, since without its participation, the foundation of this company would not have been possible. Additionally, due to the lack of incentives for the private capital, the state was forced to face the need to directly build and explore the entire basic network north of the Douro and most of the main lines south of Tejo.

Even though both periods were characterised by a significant development of the railway, they ended up in severe financial crisis mainly due to insufficient traffic. A fundamental difference between the Spanish and Portuguese railway concessionaires arised as consequence of the second crisis: the main Spanish companies were forced to stop the remuneration of the shareholders capital, however they were never in danger of bankruptcy. On the contrary, the Portuguese companies such as Real, Nacional and later on Beira Alta, faced the end of amortization of bonds debt entering a technical bankruptcy. This was what drove them to their reorganization, realized under the form of a convention. As a result, the core of their management was transferred from the shareholders to bonds owners mainly of foreign origin (namely French, and to a lesser degree, German), which was reflected in the case of Real and Beira Alta, where a committee located in Paris was in fact managing these companies. The condition for their acceptance banning the companies of making first establishment expenses as well as new bonds issues, and therefore the access to credit was restricted. In the long term, this would prove to be a serious disadvantage, limiting their possibility of technological update as well as in some cases of having a good quality maintenance of the infrastructure and rolling stock.

TRAFFIC INCREASE IN THE BEGINNING OF THE 20TH CENTURY During the 19th century, the railway service supply was always precedent and higher than the demand. However, starting in the beginning of the 20th century, the railway management on both sides of the frontier would be caught by surprise by a significant traffic increase. For the first time, the railways were responsible for a slow, yet effective economic modernization, which was now reflected by the increase in the volumes of transportation.

Nevertheless, the railway administration’s response was to be immediate and adapt to the increase in traffic by using exclusively their own resources. In Spain, an extensive investments program affecting infrastructure and rolling stock was backed up by the issue of new bonds.1 In Portugal, Companhia Real, 1 Cf. MAR MERINO, Maria de, “Tiempos de Acero” in MOPT, Revista del Ministério de Obras Públicas e Transportes; Historia del Ferrocarril en España, 1842-1992, Nº 400, July-August of 1992, pp. 42.

Luís Lopes dos Santos

while still limited by the convention would also put in practice an important investment program using the funds destined to remunerate the shareholders capital as well as to the payment of the 2nd degree bonds which was optional.2 As for the state owned railways, they financed their need using the Special Fund for State Railways, an autonomous organism that depended on public administration to which part of the income was destined.3 It is true that the remaining companies had less resources available, however they were less pressed by the traffic increase and therefore were capable of putting into practice the modernization programs suitable to their needs.4 The period between the beginning of the century and 1914 was essentially a prosperous one for the Iberian railways business. At the end of this period, the CP, new name for Companhia Real since the instauration of the Republic, faced a new problem: the electrification of tramcars in Lisbon explored by Carris raised the question of the competition for Cascais railways since the tramcars had a parallel route between Rossio and Cruz Quebrada. The comparative advantages of the tramcars could no longer be fought using the steam traction which required the line electrification. However, due to its costs this operation was not within the financial possibilities of a company under the Convention which had restricted access to credit. The solution was the concession of the line exploration to another company, Sociedade Estoril, with the condition that the concessionaire should perform the line electrification. The contract was signed in 1918.5

THE FIRST WORLD WAR AND “THE RAILWAYS PROBLEM” The First World War was the key event that would forever change the way the railways were explored since their creation in the mid-19th century.

The entire temporary concession to private companies system was based on the idea that the concessionaire would make an investment for first settlements 2 Cf. CORREIA, António Vasconcelos, A Vida da CP desde o Convénio de 1894, Dificuldades e Soluções; Conferência Realizada na Sociedade de Geografia de Lisboa, em 9 de Junho de 1938 pelo Eng. António Vasconcelos Correia, Presidente do Conselho de Administração da Companhia dos Caminhos-de-Ferro Portugueses, Lisboa, ed. Companhia dos Caminhos-de-Ferro Portugueses, 1938, pp. 11-27.

3 Cf. “Proposta de Lei Apresentada à Câmara dos Senhores Deputados em Sessão de 24 de Abril de 1903”, Lisboa, Ministério das Obras Públicas, Comércio e Indústria; Caminhos-de-Ferro do Estado, 1903, pp. 9-10.

4 Cf. “Caminhos-de-Ferro” in Revista de Obras Públicas e Minas, Lisboa, Tomo XXXII, Nº 373375, January to March 1901, ed. Associação dos Engenheiros Civis Portugueses, 1938, pp. 11-27.

5 Cf. CP Archive, Lisboa, “Arrendamento e Electrificação do Ramal de Cascais” in Actas do Conselho de Administração da Companhia dos Caminhos-de-Ferro Portugueses, Sessão de 9 de Maio de 1914, Nº 26, 18 of April 1913 to the 28 of June 1915, pp. 202-203.

• 47

Railroads in historical context: construction, costs and consequences

48 •

which he expected to recover with interests along the concession term. Ending this term, the lines would revert to their initial owner, the State, free of financial charges.

In order to avoid overcharging the users, the maximum prices were fixed in the concession contracts and could be not overpassed without authorisation from the state. The problem laid with the fact that the first settlement expenses were never a closed chapter in a railways concession. This needed regular investments not only destined to absorb the increase in the demand, but to also keep the pace of the technological evolution of the equipment. It was obvious for the temporary concession system that a point would come where the proximity of the concession term would prevent the concessionaires to further invest in the technological update of the system due to the impossibility of recovering their investment on time. Additionally, there was the potential effect of a sudden increase in the exploitation costs on the financial equilibrium of the railway companies since they could not increase the prices to match a sharp increase in the inflation rate.

The war effect served to point out sooner the limits of the temporary concession systems: as a result of price inflation, the cost of fuel and other materials necessary to the exploitation increased over 100%, reaching in some cases 1000% resulting in a profitability loss for the railway companies. In addition to this, in light of the new costs and the consequent profitability loss it was not possible to recover the investments needed for the good functioning of the system within the concession period. From this moment on, the railways lost the capacity of surviving on their own.6 Initially, the state and the concessionaires did not realize that they were facing a structural problem and that the temporary concessions system had reached its limits without possibility for its renewal. For the companies, the problem was reduced to the need of updating the maximum prices in accordance to the inflation rate, keeping the essence of the railways business unchanged. However, the solution was not a peaceful one under the circumstances: the railways were the only terrestrial transportation mean and therefore any increase in the prices practices would reflect on the cost of living and economic competitivity as occurs nowadays with oil price. Therefore, it is not surprising that the government would be reluctant to approve any increase of prices since this action, in addition to being unpopular, it would bear effects on the competitivity of a various range of economic sectors. On the other hand, it was not possible to let the railway companies go bankrupt, since transport service was essential to society as well as to the entire economy. It is from this moment on that the “Railways Problem” becomes a topic.

6 Cf. IZQUIERDO, Rafael, Cambó y su Visión de la Política Ferroviaria, (El Inicio de un Cambio), Madrid, ed. Fundación de los Ferrocarriles Españoles; Colegio de Ingenieros de Caminos, Canales y Puertos, Colección de Ciencias, Humanidades e Ingeniería, Nº 64, 2000, pp. 95.

Luís Lopes dos Santos

In the peninsula, the first study made dedicated to finding a structural solution for the “Railways Problem” was made by Francesc Cambó, Development Minister in the Maura government named by King Afonso XIII on 21st March 1918. His project was the first that forecasted a breakout from the legal framework that regulated the railways sector in Spain since the 19th century. Additionally, it presented some visionary measures that only decades after would be implemented in Europe. His proposal consisted in rescuing the railways network followed by its concentration under big blocks corresponding to the main traffic flows. These would match sensibly the influence areas of the main active concessionaires in North, South, West and East.

From then now on, the split between the infrastructure management and service exploitation was forecasted: while the first was the state’s responsibilty, the second could be an alternative managed by the private sector under a renting agreement. Under these premises, he sustained that since the state would have to face the costs of the railway services, the best solution was the recovery of the lines and their concentration in order to obtain scale economy. The state would be in charge of maintaining and modernising the infrastructure as well as adding new lines to the network. Even if the short existence of the Maura governance would prevent the plan’s approval, the Cambó project profoundly influenced all the future studies in Spain. Additionally, he was one of the few that counted with the railway companies’ approval, since their assets and capital was favourably evaluated in terms of rescue payments. Moreover, they had the possibility to regain the exploitation of a network optimized by the state and could concentrate exclusively on this service once free from the weight of the infrastructure costs.7 For the moment being, the only measure that was implemented given the need to maintain the railway companies working was a price increase of 15% approved in December 1918. This was a key moment for the history of the Spanish railways since the state considered the income resulting from this increase as a subsidy. Consequently, the companies could not use these amounts as an ordinary income. The hidden goal was to prepare the rescue of the network under conditions favourable to the state: effectively, the redemption amount was calculated as an average of the net income obtained in previous years at the moment when the process was initiated. By reducing the net income, the state was subsequently reducing the amount it had to pay for an anticipated redemption of the concessions.8 Since the Portuguese state was under financial difficulties, similar solutions were adopted with a price increase of 10% in 19159, of 25% in 1916, of 40% in 7 Cf. SANTOS, Luís António Lopes dos, Política Ferroviaria en España; Desde Primo de Rivera a la Guerra Civil, Madrid, Study of research capability, as a part of the Madrid Complutense Univesity phD. Program, 2002, pp. 49.

8 Cf. Ibidem, pp. 46-47.

9 Cf. FERNANDO DE SOUSA, J., “Alteração das Sobretaxas” in Gazeta dos Caminhos-de-Ferro,

• 49

Railroads in historical context: construction, costs and consequences

50 •

1917, and 57% in 1918.10 With each increase, the cost of living was affected and therefore the price increase effects were cancelled, leading to the need of another price increase. As a most obvious result of these measures, we may point out that an increase of 157% on staple foods was noticed in Lisbon.11 While the railway service was maintained, it cannot be denied that it was being deficiently performed.

The last consequence of the First World War on the Iberian railways would be the switch of the main railway shareholders capital to the peninsula. The depreciation of the French currency created the ideal circumstances for buying capital from the main concessionaires who were of French origin. However, the process was distinct in the two countries. In Spain the private capital and mainly the most important banks took the initiative to buy the main companies, which was the case of Banco Bilbao and Banco Vizcaya which would become the owner of the Compañia del Norte de España.12 The process was part of a strategy that aimed to integrate under the same group as the Spanish factories of railways equipment (also bought by the banks) and the concessionaires. The lobby thus created would have a decisive weight in the Spanish railways policy from now on. On the contrary, in Portugal, no private capital was interested in buying companies whose activity was regulated by the Convention, which were unable since a long time to remunerate the shareholders investments. It was the State who will take the chance to increase its participation to CP capital. The decision came as a result of pressure from public opinion, which criticized the excessive weight of the foreign bonds owners in the company. From then on, as a major shareholder of CP and owner of the Minho and Douro and South and SouthEast line the state would become the main railways actor in the country.13 This fact also had a significant influence on the Portuguese railways policy.

FIRST ATTEMPTS TO OVERCOME “THE RAILWAYS PROBLEM” The end of the First World War brought no significant improvement to the railways situation since the prices were maintained frozen and could not be increased without the state approval while the costs were constantly increasing.

Lisboa, year 37, Nº 872, 1 of April 1924, pp. 80.

10 Cf. CP Archive, Lisboa, “Sobretaxas”.

11 Cf. FERNANDO DE SOUZA, J., “Providência que se impõe” in Gazeta dos Caminhos-de-Ferro, Lisboa, year 31, Nº 728, 16 April 1918, pp. 48.

12 Cf. SANTOS, Luís António Lopes dos, Op. cit., pp. 48.

13 Cf. PINHEIRO, Magda, “Les Chemins de Fer Portuguais: Entre l’Explotation Privée et le Rachar” in Revue d’Histoire des Chemins de Fer ; Actes du 5éme Colloque de l’Association Pour L’Histoire des Chemins de Fer en France en Collaboration avec l’Union Internationale des Chemins de Fer, Nº 16-17, Spring –Autumn 1997, ed. Association pour l’Histoire des Chemins de Fer ; Union Internationale des Chemins de Fer, pp. 156.

Luís Lopes dos Santos

In Spain, the implementation of the legal work day of 8 hours made the state support the cost increase derived from this regulation, since the concessionaires had no means to do it. In 1920 the evolution of the situation lead to the threat of strike from the railway workers, forcing the state to directly support an increase in wages. On October 15th of the same year, the public intervention reached new dimensions since due to the traffic increase which the railway companies were unable to balance, the state decided to assign funds destined to sponsor the purchase of materials. Since the concessionaires were not able to keep the pace of the increasing exploiting costs, the state was forced to give subsidies to the railways in order to prevent the cessation of a vital transportation service. Its intervention was no longer limited to assuming the financial costs of the system which were out of the companies reach, but was extended to financing the adaption of transportation capacity to the increasing demand.14 In Portugal, social measures such as the legal 8 hours work day or wages increase were accompanied by the depreciation of the escudo until 1924 further degraded the railways companies’ financial situation. Not only did the bonds debt have to be paid but the fuel and metallurgical materials necessary for exploitation also had to be imported. With each further depreciation of the escudo, the financial health of the railway companies grew worse, increasing the financial burden and the prices of the materials. It is not therefore surprising that CP, like the majority of the private concessionaires, was forced to suspend the payments of the bonds debt. The public owned companies were also facing a deficient situation which maintained until 1922 for SS and 1924 for MD.15 In both countries as well as in the entire Europe, the State realized that from then on, it would have to face the increasing functioning costs of the railway system. Therefore, its attention turned more to a concentration that would optimize the scale economy and to the definition of a legal framework for its investments than to the creation of new lines.

In consequence of this evolution, three new projects of railways legislation were studied in Spain after the war, all influenced by the Cambó project. They are known as La Cierva, Maura and Argüelles. All of them were discussed in the Parliament and one of them was partially put in practice. However, in opposition to their predecessor, none of these projects counted with the approval of the railway companies; effectively their activity until the rescue term was more subject to the State vigilance and intervention materialized by the participation of State representatives in the Boards of Directors and the redemption formula was not to their advantage. While the Cambó project allowed them to recover 14 Cf. MUÑOZ RUBIO, Miguel, “El Estado como Empresario Ferroviario” in MUÑOZ RUBIO, Miguel; SANZ FERNÁNDEZ, Jesús; VIDAL OLIVARES, Javier, Siglo y Medio de Ferrocarril en España, 1848-1998; Economía, Industria y Sociedad, s.l, ed. Fundación de los Ferrocarriles Españoles; Instituto de Cultura Juan Gil Albert; Diputación Provincial de Alicante; Caja de Ahorros del Mediterráneo, 1999, pp. 305-306.

15 Cf. PINHEIRO, Magda, in OP. cit., pp. 156

• 51

Railroads in historical context: construction, costs and consequences

52 •

the network exploitation, this aspect was not clear for the following projects since no specific mention on who would assure the exploitation at the rescue deadline was included.16 In Portugal, the government nominated in 1919 a specific commission destined to perform a study on the “Railways Problem”. While never published, its conclusions were disseminated in the railways and financial environment having been discussed by Gazeta dos Caminhos de Ferro. Given the impossibility to renew the previous formula of temporary concessions, the Government aimed at the rescue of private-owned networks and their concentration in a common exploiting system.17 While the subject was being studied, in January 1920, there was a need to approve new prices and the overtaxation of 50% would reach 100% in March.

However, the most important aspect of the new overtaxation was the contents of the Law nº 952 that approved it. This established that the revenue originated by the overtaxation could not be booked by the companies as income, being the state in charge of defining its use. It was essentially a subsidy belonging to the public finance. First of all, the funds were used to cover the deficit, secondly to finance first settlement works and thirdly to cover financial costs. The amounts exceeding the needs for these three chapters were reintegrated in the Treasury.18 One of the goals of this procedure was to force the companies to make the necessary works in order to increase the capacity of the networks they exploited. However, as is the case in Spain, the final goal was to reduce the amount to be paid as redemption by decreasing the concessionaires net income. The State was looking to create the necessary conditions for a rescue at the lowest value possible.19 The concessionaire’s reaction was immediate and complete using a violent press campaign. The state intentions were labelled as theftas the irregular nature of the redemption formula was denounced. On the other hand, a campaign destined to denigrate the state railway administration was initiated by comparing its inefficiency with the situation of the private-owned companies. In mid-1921 after facing the public hostility, the Government had dropped the idea of the redemption.20 However, due to the constant increase in costs and wanting to maintain the service running, the increase in prices was further used in order to allow the railway system to survive, even if barely. In 1920, an overtaxation of 200% was approved in January, followed by 300% in September of the same year, and 500% to be maintained in December 1922. In 1923, a new overtaxation of 700% 16 Cf. SANTOS, Luís António Lopes dos, Op. cit., pp. 48-50.

17 Cf. FERNANDO DE SOUSA, J., “O Resgate” in Gazeta dos Caminhos-de-Ferro, Lisboa, year 33, Nº 796, 1 of January of 1920, p. 2-4.

18 Cf. FERNANDO DE SOUSA, J., “A Propósito do Resgate” in Gazeta dos Caminhos-de-Ferro, Lisboa, year 33, Nº 776, 16 April 1920, pp. 113-115.

19 Cf. CUNHA LEAL, “A Política Ferroviária Portuguesa Depois da Guerra” in Revista Portuguesa de Comunicações, Lisboa, year 1, Nº 1, 1 July 1929, pp. 4-5.

20 Cf. FERNANDO DE SOUSA, J., “A Propósito do Resgate”in Op. cit., pp. 115.

Luís Lopes dos Santos

was approved for state-owned railways and of 600% for private-owned railways.

In order to decrease its impact on the cost of living, some primary necessity goods were exempted of the new overtaxation being applied lower percentages.21 Until 1923, it was obvious that the measures taken by the two countries had allowed maintaining the functioning of the basic railway system, even if with a huge financial effort and a deficient service. However, everybody was aware that the solutions put in practice until that time were mere palliative measures to a structural problem.

FIRST ATTEMPTS FOR A STRUCTURAL SOLUTION TO THE “RAILWAYS PROBLEM” The first attempt to find a structural solution to the “Railways problem” in Spain was performed by the dictatorship regime of Primo de Rivera which gained power after the coup of 13th September 1923. The key figure of the railways policy under the military dictatorship were the Railways Statutes from 1924 which would mark a breaking point with the previous legal framework in Spain in place since the 19th century. In the name of common interests, the state and the concessionaries would work together in the exploiting of the railways network.

The state was in charge of all the first settlement expenses both related to infrastructure and rolling stock. On their part, the companies were dedicated to the exploiting of the network. The state investments gained legal statute which was reflected in the sharing of results proportional to public and private capital.

The financing would be assured by an autonomous organism, Caixa Ferroviária do Estado, whose main income source was the state Railways Debt. A Superior Railways Council acted as liaison between the State and the companies being in charge of the enforcement of the Statutes and the entire railway policy as well as of the management of the Caixa Ferroviaria. The effects generated by the works and renewal occurred in the railways system were directed to the internal industry.

The Superior Railways Council was also responsiblefor making a study regarding the current network concentration in order to potentiate scale economy to maximise the state efforts. In exchange for a certain loss of independence of the concessionaires, the state was recognizing the need for the prices to cover all exploitation costs allowing the capital remuneration. It was assumed the obligation to study the possibility to apply new prices that would allow the railway network to sustain itself without the need for subsidies.22 In the following years, a real modernisation of the existing network was noticed as well as the construction of some new lines offering the companies as 21 Cf. “A Nova Sobretaxa e a sua Aplicação” in Gazeta dos Caminhos-de-Ferro, Lisboa, year 36, Nº 859, 1 October 1923, pp. 6.

22 Cf. Royal Law -Decree from 12 July 1924.

• 53

Railroads in historical context: construction, costs and consequences

54 •

much their best years in terms of income. The goal of the investments promoted

by Primo Rivera was to improve the transportation service but also to boost the development of certain basic industries. However, the companies began to lose their confidence as the sensitive problem of the new prices was always delayed.

Concerning Portugal, the conditions would change starting with Marshal Carmona’s coup in 28th May 1926. The actions taken by the recently implemented dictatorship concerning the “railways problem” would focus on the network concentration recovering a formula already studied in 1924. This approach would revert the terms of the equation handing over the state railways to CP under the renting form. Thus, the state would group the main part of the basic network overcoming the rescue obstacle and maintaining significant control over the system due to its major participation to the company capital whose owner in fact was the state.23 The entire operation was organized under the form of a public tender held in 1927 whose terms were negotiated exclusively between the State and the company Board of Directors. It was therefore no surprise that CP won the public tender becoming the owner of two state networks and the entire capital of other companies which already belonged to its Group: Caminhosde-Ferro Meridionais, and Caminho-de-Ferro do Mondego.24 For now, the only networks still out of its direct exploitation were Beira Alta and the small private networks of narrow gauge.

More than a railway policy option, the primary goal of the state-owned railways renting was to dismember the trade union structure that dominated its workers and which being a strategic element for military troops movements, had backed up all the military coups since the implantation of the Republic including the one occurred on 28th May. The goal of the army was to eliminate a potentially threatening forces. It was mainly a strategy to maintain the power.

In a smaller way, the renting was also helped by the pressure made either by groups that were interested in the exploitation - which is the case of Alfredo da Silva, owner of the CUF - or that considered their activity affected by deficient transportation systems as was the case of the Commercial Association in Lisbon and Oporto.25 The prosperity of the Spanish railways under the Statute would be short lasting. With the investments of half of the State Railways Debt, 1,300 out of 2,600 million pesetas, the the possibilities of the Spanish capital market were drawn.

Beginning at the end of 1928, the government began a series of negotiations with the two main concessionaires (Norte and MZA) admitting that there was 23 Cf. FERNANDO DE SOUZA, J., “O Arrendamento das Linhas do Estado” in Gazeta dos Caminhos-de-Ferro, Lisboa, year 40, Nº 948, 16 of June of 1927, pp. 171.

24 Cf. CP Archive, Lisboa, “Aquisição das Linhas dos Meridionais” in Actas das Sessões do Conselho de Administração; Nº 385, Sessão de 29 de Dezembro de 1926, Nº 31, 30 of March of 1926 to 19 of December of 1927, pp. 163.

25 CF. BARRETO, José, “Sindicalismo e Política nos Caminhos-de-ferro, 1872-1961” in coord. by GOMES, Gilberto, Estudos Históricos I, Lisboa, ed. por CP, Caminhos-de-Ferro Portugueses, colecção Para a História dos Caminhos-de-Ferro em Portugal, vol. 2, 1999, pp. 133-134.

Luís Lopes dos Santos

no possibility of maintaining the Railway Statutes. At the same time, due to its social impact the State could not allow the application of new prices. Thus, its goal was to renegotiate a new legal framework returning the first settlement costs to the companies. At the same time, the state intended to receive the return of the capital invested in the railway system. The companies rejected this proposal, which put them in a disadvantage to the State. The fall of the dictatorship interrupted the negotiation and the virulence of the discussion around the “Railways problem” increased.26 During General Berenguer’s government, the payments originated by the Caixa Ferroviaria were scarce and at the same time, the modernization and increase of the network almost slowed to a stop. At the same time, the negotiations for the new legal framework kept going and the state kept insisting on the points already defended in the last phase of the dictatorship: return of the first settlement cost to the concessionaires, no approval for changes in maximum prices, and remuneration of the state capital. The possibility to maintain the investment flow was dependant on a concentration that would allow the maximization of the effects of the state financial effort. However, the fall of the monarchy prevented to reach a solution.27 The disappointment provoked by the results of the railway renting appeared at the same time that Primo de Rivera railway policy fell. When in 1928 the financial results of the state owned networks were published, unknown until the moment, it was discovered that MD and SS had positive results before the renting. This reality was in contrast with the subsequent situation since the state began to undergo losses in the state-owned networks.

The very nature of the renting contract was frustrating all the goals of an amalgamation: it was a 30 year contract after which the state recovered the exploitation of the lines. For that to happen the people, the mobile equipment and the workshops should have stayed active and at their original places making impossible the rational concentration of resources along the network exploited by CP in accordance to traffic needs. Even the need to control the contract fulfillment as well as the amounts to be charged by the state required the management and accounting to be maintained separately. Additionally, it was not possible to start a modernization plan since CP, like BA, was a company regulated by the Convention without the possibility to access credit. In turn, the state did not want to disburse any amounts which even if it would benefit the public, it would also revert to the foreign bond owners. As an immediate result of the renting, the entire basic network was now in the hands of foreign bond holders committees.28 26 Cf. El Problema Ferroviario. Estudios Realizados y Medidas Legislativas Adoptadas en España Para Intentar su Resolución, 1918-1932, Madrid, ed. by Compañías del Norte de España and Madrid a Zaragoza y a Alicante, September 1932, pp. 83.

27 Cf. Idem, pp. 97-98.

28 Cf. “O Estado e o Problema Ferroviário” in Revista Portuguesa de Comunicações, Lisboa, year 1, Nº 1, 1 June 1929, p. 28-30.

• 55

Railroads in historical context: construction, costs and consequences

THE FIRST TWO YEARS OF THE SPANISH REPUBLIC, THE CP DEBT CONVERSION AND THE STRENGTHENING OF THE PORTUGUESE STATE CONTROL OVER THE RAILWAYS.

56 •

The “Railways problem” was part of the priorities of the left provisional government of the Second Spanish Republic, with Albornoz as Minister of Development. On May 1931, a committee in charge of elaborating a new study on the railways legal framework was nominated with the goal to replace the 1924 statutes. Within the committee agenda, it was mentioned that the railways should be directed towards nationalisation, network concentration and an increase of the state participation in its management.

The need for the state to start receiving the return of its participation to the companies’ capitalm as well as for regulation of the labour relations, the resulting project was known as the “Albornoz Project” and was approved by the Council of Ministries in October 1931. However, it never came into force.29 Nevertheless, its fundamental principles were put in practice by the next Government which resulted from a bad coalition of left-wing parties such as PSOE, with Indalecio Prieto named as Public Works Minister.

As an example, we may point out the decision taken by Prieto in May 1932, according to which for the first time the state was able to withdraw from exploiting lines whose costs were not related to social benefits. This measure was taken to avoid overcharging the state with deficient lines abandoned by the concessionaires.30 The local public administrations were encouraged to collaborate with the state in bearing these costs, in case they pretended to keep them running.31 In July 1932, another important decision was taken related to the possibility of increasing the prices by 3% in order to be able to finance an increase in the railway workers’ salary.32 However, perhaps the most important measure taken by Prieto, equally included in “Albornoz project”, was the decision to introduce state representatives in the companies’ Boards of Directors having a veto right on all decisions as of September 1932. To this decision a declaration of incompatibility between top management position in railway companies and companies producing railways equipment was made.33 The government was not limiting its action to increasing the control over the railways management, but was also trying to break the promiscuous relationships between the manufacturers of railway materials and the concessionaires, sources of many expenses to the state finances.

Initially limited to the companies explored as a state consortium, the measure 29 Cf. Decree of 30 May 1931.

30 Cf. Law of 10 May 1932.

31 Cf. Decree of 29 July 1932.

32 Cf. Law of 7 July 1932.

33 Cf. Law of 9 September 1932.

Luís Lopes dos Santos

was soon extended to the companies, which had not adhered to the 1924 statutes reaching the mention of incompatibilities with political functions.34 At the same time that the state control was increasing, a study was being made concerning a project for the network concentration and solutions for making the public bodies immune to pressures from the railway companies that were being developed.35 Within the same period of time, the Portuguese railway policy advanced significantly: the topic of the moment was to release CP from the Convention bonds a necessary condition to allow the state to assume the first settlement expenses without transferring the benefits of its financial effort to the foreign bonds owners. A new depreciation of the escudo gave the pretext to negotiate a new Convention. This consisted in exchanging old bonds for others with lower nominal value however bearing higher interests. Even if on the short-term the financial costs increased, the liabilities decreased and the company had conditions to guarantee immediate payments. To compensate, the Board of Directors where the state had the majority, was recovering the company control as well as the possibility to access to credit.36 All these events occurred in a period of significant road network expansion as opposed to railway systems that were becoming gradually obsolete, which was reflected in the traffic decrease. As the company situation was degrading, the state was forced to give it subsidies even if in a hidden form and without assuming this obligation.37 The state importance in the Portuguese railway sector further increased with the rescue of the Companhia do Norte de Portugal in 1933. Since its foundation, the company was maintaining an extensive modernization program including the construction of new narrow gauge line right when traffic was reducing due to the competition brought on by the automobile. The concession contracts were forcing the state to guarantee high interest rate for the exploitation of its line which became a heavy burden for the Public Treasury. In mid-1933 the works undergone left the company with a temporary treasury problem, which favored the rescue by the government which immediately replaced the Board of Directors by a Management Committee appointed by the state. One of the missions assigned to this Committee was to perform new statutes and concession contracts under conditions less restrictive for the state.38 In addition to the immediate goal of reducing costs with the company, functioning its rescue also served the purpose of network concentration.

34 Cf. Order of the 13 December 1932.

35 Cf. Decree of the 13 December 1932.

36 Cf. “Como vão Ser Remodelados os Estatutos da CP” in Revista Portuguesa de Comunicações, Lisboa, year 3, Nº 58, 15 November 1931, pp. 424.

37 Cf. BARRETO, António, Op. cit., pp. 134-140.

38 Cf. “Companhia dos Caminhos-de-Ferro do Norte de Portugal” in Revista Portuguesa de Comunicações, Lisboa, year 5, Nº 83, August of 1933, pp. 366-367.

• 57

Railroads in historical context: construction, costs and consequences

LERROUX CABINET, CEDA GOVERNMENT AND THE PORTUGUESE RAILWAYS TRANSITION TO AMALGAMATION.

58 •

The Spanish railway policy changed course in September 1933 when the Lerroux Cabinet came to power, with Guerra del Rio nominated Public Works Minister, marking the beginning of a two-year period dominated by right-wing parties. Once again, a committee dedicated to produce a new project of legal framework was nominated. A polemic discussion had been started between the users and the representatives of the railway companies. For the users, the railways were essentially a public service, and they defended the lowest possible prices. For them, the profitability issue was irrelevant.39 The representatives of the companies on the other hand, wanted the enforcement of the railway statutes until its last consequences, and at the same time they were requiring coordination measures with automobiles.40 Guerra del Rio’s project for a railway legal framework was close to Cambó’s principles, since he was proposing the rescue of the network, the transfer of first settlement costs to the state, the partial networks concentration based on geographical criteria, as well as the rental of the exploiting to private companies. As a novelty, it also presented a set of coordination measures between the railways and the new transportation mean the automobiles. Nevertheless as it become soon evident the companies could not accept a redemption formula that left them in technical bankruptcy.41 Since the legal project was never approved, the most important measure taken by Guerra del Rio government was a price increase by 15% destined to absorb the deficit and allow the payment of the financial charges.42 It equally authorized the issue of two series of treasury bonds of 25 million pesetas each destined to finance first establishment expenses for Norte and MZA. At a time when the companies were on the verge of bankruptcy, the CEDA government, with Lucía as Public Works Minister, presented what would be the third project for railway legislation under the republic. This project intended to approximate the current situation to the conditions of the original concessions as much as possible, freeing the State from the costs of the railways services and giving back their management freedom to the companies trying to create the conditions necessary for their self-sufficiency.

In order to overcome the lack of profitability, no new prices were established.

Instead, the prices fixed in the original concession contracts updated using a formula based on real prices. The three overtaxations proved in the meanwhile, 39 Cf. SANTOS, Luís António Lopes dos, Op. cit, pp. 230-231.

40 Cf. Ibidem, pp. 228-230.

41 Cf. ARTOLA, Miguel, “La Acción del Estado” in ARTOLA, Miguel, Los Ferrocarriles en España, 1844-1943; 1. El Estado y los Ferrocarriles, Madrid, 1º volume, ed. por Servicio de Estudios del Banco de España, 1978, pp. 436.

42 Cf. Law of the 29 May 1934.

Luís Lopes dos Santos

were also maintained. The financing problem was overcome by authorising the concessionaires to issue bonds payable after the concession reversal date, being that their financial services were assured by the state after that date. Finally the State was giving up the direct exploiting of railways and at the rescue date was handing them over again to the private sector using a public tender call.43 With the profitability and the financing problems solved and before the perspective to recover the service exploiting once the concession was over, the concessionaires started to lobby for the approval of the Lucía project.

Meanwhile, in Portugal, under a politically stable environment with the consolidation of the “Estado Novo” regime, no discussions concerning the railways legal framework occurred. All concessionaires were in bad financial shape maintaining CP its regular functioning only due to more or less hidden public aid. Concerning the remaining companies, since public services were not interrupted, the political power was not interested in helping them out, and even went as far as suggesting that it was waiting for their financial breakdown in order to impose the amalgamation as a condition for direct state aid.44

THE POPULAR FRONT RAILWAY POLICY IN SPAIN AND THE LACK OF INVESTMENT IN THE PORTUGUESE RAILWAYS. When the Popular Front rose to power, social tensions were increasing and the railway companies - even the bigger ones - were on the verge of financial breakdown and walking towards bankruptcy.

The main problem related to the railways that the Popular Front had to face was the rescue of the railways company Andaluces. The concessionaires bankruptcy was a common fact at the time, however this was the second time when an important company had to be rescued (after MCP-Oeste in 1928). The solution found was to hand over its exploitation to Oeste, partially state owned, and maintain it as an independent company. Nevertheless, a Board of Directors was nominated in charge of taking all strategic decisions.45 Facing bankruptcy, the companies and the business interests associated to them started making pressure for the application of the Lucía project. However, the Public Works Minister, Velao Oñate, presented a conjoint management project between the state and the concessionaires, represented in practice by the permanent intervention of the state representatives in the Board of Directors 43 Cf. DE LA TORRE, Enrique, Anuario de Ferrocarriles, Madrid, Autoedición, 1935, pp. 257-265.

44 Cf. CP Archive, Lisboa, “Companhias de Via Reduzida –Vale do Vouga e Nacional” in Actas das Sessões do Conselho de Administração; Nº 521, Sessão de 1 de Janeiro de 1935, book Nº 36, 2 of November of 1934, to the 27 Jannuary 1937, pp. 19.

45 Cf. COMÍN, Francisco; MARTIN ACEÑA, Pablo, MUÑOZ RUBIO, Miguel, VIDAL OLIVARES, Javier, Op. cit., pp. 316.

• 59

Railroads in historical context: construction, costs and consequences

with veto rights meaning that in practice the states was assuming full control of the companies without paying any redemption. The decision was based on the argument that with all the investments and subsidies made since the First World War and mainly under the railway statutes, the state was in fact the true owner of the railways, its participation to the capital being equal to that of the private interests. The companies Norte and MZA contested the demand while trying to convert their debts, distribute the reserves to the shareholders, and transform in new companies with a clean financial situation.46 The civil war ended this process.

In Portugal, the discussion was centered on the technical obsoleteness of a network without significant modernisation since the First World War. Its lack of efficiency was now underlined by the savage competition made by the road transport. However, the companies had no resources to invest and therefore required the state intervention which kept maintaining its action to a minimum necessary to guarantee that the service existed, concentrating its attention on road transport and sea ports.

60 •

THE FINAL SOLUTION TO THE IBERIAN “RAILWAYS PROBLEM” In Spain, the civil war sped up a solution to the “Railways Problem” under General Franco’s regime. The conflict completed the technical and financial degradation of the railways companies started by the First World War and enhanced by the republican phase. Once the conflict was over, the network was severely damaged and depreciated, and the concessionaires bankrupt and unable to recover or maintain the exploitation on their own. In addition to this scenario, Spain was under a dictatorial regime, almost almighty, winner of a bloody civil war which could not be subject to any mechanism of control or restriction. The circumstances allowed the government to impose the rescue of the exhausted companies under favorable conditions for the state.47 On the other hand, the assumption that rejected the constitution of public companies considering the state as incapable of an efficiency management was overcome in all of Europe by the formation of a managerial public sector grouping strategic domains having as main goal to support the private initiative. This international evolution matched the own dictatorship vision freshly out of the war under the command of General Franco. Transport was a strategic sector which the regime wanted to control in the name of common interests and the 46 Cf. ARTOLA, Miguel, “La Acción del Estado” in Op. cit. pp. 442.

47 Cf. ORTUÑEZ GOICOLEA, Pedro Pablo, El Proceso de Nacionalización de los Ferrocarriles en España. Historia de las Grandes Compañías Ferroviarias, 1913-1943, Valladolid, ph Degre thesis, Valladolid University, Facultad de Económicas y Empresariales; Departamento de Historia e Instituciones Económicas y Economia Aplicada, 2003, pp. 451-452.

Luís Lopes dos Santos

need to support the private initiative. In this context, a double role was being assigned to the railways: on one hand, it should transport raw material and finished products at the lowest possible price even if the entity that exploited the service made no profit, and on the other hand, by its reproductive effects over the economy it should enhance an entire range of basic industries or their suppliers such as siderurgy, metallurgy, mechanics, electrical and chemical industry mining, etc.48 The normal history of the railway companies would in fact end in May 1939 when the government forced the resignation of the Boards of Directors, replacing them by Management Committees named by the state itself where the private capital was not represented, an action which nearly bordered expropriation. The process was completed by the Law on the Basic Railways and Road Transport Organisation issued in January 1941. This gave legal shape to the anticipated rescue of all broad gauge companies (in practice already performed in 1939) which was integrated into RENFE, a public company in charge of the exploitation of the entire network. As to the narrow gauge lines, even if they remained temporarily out of the nationalisation process, their financial and technical deterioration as well as the competition of the road transportation lead to their slow integration in the FEVE during the following years. The Spanish “Railways Problem” was thus definitively solved.49 The Portuguese “Railways Problem” would find a solution due to a series of apparently isolated events which proved to be related and which occurred between 1933 and 1951.

In 1933, the polemic rescue of the Norte de Portugal company came to an end. Since the company owned 12,000 shares waiting to be placed, the Management Committee nominated by the government sold them to the state, at an advantageous price. Thus, the state became the company’s true owner. This advantage allowed the State to call for a General Assembly which approved new statutes and contracts.50 In 1943, the financial reorganisation of the Beira Alta Company occured under circumstances similar to the ones that occurred for CP: original bonds were exchanged for new titles with a lower nominal value and higher interest rate. The situation was favorable to negotiations due to the depreciation of the French currency. Even if short term financial costs increased, the liabilities decreased and the company was able to assure immediate payments. The Board of Directors recovered its autonomy as well as the access to credit and the company financial situation was solved.51 48 Cf. Ibidem, pp. 454.

49 Cf. COMÍN, Francisco; MARTÍN ACEÑA, Pablo; MUÑOZ RUBIO, Miguel; VIDAL OLIVARES, Javier, Op. cit., Vol. II, pp. 16-19 50 Cf. FERNANDO DE SOUSA, J., “Companhia dos Caminhos-de Ferro do Norte de Portugal” in Gazeta dos Caminhos-de-Ferro, Lisboa, year 49, Nº 1.192, 16 of August of 1937, pp. 397.

51 Cf. “Companhia dos Caminhos-de-Ferro da Beira Alta” in Gazeta dos Caminhos-de-Ferro, Lisboa,

• 61

Railroads in historical context: construction, costs and consequences

62 •

The final solution began to take shape with the Law no. 2,008 issued in September 1945, also known as The Law for the Railways amalgamation and terrestrial transportation coordination. With this law, the state proposed to grant a single railways concession to a company that included all the existing public service lines. Additionally, without stopping trying to create conditions for the railways to reach self-sufficiency, the state assumed all first settlement costs, the infrastructure modernisation, the exploiting costs, and the equipment costs in case of a sustained deficit which would effectively happen.52 The originality of the Portuguese case consisted in concentrating the entire railway network under state control without ever paying the ransom and under the appearance of a private company. Unable to rescue CP, the state was in fact its owner. It started by handing over two networks that it exploited directly. At this time, it succeeded even if with some deficiency to perform an amalgamation of the basic network. The remaining companies were nothing more than small companies with low settlement costs and easy to absorb. The CP debt conversion allowed the state to put in evidence its dominance within the company. The subsequent rescue of the North Company made possible the integration of the most important narrow gauge network. Only the need to free BA from the Convention in order to be able to integrate the new organization with a healthy financial situation was still representing an obstacle, which was then finally surpassed in 1943. Thus, conditions were created for CP to absorb BA, Nacional, and Vale do Vouga. The state contributed with its own lines and with Norte de Portugal which it already controlled. The exploitation of the single network started on January 1st 1947. The single concession contract given legal form the already existing situation was signed May 1951.53 The Portuguese “Railways Problem” was finally solved. Even if CP, a technically private company, was having a subordinate position to the state, as clear as the RENFE position: its dominance was not only assured by the majority of the shareholder capital but by the impossibility to survive without financial aid.

Only the Cascais line remained out of the renting contract and the single exploitation due to the high redemption amount. It was a small suburban branch serving Lisbon which was functioning more or less isolated from the remaining network. Nevertheless, the renting contract was not necessarily in contradiction with the single concession since the concessionaire was CP and the line integrated with all the others in the concession contract while its exploitation was rented to another company. This reality did not prevent the line from being integrated into the single exploitation once the renting contract was over , which occured in 1976. year 49, Nº 1.331, 1 June of 1943.

52 Cf. Diário da República, Lisboa, 7 September 1945, 1ª serie, pp. 123-124.

53 Cf. CP Archive, Lisboa, “Concessão Única” in Actas das Sessões do Conselho de Administração, Nº 6, 14 March 1951 to the 3 January 1952.

Luís Lopes dos Santos

BIBLIOGRAPHY

“A Nova Sobretaxa e a sua Aplicação” in Gazeta dos Caminhos-de-Ferro, Lisboa, year 36, Nº 859, 1 October 1923.

ARTOLA, Miguel, “La Acción del Estado” in ARTOLA, Miguel, Los Ferrocarriles en España, 184-1943; 1. El Estado e los Ferrocarriles, Madrid, 1º Volúmen, ed. por Servicio de Estudios del Banco de España, 1978.

“Caminhos-de-Ferro” in Revista de Obras Públicas e Minas, Lisboa, Tomo XXXII, Nº 373375, January to March 1901, ed. Associação dos Engenheiros Civis Portugueses, 1938.

BARRETO, José, “Sindicalismo e Política nos Caminhos-de-Ferro, 1872 -1961” in coord. by GOMES, Gilberto, Estudos Históricos I, Lisboa, ed. by Caminhos-de-Ferro Portugueses, coleccion Para a História dos Caminhos-de-Ferro, vol. 2, 1999.

“Como Vão Ser Remodelados os Estatutos da CP” in Revista Portuguesa de Comunicações, Lisboa, year 3, Nº 58, 15 November 1931.

“Companhia dos Caminhos-de-Ferro do Norte de Portugal” in Revista Portuguesa de Comunicações, Lisboa, year 5, Nº 83, August 1933.

“Companhia dos Caminhos-de-Ferro da Beira Alta” in Gazeta dos Caminhos-de-Ferro, Lisboa, year 49, Nº 1.331, 1 June 1943.

CORREIA, António Vasconcelos, A vida da CP desde o Convénio de 1894, Dificuldades e Soluções; Conferência Realizada na Sociedade de Geografia de Lisboa, em 9 de Junho de 1938 pelo Eng. António Vasconcelos Correia, Presidente do Conselho de Administração da Companhia dos Caminhos-de-Ferro Portugueses, Lisboa, ed.

Companhia dos Caminhos-de-Ferro Portugueses, 1938.

CP Archive, Lisboa, “Aquisição das Linhas dos Meridionais” in Actas das Sessões do Conselho de Administração; Nº 385, Sessão de 29 de Dezembro de 1926, Nº 31, 30 March 1926 to the 19 December 1927.

CP Archive, Lisboa, “Arrendamento e Electrificação do Ramal de Cascais” in Actas do Conselho de Administração da Companhia dos Caminhos-de-Ferro Portugueses, Sessão de 9 de Maio de 1914, Nº 26, 18 April 1913 to the 28 June 1915.

CP Archive, Lisboa, “Companhias de Via Reduzida –Vale do Vouga e Nacional” in Actas das Sessões do Conselho de Administração; Nº 521, Sessão de 1 de Janeiro de 1935, book Nº 36, 2 November 1934, to the 27 Jannuary 1937.

• 63

Railroads in historical context: construction, costs and consequences

CP Archive, Lisboa, “Concessão Única” in Actas das Sessões do Conselho de Adinistração, Nº 6, 14 March 1951 to the 3 January 1952.

CP Archive, Lisboa, “Sobretaxas”.

CUNHA LEAL, “A Política Ferroviária Portuguesa Depois da Guerra” in Revista Portuguesa de Comunicações, Lisboa, year 1, Nº 1, 1 July 1929.

DE LA TORRE, Enrique, Anuario de Ferrocarriles, Madrid, Autoedición, 1935.

Decree of 30 May 1931.

Decree of 29 July 1932.

Decree of the 13 December 1932.

Diário da República, LIsboa, 7 September 1945, 1ª serie.

El Problema Ferroviario. Estudios Realizados y Medidas Legislativas Adoptadas en España Para Intentar su Resolución, 1918 -1932, Madrid, ed. by Compañías del Norte de España and Madrid a Zaragoza y a Alicante, September 1932 FERNANDO DE SOUSA, J., “A Propósito do Resgate” in Gazeta dos Caminhos-de-Ferro, Lisboa, year 33, 16 April 1920.

FERNANDO DE SOUSA, J., “Alteração das Sobretaxas” in Gazeta dos Caminhos-deFerro, Lisboa, year 37, Nº 872, 1 April 1924.

64 •

FERNANDO DE SOUSA, J., “Companhia dos Caminhos-de-Ferro do Norte de Portugal” in Gazeta dos Caminhos-de-Ferro, Lisboa, year 49, Nº 2.192, 16 August 1937.

FERNANDO DE SOUSA, J., “O Arrendamento das Linhas do Estado” in Gazeta dos Caminhos-de-Ferro, Lisboa, year, 40, Nº 948, 16 June 1927.

FERNANDO DE SOUSA, J., “O Resgate” in Gazeta dos Caminhos-de-Ferro, Lisboa, year 33, Nº 796, 1 January 1920.

FERNANDO DE SOUSA, J., “Providência que se impõe” in Gazeta dos Caminhos-deFerro, Lisboa, year 32, Nº 728, 16 April 1918.

IZQUIERDO, Rafael, Cambó y su Visión de la Política Ferroviaria, (El Inicio de un Cambio), Madrid, ed. Fundación de los Ferrocarriles Españoles; Colegio de Caminos, Canales y Puertos, Colección de Ciencias, Humanidades e Ingeniería, Nº 64, 2000.

Law of 10 May 1932.

Law of 7 July 1932.

Law of 9 September 1932.

Law of the 29 May 1934.

MAR MERINO, Maria de, “Tiempos de Acero” in MOPT, Revista del Ministério de Obras Públicas e Transportes; Historia del Ferrocarril en España, 1842 -1992, Nº 400, JulyAugust 1992.

Luís Lopes dos Santos

MUÑOZ RUBIO, Miguel, “El Estado como Empresario Ferroviario” in MUÑOZ RUBIO, Miguel; SANZ FERNÁNDEZ, Jesús; VIDAL OLIVARES, Javier, Siglo y Medio de Ferrocarril en España, 1848-1998; Economía, Industria y Sociedad, s.l., ed. Fundación de los Ferrocarriles Españoles; Instituto de Cultura Juan Gil Albert; Diputación Provincial de Alicante; Caja de Ahorros del Mediterráneo, 1999 “O Estado e o Problema Ferroviario” in Revista Portuguesa de Comunicações, Lisboa, year, 1, Nº 1, 1 June 1929.

Order of the 13 December 1932.

ORTUÑEZ GOICOLEA, Pedro Pablo, El Proceso de nacionalización de los Ferrocarriles en España. Historia de las Grandes Compañías Ferroviarias, 1913- 1943, Valladolid, phD Thesis, Valladolid University, Facultad de Económicas y Empresariales; Departamento de Historia e Instituciones Ecinómicas y Economia Aplicada, 2003.

PINHEIRO, Magda, “Les Chemins de Fer Portuguais: Entre l’Explotation Privée et le Rachat” in Revue d’Histoire des Chemins de Fer ; Actes do 5éme Colloque de l’Association pour l’Histoire del Chemins de Fer en France en Collaboration avec l’ Union Internationale des Chemins de Fer, Nº 16-17, Spring –Autumn 1997, ed.

Association pour l’Histoire del Chemins de Fer ; Unin Internationale des Chemins de Fer. “Proposta de Lei Apresentada à Câmara dos Senhores Deputados em Sessão de 24 de Abril de 1903”, Lisboa, Ministério de Obras Públicas, Comércio e Indústria; Caminhos-deFerro do Estado, 1903.

Royal Law –Decree from the 12th July 1924.

SANTOS, Luís António Lopes dos, Política Ferroviaria en España; Desde Primo de Rivera a la Guerra Civil, Madrid, Study of research capability, as a part of the Madrid Complutense University phD program, 2002.

• 65

Hugo Silveira Pereira

RAILWAYS IN TRÁS-OS-MONTES DURING THE SECOND HALF OF THE 19TH CENTURY: PROJECTS AND ACHIEVEMENTS CAMINHOS DE FERRO EM TRÁS OS MONTES NA SEGUNDA METADE DO SÉCULO XIX: PROJETOS E RESULTADOS Hugo Silveira Pereira (U. Porto, Portugal) Hugo José Silveira da Silva Pereira was born in Oporto in 1979. In 2005, he completed his History undergraduate program in the Arts and Humanities College of Oporto University. Three years later he completed his master program in Contemporary History in the same institution with an investigation about the relationships between the lower house of the Portuguese parliament and the construction of railways between 1845 and 1860. In 2008, he began his PhD program still in Oporto University.

Hugo Silveira Pereira nasceu no Porto em 1979. Em 2005, licenciou-se em História na Faculdade de Letras da Universidade do Porto. Aqui obteve, três anos depois, o grau de mestre em História Contemporânea com uma tese sobre a relação entre o parlamento português e a construção ferroviária no período compreendido entre 1845 e 1860. Em 2008 iniciou o seu curso de doutoramento ainda na mesma universidade.

Abstract Resumo

In the second half of the nineteenth century, after 30 years of political turmoil, the Portuguese monarchy was finally solid enough to begin a new economic strategy based on the development of transport infrastructures, essentially railways. The deployment of the very first rails occurred in 1852, however the first law that decreed the building of a rail line to Trás-os-Montes (the one between Oporto and Pinhão, through the Douro valley) was only published in 1867, even though the projects and suggestions presented since the beginning of the second half of the nineteenth century in the Portuguese parliament to build railways in the far northeastern part of Portugal were numerous. Despite all this, the first trains would only travel through the southern border of that region in the late 1870s. This paper aims to shed some light on this issue and to try to explain the reasons for that delay. We will then focus our analysis on the first railroad that served the heart of Trás-os-Montes: the Tua railway. When the Douro line entered Trás-os-Montes, the efforts to grant other railroads through that province grew stronger. Several solutions were presented by the Portuguese engineers, but the Tua railway (from the Douro line to Mirandela)

• 67

Railroads in historical context: construction, costs and consequences

ended up being the chosen one. To do all this, we will mostly use the debates that took place in both houses of the Portuguese Parliament and the reports of the Portuguese engineering since the beginning of the second half of the 19th century until the 1890s crisis which put the investment on public works on hold.

68 •

Na segunda metade do século XIX, e depois de trinta anos de agitada discussão, a monarquia portuguesa teve finalmente forças para iniciar uma nova estratégia económica baseada no desenvolvimento de infraestruturas de transportes, essencialmente baseada nos caminhos de ferro. Os primeiros troços começaram em 1852, mas a primeira lei que define a construção da primeira linha em Trás os Montes (entre Porto e Pinhão, ao longo do vale do Douro) só foi publicada em 1867, embora tenham sido numerosos os projetos e sugestões apresentadas ao Parlamento, desde o inicio da segunda metade do século XIX, para construir linhas de caminhos de ferro na região nordeste de Portugal. Apesar de tudo isso, os primeiros comboios só começaram a circular através da fronteira meridional dessa região em finais dos anos 1870.

Este trabalho pretende lançar alguma luz sobre o assunto e tentar explicar as razões desse atraso. Focaremos a atenção na primeira linha a servir o centro de Trás os Montes: a linha do Tua. Quando a linha do Douro entrou em Trás os Montes, aumentaram os esforços para se criarem outras linhas na região. Os engenheiros portugueses propuseram várias soluções, mas a linha do Tua (entre a linha do Douro e a cidade de Mirandela) acabou por ser a escolhida. A nossa análise baseia-se largamente nos debates parlamentares que tiveram lugar em ambas as câmaras do Parlamento português e nos relatórios da engenharia portuguesa desde o principio do século XIX até à crise de 1890, que suspendeu muitos dos investimentos em obras públicas.

Hugo Silveira Pereira

Railways in Trás-os-Montes during the second half of the nineteenth century: projects and achievements Hugo Silveira Pereira

INTRODUCTION In the first half of the nineteenth century, turmoil in the Portuguese political affairs prevented any kind of investment in transport infrastructures1. Only Costa Cabral, in the 1840s, was able to introduce some stability to the political system in order to sign the first contract to build a railway in Portugal. This contract was not fulfilled; however the Portuguese rulers understood that cutting the expenses wasn’t enough and that investing in public works and transport infrastructures was a pressing need2.

The coup of 01-05-1851 marks the beginning of a historical period known in Portugal as Regeneração (Regeneration) that ended in 1892 with the State default. These four decades were characterized by an enlarged political consensus around the concept of progress that would be brought by investing in railways, roads and harbours, or so the Portuguese politicians hoped3.

The great objective of this strategy of investment was to draw Portugal closer to Europe, both in terms of distance and economic development. In the mid 1850s, world commerce had reached an all-time high. Railroads had met the need for better and larger transportation but they also played an important role in the growth of commercial transactions4. In countries like England, France, Germany, Belgium or the United States of America, commerce grew in tandem 1 MARQUES, 2002: 552-621. MATA & VALÉRIO, 1993: 142.

2 SANTOS, 1884, n.er 174: 1. SOUSA & MARQUES, 2004: 467.

3 BONIFÁCIO, 1992: 96-98. PINHEIRO, 1983: 53.

4 BAIROCH, 1976: 33-36.

• 69

Railroads in historical context: construction, costs and consequences

with their railway networks5. For a nation like Portugal it was very tempting to connect economic development with railway mileage, even though those other nations had other characteristics and structures that justified their level of economic growth6.

At an internal level, railways would also serve as means to modernize the national transport system and unify the Portuguese market. By mid-19th century, the kingdom’s transport grid was archaic to say the least. There were only 150 miles of roads, most of the rivers had limited navigability and 30 to 40% of the Portuguese territory had absolutely no access to those natural waterways7.

In this particular matter, Trás-os-Montes was, at the beginning of the Regeneração, one of the most underprivileged provinces of Portugal8 and this is precisely the focus of this paper.

Map 1 – Trás-os-Montes

70 •

5 MITCHELL, 1976: 789-790.

6 PEREIRA, 2010: 5.

7 VALÉRIO, 2001: 361. GUILLEMOIS, 1995. JUSTINO, 1988-1989: 189-190.

8 ALEGRIA, 1990: 161 and 335.

Hugo Silveira Pereira

Trás-os-Montes had some economic interests to exploit and deserved the attention of the government as much as any other region. Nevertheless, only in the late 1870s, roughly twenty years after the opening of the first rail service, would railways reach this part of the kingdom. After that, and until the default of 1892, only one more railroad would be open for business. In the following lines we will try to explain why this happened. Why was Trás-os-Montes apparently forgotten whereas in other regions the investment was far greater (by 1892 Portugal had built around 1500 miles of tracks9). To do so we will use the debates that took place in both houses of the Portuguese parliament (published in those chambers journals) and the reports of the national engineering spawned from the offices of the ministry of Public Works (preserved in its Historical Archive or published in its official paper) and from the Portuguese Engineers Association (edited in its own magazine).

AN EARLY OBJECTIVE The idea to build a railway straight to the heart of Trás-os-Montes goes back to the 1840s. M. Huguin, Augusto Garvelle and Charles Jucqeau thought about a rail line between Lisbon and Porto with a branch to Braganza and the Portuguese frontier with Spain. Afterwards, Samuel Clegg, William Law, Arriaga e Cunha, Sousa Botelho and Pinto Soveral preferred a shorter route that would connect Porto to Régua. A third party (self denominated Alto Douro Railway) suggested a railway between Porto and Salamanca or Torre de Moncorvo10. In the parliament, a railway through the Douro valley was also regarded as the only one that would justify the investment, at least in the mind of the viscount of Vilarinho de São Romão, a rich landlord in that region11. All of these solutions aimed to connect Trás-os-Montes to Porto, the main harbour of the north of Portugal and ending point of commercial routes with centuries of tradition12.

In the following decade, the project of driving trains to Trás-os-Montes was still in the agenda of the engineers and politicians of the Regeneração. In 1851, the engineer Albino de Figueiredo, while sketching a nationwide network, recommended the building of a railway along the Douro valley to Régua and from here to Vila Real13. A year later, when Fontes Pereira de Melo, minister of Finance and Public Works who was also an engineer, ordered the survey of a 9 VALÉRIO, 2001: 373.

10 VIEIRA, 1983: 84-88.

11 Diário do Governo (DG), session of the house of Peers, 28-03-1843: 134; 06-06-1843: 777; 0805-1843: 785. For the biographical background of this and other congressmen, check MÓNICA, 2005-2006.

12 ALEGRIA, 1990: 63, 71-72 and 102-103. SOUSA & PEREIRA, 1988: 37-38.

13 ALMEIDA, 1851: 21-23.

• 71

Railroads in historical context: construction, costs and consequences

railway between Porto and Lisbon (09-11-1852 and 28-10-1852 ordinances), he states that the arrival to Porto should be such that could allow the continuation of the track northwards across Minho and eastwards to Trás-os-Montes14.

All of this enthusiasm however, was cooled down by a report of a French engineer that arrived in Portugal in the mid 1850s (hired by the minister Fontes) to study railway construction in Portugal. According to Mr. Watier, any line deriving from Porto apart from that one through the shoreline to the northern frontier with Spain was impossible to build15. Due, perhaps, to this report, the Douro railway would be forgotten in the following years, whilst the Minho line gathered the attention of some entrepreneurs, namely the Spanish count of Reus and the Portuguese baron of Lagos, whose offers were acknowledged but not approved16.

THE MATURATION OF THE PROJECT

72 •

Since the businessmen seemed uninterested in the Douro line, some congressmen of that region tried to use their position in the parliament to draw their attention. In the low house, Júlio do Carvalhal Sousa Teles and Afonso Botelho proposed the building of a railway between Porto and Régua or Tua thrice17.

Those men deemed such an endeavour as the salvation of the Douro valley and the province of Trás-os-Montes. At the time, the vineyards of these regions were dealing with the oidium18 and a railroad was considered a solution to the problems of the area. However the lack of surveys admonished an immediate construction and so the government preferred to order the assessment of that work to the Portuguese engineer Sousa Brandão (17-07-1862 ordinance)19.

Even though the work of Sousa Brandão was commended by the ministry of Public Works advisory body (the General Council of Public Works)20, the entrepreneurs still felt more attracted by a hypothetical investment in Minho.

Four proposals were presented to the government (Grouselle & Companhia, José de Salamanca – contractor of the northern and eastern railways –, Mare de la 14 Ministry of Public Works Historical Archive (MPWHA). Conselho de Obras Públicas e Minas (COPM). Book 1 (1852-1853): 1-6. Colecção Oficial de Legislação Portuguesa (COLP), 1852: 628630. FINO; 1883: 19-20. DINIS, 1915-1919, vol. 1: 41-43.

15 WATIER, 1860.

16 COLP, 1857: 408. Boletim do Ministério das Obras Públicas (BMOP), 1863, n.er 11: 429-432. DG, 1859, n.er 61: 332-333.

17 Diário da Câmara dos Deputados (DCD), 1860.5.22: 234-235. DL, sessions of the house of commons 1862.1.17: 182; 1862.6.12: 1638 and 1862.6.26: 1775-1776.

18 SOUSA & MARQUES, 2004: 78 and 81.

19 COLP, 1862: 217.

20 MPWHA. COPM. Book 19 (1864): 53-54.

Hugo Silveira Pereira

Caine and Piombino & Companhia)21 and in the parliament, the congressmen approved a bill (05-10-1864) that granted in a public auction financial support to the company that was willing to build a railroad between Porto and the Spanish province of Galicia. Unfortunately, the house of Peers was never heard as far as this issue was concerned and so the bill never became a law22.

Fearing that the Douro line could be forgotten and surpassed by the Minho railway, some Trás-os-Montes congressmen drew once again the attention of the government to that line. Some of them (led by Júlio do Carvalhal Sousa Teles) went as far as asking the cabinet for a road to Abreiro and a tram to Mirandela as a complement of the Douro railway. Afonso de Botelho, a rich landlord in Porto native of Trás-os-Montes, proposes that the railway goes as far as Torre de Moncorvo where a road from Barca de Alva (near the border) would branch23.

At the time, the government of the so-called Historical Party tried to please everybody and ordered that both lines be studied (Porto – Braga and Porto – Régua – Barca de Alva – Salamanca), even though there were those amongst the military who feared the connection to Salamanca since it would become a second route of invasion if the Beira Alta railway was also deployed24. In any case, no rail was to be settled for the time being. They were only talking about studies and once again Sousa Brandão was man for the job (30-08-1864 and 21-10-1864 ordinances)25.

However, the fear of some army men was shared by a few engineers (most of whom were also army officers) in the ministry of Public Works, who, in a draft of the Portuguese desired rail network, proposed that the Douro line cross the border alongside the Beira Alta railway in Almeida26. The marquis of Sá da Bandeira suggested yet another way: the line through the Douro valley should turn north in Regua in the direction of Vila Real and then turn east again to Mirandela and Braganza before crossing the border towards Zamora27.

Anyhow, this subject was far from being a matter of concern in the short run, since the border was still 150 miles away. So, in 1865, the coalition in the government promised the Douro line as a compensation for the law that opened the Douro harbour to the exportation of wine to any part of the country28. By this time, Sousa Brandão had completed his survey of the track between Porto (Fontainhas) and Marco de Canaveses29. Meanwhile, back to the parliament, other congressmen tried to 21 MPWHA. COPM. Book 19 (1864): 361-364 and 366-370.

22 PEREIRA, 2011.

23 DL, sessions of the house of commons, 1864.2.16: 436; 1864.3.12: 779; 1864.4.19: 1207; 1864.4.20: 1223-1224 and 1864.4.25: 1295.

24 PIMENTEL, 1865.

25 BMOP, 1864, n.er 11: 589.

26 MPWHA, CSOP, Book 22 (1865): 18-29v.

27 DL, session of the house of Peers, 1865.12.23: 3004-3005.

28 DL, session of the house of commons, 1865.11.8: 2497-2498.

29 MPWHA, COPM, Book 21 (1865): 58v-61.

• 73

Railroads in historical context: construction, costs and consequences

place Chaves in the railway route, both by the Corgo and the Tâmega valley30.

THE 02-07-1867 LAW

74 •

Despite all the surveys and all the efforts of Trás-os-Montes men in the parliament, the locomotives were still far away from that region. However, that was about to change...or so it seemed. On 02-07-1867, after a lively two weeks debate, the congress approved a bill that authorized the government to build the railways between Porto, Galicia and Pinhão (in the Douro country). Regretfully, the Portuguese treasury was in no conditions to take on such enterprise, which had to be postponed.

Five years later, the realm was once again able to invest in railways. Furthermore, the government was still authorized to build the Minho and Douro lines. So, in July, the construction began in Porto. A year later, the work towards Pinhão also began in Ermesinde. However, the progress of the construction was very slow due to the hilliness of the region. Two years later, the tracks reached Caíde, 25 miles away from Ermesinde. On the other railway, this year (1875) marked the beginning of railway operations between Porto and Braga. Once this was accomplished, several men advised that the line should be extended to the northwest of Trás-os-Montes and Spain31. Meanwhile, the railway finally entered Trás-os-Montes (Rede) in the beginning of 1879. Afterwards, Régua (July 1879) and Pinhão (1880) would also receive the first locomotives and carriages32. By this time, the Porto wine vineyards had already migrated to the High Douro, where rails were yet to be deployed33.

NEW PROJECTS In 1880, the Douro railway only served the southernmost part of Trás-os-Montes.

By this time, there had been several projects to expand the network in this part of the country and the objective to drive the Douro line straight into the heart of Spain was definitely there. However, none of this was accomplished in the short run.

30 DL, sessions of the house of commons, 09-05-1864: 1475 and 10-05-1867: 1467.

31 DCD, 1875.3.10: 772-773 and 831-834.

32 SILVA, 2004, vol. 1: 48.

33 SOUSA & MARQUES, 2004: 80.

Hugo Silveira Pereira

On 01-03-1875, a group of congressmen proposed the construction of a railway between Régua, Vila Real and Chaves. The richness of the soil of that region, the intense wine production, the fame of the Barroso cattle and the Pedras Salgadas and Vidago baths and springs were reasons enough for that party to justify the investment. Besides, all of its members had connections or interests in such an enterprise. Some of them were from Trás-os-Montes; others were elected by Trás-os-Montes voters; and two of them (Falcão da Fonseca and José Pedro António Nogueira) founded the company that explored the baths and springs around Chaves34. The bill was sent to the Public Works and Finances parliamentary committees – standard procedure according to the rules of the Portuguese congress, necessary to the debate on both houses –, but these never stated their opinion on the matter.

Further to the east, there was also an idea to build a railroad through the Sabor valley, ending in the Douro line and putting an end to Trás-os-Montes seclusion. Such a track would benefit the Miranda do Douro plateau and the mining regions of Roboredo, Freixo de Espada à Cinta, Moncorvo, Santo Adrião and Mogadouro. Furthermore, it could attract traffic from Zamora, Spain. This acted as a powerful motive, since in normal circumstances the construction of international railways had precedence over any other line35. However, a few years later, the chosen route would take the Douro line to Barca de Alva. So that project had to be placed on hold.

A similar fate would be met by another project of another company to employ Winterthur locomotives (capable of beating steep slopes and tight curves) in a line along the Corgo valley to Vila Real. The first experiments occurred in July 1877, but the ministry of Public Works didn’t buy the idea36.

Some of these railways (alongside the lines between Mirandela, Torre de Dona Chama and Vilar Seco or Braga and Montalegre through the Cávado valley) were considered during the debate that took place in the Portuguese Engineers Association (PEA) about the general network plan in the second half of the 1870s. In these discussions, the Portuguese engineering forecasted a bright future for the lines of the Tua and Sabor basins, both as part of international routes and as part of a railroad parallel to the Spanish frontier. The debate lasted for about a year and a half, but in the final report of the PEA, only the Tua line (as part of the Douro line in its route to Spain) and the railway between Guimarães and Chaves were proposed.

34 DCD, 28-03-1877: 568-569. MÓNICA, 2005-2006.

35 DCD, 28-03-1877: 823-826.

36 MPWHA. JCOPM, 1876-1877, box 20, report 7637.

• 75

Railroads in historical context: construction, costs and consequences

Map 2 – PEA’s first proposal (1876)*

* ASSOCIAÇÃO…, 1878a

Map 3 – Engineer Valadas’ proposal (1876)**

76 •

** VALADAS, 1878.

Map 4 – Engineer Sousa Brandão’s proposal (1876)***

*** BRANDÃO, 1878.

Hugo Silveira Pereira

Map 5 – PEA’s second proposal (1876)*

* ASSOCIAÇÃO…, 1878b.

Map 6 – Engineer Mendes Guerreiro’s proposal (1877)**

• 77

** GUERREIRO, 1878.

Map 7 – PEA’s final proposal (1877)***

*** ASSOCIAÇÃO…, 1878c. ALEGRIA, 1990.

Railroads in historical context: construction, costs and consequences

Map 8 – Engineer Pimentel’s proposal (1877)*

* PIMENTEL, 1878a. PIMENTEL, 1878b.

78 •

This report was probably delivered to the government, but the minister was still looking for other opinions and he instructed João Crisóstomo, former minister of Public Works, to present yet another survey. In 1878, the report was published in the crown’s journal, proposing the lines highlighted in the following map.

Map 9 – Engineer João Crisóstomo’s proposal (1878)**

** DG, 1878, n.er 210: 2260-2266. ALEGRIA, 1990.

Hugo Silveira Pereira

The network that would be proposed to the parliament in 1879 by the ministers of Public Works (Lourenço de Carvalho) and Finance (António de Serpa) was rather different. Trás-os-Montes would count several lines: following the Douro river to the border; through the Tua valley to Bragança and Vinhais (this line also had powerful supporters in and outside the parliament, as we shall see in due time); connecting Famalicão and Régua to Chaves; a railway between Pocinho and Miranda do Douro; and a track through the Cávado basin. Unfortunately, the parliamentary committees never gave their opinion about this proposal.

Map 10 – Government’s proposal (1879)*

• 79

* DCD, 07-02-1879: 345-353. ALEGRIA, 1990.

In 1879, the government was still being advised to build more railways in Trás-os-Montes. The company that owned a coastland railway (Porto – Póvoa de Varzim with a branch to Famalicão) wanted to expand it to Chaves and Régua and asked the government for a subvention (a guaranteed income)37. Since this project implied an increase in the public expenses, it had to be presented to the parliament, which happened in 24-03-1879. The advisory board of the ministry of Public Works and the parliamentary committees gave their approval to such a venture38. However, their reports were never presented for discussion. A year later, the project was resurrected in the parliament at the same time a similar one (railways towards Trás-os-Montes beginning in Braga) was also proposed39.

Nonetheless, all these efforts would be disdained. A few years, later the company that operated the line between Guimarães and the Minho rail would also suggest its expansion to Chaves through the Tâmega valley, but their lack of success was the same as all the others40.

37 OLIVEIRA, 1979: 7.

38 SANTOS, 1884. MPWHA. Junta Consultiva de Obras Públicas e Minas (JCOPM), 1879, box 22, report 8156.

39 DCD, 1880.4.24: 1560 and 28-05-1880: 2360-2361.

40 Ministry of Foreign Affairs Historical Archive (MFAHA). Caminhos-de-ferro. Ligações por Intermédio de Pontes, box 38, pack 8, document 281.

Railroads in historical context: construction, costs and consequences

Map 11 – Guimarães Railway Company and Porto, Póvoa and Famalicão Company proposals

80 •

A year later, Sousa Brandão added new information to this dossier when he presented new studies for a narrow gauge network north of the Douro. His purpose was to bring trains to Trás-os-Montes and to increase the income of the Douro line, since all of the projected railways would be its branches. And just like the rivers Tua, Sabor, Corgo and Tâmega were tributaries of the Douro river the proposed railroads would also be tributaries of the Douro line and they would follow those rivers’ valleys. So, great cities of Trás-os-Montes (like Mirandela, Braganza and Chaves) would also be connected to the national rail system. Furthermore, other railroads could be established, but their construction wasn’t a pressing need.

Map 12 – Narrow gauge network north of the Douro*

* BRANDÃO, 1880.

Hugo Silveira Pereira

THE SALAMANCADA In any case, any enterprise of this sort had to wait for the completion of the Douro line to the frontier, the most important objective of the Portuguese national railway strategy. This occurred in the mid-1880s in a process that was far from being straightforward.

In 1878, the Portuguese had placed their hopes and dreams on a new track, christened Beira Alta railway, connecting Pampilhosa to Vilar Formoso connecting to the west to Figueira da Foz and to the east to Salamanca. Since the Douro line was only a national railway, the merchants of Porto feared that all traffic from Salamanca would flee to the harbour of Figueira. Worsening the scenario, the company that owned the Beira Alta railway – the Societé Financière de Paris (SFP) – also owned the section between Salamanca and Medina del Campo and furthermore had grasped the concession of both railroads from Salamanca to the Portuguese frontier, but naturally only had real interest in the connection to Vilar Formoso41.

To avoid this scenario, the Portuguese government began talks with Spain.

Due to those negotiations and to new surveys instructed to a team of Iberian engineers (Martinez Gordon, Eusébio Page, Boaventura José Vieira, Avelar Machado and Bandeira Coelho de Melo), it was decided that the Douro line would also be built to the frontier and that it would branch with the Beira Alta railway in Spanish territory, despite the military’s preference that the railway went through Almeida, a military stronghold42. In the parliament, the government proposed the extension of the railway along the Douro shores to Spain instead of taking it through the heart of Trás-os-Montes, on the condition that an agreement with Spain was achieved. This indecision was due to the fact that Spain and Portugal had different understandings about the point of embranchment of both lines. Spain wanted Ciudad Rodrigo, which was against the interest of Porto, since in this fashion Salamanca was closer to Figueira and further from Porto. Portugal, on the other hand, wanted Boadilla. In May 1880, the Portuguese ambassador was able to convince his counterpart and Spain agreed on Boadilla.

However, there was still a loose end in this issue: would the SFP be willing to build both lines at the same time, even if that company only had real interest in one of them?43 To avoid such a scenario some banks at Porto gathered in a joint venture (01-07-1881) and took the responsibility of building and exploring both railroads. The Portuguese government also agreed to aid this joint venture with a financial subvention, despite the demonstrations in and outside the parliament 41 SOUSA, 1978: 5-7.

42 MFAHA. Entroncamento da linha ferrea do Douro, box 1036, pack 5. MACHADO, 1879.

43 MFAHA. Entroncamento da linha ferrea do Douro, box 1036, pack 5; Caminhos de ferro. Ligação por intermédio de pontes, box 1066, pack 8, documents 174 and following.

• 81

Railroads in historical context: construction, costs and consequences

against such an uncanny decision44. On the other hand, the government also had to hasten the construction of the Douro line so that the entire railway from Porto to Salamanca could be opened for business in one single stroke. Nonetheless, the hilliness of the region delayed the work. Only in 1883 did the railway reach Tua, barely 10 miles away from Pinhão. A short distance further, the railway would cross to the other shore of the Douro45. Only in 1887 would the line be completed in Barca de Alva46. As far as the lines in Spain were concerned, they would become an enormous financial burden to the Porto banks that forced the government to renew its aid a few years later47.

THE TUA LINE

82 •

As we saw, in the mid-1880s, Trás-os-Montes only had a few miles of tracks going close to its southern border. It was clearly not enough. In this chapter, we will analyze the process that culminated in the opening of the Tua line.

In the mid-1870s, the businessman Clemente Meneres invested large amounts of money in Trás-os-Montes and started to feel the need for a more effective transport system towards Porto. He put some pressure upon the government in order to build a railway through the heart of Trás-os-Montes. He wrote several texts in the papers and he even went as far as making up demonstrations of people demanding the railways, since he believed that the people of Trás-os-Montes were too phlegmatic. At the same time, he also took advantage of the interest of some powerful men in the parliament and in the government in the purchase of some properties in this region48. In fact, in the parliament, a few congressmen from both houses weren’t shy to show their support to this enterprise. Between 1878 and 1882, Eduardo José Coelho and the bishop of Braganza (brother of Martens Ferrão, a high-ranked member of the Regenerador Party, who was often called by the king to rule the nation) proposed the construction of the Tua railway49.

Meanwhile, there were those who had a preference for other railways across Trás-os-Montes. However, the decision to take the Douro line by the left bank of that river (1882) hastened the need to build other railways that would bring 44 SANTOS, 1884. Documentos relativos ao caminho de ferro de Salamanca publicados no Diário do Governo.

45 MPWHA, JCPOM, box 24 (1882), report 10050.

46 SILVA, 2004: 154. SILVA & RIBEIRO, 2007-2009, vol. 3: 162. MFAHA. Caminhos de ferro.

Ligação por intermédio de pontes, box 1066, pack 8, documents 180 and following.

47 SOUSA, 1978.

48 ALVES, 2007: 124-125.

49 DCD, 18-03-1881: 1065. Diário da Câmara dos Dignos Pares do Reino, 03-04-1878: 277-278, 1881.2.9: 139-140 and 1882.4.19: 427-428.

Hugo Silveira Pereira

the benefits of steam traction to Trás-os-Montes. So it’s no wonder that in 1883, several congressmen asked the government for new surveys for lines connecting Régua, Vila Real, Vila Pouca de Aguiar, Mirandela, Chaves, Braganza and Zamora to the Douro railway50.

Nevertheless, it would be the Tua railroad the chosen one, thanks to the interest taken by the count of Foz, who in a short time would become the leader of the Royal Company of the Portuguese Railways, the most powerful company operating in Portugal that held the northern and eastern railways51. Furthermore, the advisory board of the ministry of Public Works also believed that the Tua line was the best choice at the moment, because it was already surveyed and it would go through the richest regions of the province52.

In the parliament, the line was debated since February 1883 in a bill that also proposed the construction other railways (through Beira Baixa and to Viseu).

The main issue during the discussion was to know in which bank of the Tua river should the rails be deployed, which motivated a feud between the congressmen of both banks. In any case, that was a concern to be settled by engineers (that would advise the left bank, for being less ridged, a year later53) and not by politicians and in 26-04-1883 the law that allowed the government to help the company who wished to build this narrow gauge railway was published.

Unfortunately, no one was interested in taking such an obligation, due to the conditions of the forced repurchase by the government. Clemente Meneres tried to attract some investors, like Henry Burnay or the Pereire brothers, but he was unsuccessful. He even considered bidding himself, but he realized that such a decision was no longer necessary54. The government would alter the conditions of the compulsory repurchase and the count of Foz accepted the work (24-121883 and 30-06-1884 contracts and 26-05-1884 law)55. The contractor took the endeavour very seriously and a few weeks later he presented its working projects that were highly praised by the ministry of Public Works, even though its engineers were very unconvinced about the economic viability of the railway, due to the demographic shortage and the lack of roads that would make the access to the stations very difficult56. As soon as his projects were approved, the count of Foz created the Railways National Company (01-10-1885), which became responsible for the contract.

The work was far from easy, due to the mountainous characteristics of the terrain. This led to a track with many curves and to the postponement of the con50 DCD, 17-02-1883: 372-373.

51 GOMES, 2009: 5.

52 MPWHA, JCOPM, box 25 (1883), report 10305.

53 MPWHA, JCOPM, box 27 (1884), report 11627.

54 ALVES, 2007.

55 COLP, 1883: 359-367 and 406 and following; 1884: 190-210 and 307-313.

56 MPWHA, JCOPM, box 27 (1884), report 11627; box 29 (1885), report 12791.

• 83

Railroads in historical context: construction, costs and consequences

nection to Braganza and to the marble and alabaster mines nearby. In 29-09-1887 the line was conditionally opened. However, there were still many flaws, no agreement about the joint operation of the Tua station (state property) and lack of personnel. In 1888, the line still wasn’t fully operational57. Its net profit was also disappointing. In fact, the expenses exceeded the revenue, a scenario that went on through most of the 1890s58. If it weren’t for the State aid, the company would never survive.

Map 13 – The Tua line*

84 •

* FERREIRA, 1999.

WRAPPING UP THE ISSUE: THE NETWORK NORTH OF THE MONDEGO RIVER The Tua and Douro railways only served a slender region of Trás-os-Montes.

Soon other proposals were presented to the government in the parliament to 57 SILVA, 2004: 118. MPWHA, JCOPM, box 31 (1886), report 13877 (16-12-1886); box 34 (1888), report 15179. COLP, 1888: 405. FINO, 1883-1903, vol. 2: 256-257; vol. 3: 11-14.

58 FINO, 1883-1903, vol. 2: 245-246; vol. 3. PIMENTEL: 1890.

Hugo Silveira Pereira

increase the extension of the region’s network. As soon as 1885, two new solutions were advised, both in 09-06-1885. They proposed that Vila Real become part of the set of cities served by railway through the construction of a track to the Douro line59. Besides the usual reasons that were invoked to justify these kind of investments (the wealth and degree of commercial activity of the region), the men that asked for this work (all of them elected or born in this area) added the phylloxera plague and precisely the uselessness of the Tua and Douro line to the Vila Real district. Theses two projects would merge a year later, but the practical results were the same: none whatsoever60. The same fate would be met by another idea that allowed the Guimarães Railway Company to expand its line to Chaves and then Régua61. In 1888 and 1890, the extension of the Tua line to Braganza (and then Spain) was brought to the parliament by the hands of Eduardo José Coelho and Lopes Navarro (both of them congressmen elected by Trás-os-Montes)62.

These efforts showed that there was still the need to build more railways in Trás-os-Montes. In 1888, the government tried to end this matter, asking permission from the parliament to complete the network north of the Mondego. In Trás-os-Montes, this project included the connection of Chaves to the Douro line through the Tâmega and Corgo valleys and the extension of the Tua line to Braganza. Any other suggestion was disdained. The government hoped to create the conditions to take real advantage of the imagined richness of the land of Trás-os-Montes and to award the patience and tax money that the people from this part of the kingdom always shared. Moreover, building railroads was still regarded as the means to induce economic development, since it was still used abroad63.

However, this proposal was never discussed in the parliament, even though the Public Works and Finance committees gave it their approval64. Outside the parliament, quite a few engineers (Montenegro, Costa Serrão and Rego Lima) also believed that this project was possible, but their opinion wasn’t enough to get the work started65. Two years later, the project was presented once again to the congressmen, but the outcome was the same: no law published, no rails deployed66.

This way, a great part of Trás-os-Montes remained isolated from the rest of the country due to the lack of effective transportation. Even in terms of roads, 59 DCD, 09-06-1885: 2162-2164.

60 DCD, 05-02-1886: 298.

61 DCD, 20-03-1886: 660-661.

62 DCD, 20-03-1886 e 11-07-1890: 1246 e 1157.

63 DCD, 01-06-1888: 1813-1819.

64 SANTOS, 1884.

65 MONTENEGRO, 1889. SERRÃO, 1890. LIMA, 1890.

66 DCD, 10-05-1890: 158-159.

• 85

Railroads in historical context: construction, costs and consequences

Map 14 – Network north of the Mondego

MPWHA, Mapas e Desenhos, C-32-7-B.

this region was poorly served67. In fact, “the greatest terror one could inspire in the heart of a public servant is to threaten him with a transfer to Braganza. One can only imagine that Braganza is Siberia!”68.

86 •

FINAL REMARKS Even though the construction of railways began in Portugal in the 1850s, only in the late 1870s did trains travel in Trás-os-Montes. This delay was due to several reasons, the first of which the distance from the shore and from the two major Portuguese cities (Porto and Lisbon). Furthermore, Portugal had limited resources and the priority of the governments (apart from linking Lisbon to Porto) was to connect Lisbon and its harbour to Spain and the rest of Europe69.

In the 1850s and 1860s, the international connection was very far from Trásos-Montes. The governments were definitely leaning towards a railway that would cross the border somewhere between Almeida and Elvas. However, the potential investors and Spain were only willing to accept the latter70. On the other hand, it was believed that it would be easier and cheaper to build on the southern and plainer provinces of Portugal.

The Alentejo, for instance, was granted rail tracks in the 1850s. It was however a region closer to Lisbon and there were investors (rich landlords of the 67 ALEGRIA, 1990: 161 and 335. PORTUGAL, 1907.

68 DCD, 23-02-1880: 542.

69 PEREIRA, 2008: 154. RAMOS, 1996.

70 ALEGRIA, 1990: 243-251.

Hugo Silveira Pereira

region and with sufficient influence over the government and the parliament) willing to take on such enterprise. In the north, only the wine producing regions of the Douro seemed to justify such an investment. However, the expected difficulties of building there, the unlikeliness of a railway in the Douro to reach the frontier, the minimal satisfaction with the Douro river transport capabilities and the vineyard plagues kept the investors away. In fact, these men seemed more eager to invest in Minho71.

In any case, only when the northern railway connected Lisbon to Porto would any new enterprise be undertaken. However, the construction of the bridge over the Douro river was a very troublesome matter that delayed the arrival of trains to Porto. Only when there was some certainty that the company in charge of the railways would in fact build the bridge (contract signed in 1866) did the government to build a railway towards Trás-os-Montes (the 1867 law). Despite this, there wasn’t enough money to take on such a project and Portugal had more pressing matters to attend, namely the compensations demanded by the foreign companies that operated in Portugal72. The construction had to wait until these issues were settled, which happened in the beginning of the 1870s.

For economic and technical reasons, the Douro railway followed the river valley, merely serving a thin strip of land in the south of Trás-os-Montes. The extension of the line towards the frontier was dependent of Spain. Since it was decreed that such an extension would go to Barca de Alva, any railway towards the centre of Trás-os-Montes would only be of mere local interest and destined to serve as tributary of the Douro line. Under this understanding, several lines were suggested, but without a company willing to build them, their coming into reality was unlikely. The lack of money wasn’t a reason, since until 1882 several lines were adjudicated and built in Minho, Beira Alta, Estremadura and Alentejo. It was still believed that in Trás-os-Montes, only the wine production of the Douro banks was large enough to justify a railway73. Furthermore, the steepness of the region guaranteed a hard and expensive mission. It is a fact that the government could have built the entire railway. However, the Treasury was already overloaded with the construction of railroads in the Minho, Douro, Southeast and Algarve lines.

The investments made by Clemente Meneres in the region boosted the need to build more railways in Trás-os-Montes. The surveys showed that any line had its advantages and disadvantages, but in the parliament, the congressmen who supported the Tua line were the ones who held the greatest political power.

In addition, it was the railway that crossed Trás-os-Montes in its more central region, which could allow other transversal solutions suggested, for instance,

by Sousa Brandão and João Crisóstomo. The concern of the count of Foz with 71 PEREIRA, 2011.

72 SAMODÃES, 1873: 37-53.

73 ALEGRIA. 1990: 63.

• 87

Railroads in historical context: construction, costs and consequences

this line was merely circumstantial. In any case, the government’s subvention ensured that his company would have an annual net profit of 5,5% of the cost of construction.

None of these reasons alone can account for the choice of the Tua line. All of them put together justify that in the 1880s Trás-os-Montes would count with another few miles of rails within its borders. The 1892 bankruptcy placed every major project on hold. Until the end of the 19th century, no more rails would be set on this province of Portugal. However in 1900 (February 15), a building plan was approved74. The Tâmega, Corgo and Sabor railways as well as the extension of the Tua line to Braganza were included in this project. In 1906, trains stormed into that city. And as for other lines, they would be built and open to the general public in the following decades75.

88 •

74 COLP, 1900: 53-54.

75 COLP, 1900: 53-54. TORRES, 1936: 24-27 and 52-53.

Hugo Silveira Pereira

SOURCES OF INFORMATION ALMEIDA, Albino Francisco de Figueiredo e (1851) – Vias de communicação. Lisbon: Tipografia da Revista Popular.

ASSOCIAÇÃO dos Engenheiros Civis Portugueses – Revista de Obras Públicas e Minas. Lisbon: IN (several years).

ASSOCIAÇÃO DOS ENGENHEIROS CIVIS PORTUGUESES. Comissão encarregada de estudar a rede dos caminhos de ferro em Portugal (1878a) – Parecer da commissão encarregada de estudar a rede dos caminhos de ferro em Portugal. «Revista de Obras Publicas e Minas», t. 9, n.er 97. Lisbon: IN, p. 1-8.

ASSOCIAÇÃO DOS ENGENHEIROS CIVIS PORTUGUESES. Comissão encarregada de estudar a rede dos caminhos de ferro em Portugal (1878b) – Caminhos de ferro em Portugal. «Revista de Obras Publicas e Minas», t. 9, n.ers 100101. Lisbon: IN, p. 181-193.

ASSOCIAÇÃO DOS ENGENHEIROS CIVIS PORTUGUESES. Comissão encarregada de estudar a rede dos caminhos de ferro em Portugal (1878c) – Relatorio ácerca do plano da rede geral dos caminhos de ferro em Portugal. «Revista de Obras Publicas e Minas», t. 9, n.ers 102-103. Lisbon: IN, p. 289-304.

BRANDÃO, Francisco Maria de Sousa (1878) – A rede dos caminhos de ferro em Portugal. «Revista de Obras Publicas e Minas», t. 9, n.er 99. Lisbon: IN, p. 148-171.

BRANDÃO, Francisco Maria de Sousa (1880) – Estudos de caminhos de ferro de via reduzida ao Norte do Douro. «Revista de Obras Publicas e Minas», t. 11, n.ers 125-126.

Lisbon: IN, p. 145-183.

GUERREIRO, João Veríssimo Mendes (1878) – Esboço da rede geral dos caminhos de ferro em Portugal. «Revista de Obras Publicas e Minas», t. 9, n.ers 100-101. Lisbon: IN, p. 194-244.

LIMA, J. M. do Rego (1890) – Algumas palavras sobre as condições de adaptação da industria siderurgica em Portugal. «Revista de Obras Publicas e Minas», t. 21, n.ers 245-246. Lisbon: IN, p. 188-212.

MACHADO, Joaquim Emídio Xavier (1879) – A praça de Almeida e sua influencia sobre o caminho de ferro da Beira Alta. «Revista Militar», t. 31, n.er 11. Lisbon: Tipografia Universal, p. 333-341.

MONTENEGRO, Augusto Pinto de Miranda (1889) – A rede complementar dos caminhos de ferro ao norte do Mondego. «Revista de Obras Publicas e Minas», t. 20, n.ers 237-238. Lisbon: IN, p. 315-341.

• 89

Railroads in historical context: construction, costs and consequences

PIMENTEL, Frederico Augusto (1878a) – Caminhos de ferro em Portugal. Breves reflexões ácerca do relatorio sobre o plano da rede geral dos caminhos de ferro em Portugal. «Revista de Obras Publicas e Minas», t. 9, n.er 104. Lisbon: IN, p. 317-326.

PIMENTEL, Frederico Augusto (1878b) – Prolongamento do caminho de ferro do Douro. «Revista de Obras Publicas e Minas», t. 9, n.er 104. Lisbon: IN, p. 335-345.

PIMENTEL, Frederico Augusto (1890) – Caminhos de ferro. Apontamentos para calcular o rendimento inicial provável de uma linha férrea. Lisbon: IN. PIMENTEL, Luís Augusto (1865) – Considerações estrategicas sobre o caminho de ferro da Beira. «Revista Militar», t. 17, n.er 23. Lisbon: Tipografia Universal, p. 543-549.

PORTUGAL. Arquivo Histórico do Ministério das Obras Públicas, Conselho de Obras Públicas e Minas, (several books).

PORTUGAL. Arquivo Histórico do Ministério das Obras Públicas, Junta Consultiva de Obras Públicas e Minas, (several books and boxes).

PORTUGAL. Arquivo Histórico-Diplomático, several boxes.

PORTUGAL. Câmara dos deputados – Diário da Câmara dos Deputados. Lisbon: IN (several years).

PORTUGAL. Câmara dos Pares – Diário da Câmara dos Dignos Pares do Reino. Lisbon: IN (several years).

90 •

PORTUGAL. Governo – Colecção Oficial de Legislação Portuguesa. Lisbon: IN (several years).

PORTUGAL. Governo – Diário de Lisbon. Lisbon: IN (several years) PORTUGAL. Governo – Diário do Governo. Lisbon: IN (several years).

PORTUGAL. Ministério das Obras Públicas, Comércio e Indústria – Boletim do Ministerio das Obras Publicas, Commercio e Industria. Lisbon: IN (several years).

PORTUGAL. Ministério das Obras Públicas, Comércio e Indústria (1907) – Estado da viação ordinária em Portugal. Lisbon: IN. SANTOS, Clemente José dos, compil. (1884) – Caminhos de ferro. Pareceres parlamentares de 1845 a 1884. Biblioteca da Assembleia da República, 226/1910. 32333234.

SERRÃO, Manuel Francisco da Costa (1890) – O caminho de ferro do Pocinho a Miranda do Douro e a exploração do grande jazigo de ferro do Roboredo. «Revista de Obras Publicas e Minas», t. 21, n.ers 243-244. Lisbon: IN, p. 117-148.

VALADAS, Manuel Raimundo (1878) – Memoria sobre a rede geral de caminhos de ferro em Portugal. «Revista de Obras Publicas e Minas», t. 9, n.er 97. Lisbon: IN, p. 9-35.

Hugo Silveira Pereira

BIBLIOGRAPHY ALEGRIA, Maria Fernanda (1990) – A organização dos transportes em Portugal (18501910): as vias e o tráfego. Lisbon: Centro de Estudos Geográficos.

ALVES, Jorge Fernandes (2007) – De pedras fez terra: um caso de empreendedorismo e investimento no Nordeste Transmontano (Clemente Meneres). «Revista da Faculdade de Letras – História», série 3, vol. 8, p. 113-155. Available in ler.letras.up.pt/uploads/ ficheiros/3352 .pdf (14-2-2011).

BAIROCH, Paul (1976) – Commerce extérieur et développement économique de l’Europe au XIXe siècle. Paris: École des Hautes Études en Sciences Sociales.

BONIFÁCIO, Maria de Fátima (1992) – A guerra de todos contra todos (ensaio sobre a instabilidade política antes da Regeneração). «Análise Social», vol. 27, n.er 115. Lisbon: Gabinete de Investigações Sociais, p. 91-134.

BRANCO, Rui Miguel C. (2003) – O Mapa de Portugal. Estado, Território e Poder no Portugal de Oitocentos. Lisbon: Horizonte.

DIAS, Maria Helena (1999) – Os primórdios da Moderna Cartografia Militar em Portugal: Uma História ainda por contar. «Revista da Faculdade de Letras», 5.ª série, n.er 24.

Lisbon: FLUL, p. 49-80.

FERREIRA, Tiago A. M. (1999) – O caminho de ferro na região do Douro e o Turismo. [S. l.]: CP. GOMES, Gilberto (2009) – A Linha da Beira Baixa. «XXIX Encontro da APHES» Available in web.letras.up.pt/aphes29/data/9th/Gilberto Gomes_Texto.pdf (16-4-2010).

GUILLEMOIS, Isabelle (1995) – Les transports au Portugal au XIX siècle (de 1843 à 1899) à travers les Viagens na Minha Terra de Almeida Garrett et As Farpas de Ramalho Ortigão. Bordéus: Université Michel de Montaigne.

JUSTINO, David (1988-1989) – A formação do espaço económico nacional. Portugal 1810-1913. Lisbon: Vega.

MARQUES, A. H. de Oliveira, coord (2002) – Portugal e a instauração do liberalismo. In SERRÃO, Joel; MARQUES, A. H. de Oliveira, dir. – Nova História de Portugal, vol. 9.

Lisbon: Presença.

MATA, Maria Eugénia; VALÉRIO, Nuno (1993) – História económica de Portugal. Uma perspectiva global. Lisbon: Presença.

• 91

Railroads in historical context: construction, costs and consequences

MITCHELL, B. R. (1976) – Statistical Appendix (1700-1914). Railways. In CIPOLLA, Carlo, ed. – The Fontana Economic History of Europe, 4.ª ed., vol. 4. Glasgow: Fontana/ Collins, p. 789-790.

MÓNICA, Maria Filomena, dir. (2005-2006) – Dicionário Biográfico Parlamentar (18341910). Lisbon: ICS. OLIVEIRA, Manuel Alves de (1979) – Uma gorada ligação ferroviária da Póvoa de Varzim a Trás-os-Montes. «Boletim Cultural Póvoa de Varzim», vol. 18, n.er 1. Póvoa de Varzim: [s. n.].

PEREIRA, Hugo Silveira (2008) – Caminhos-de-ferro nos debates parlamentares (18451860). Porto: FLUP. PEREIRA, Hugo Silveira (2010) – Caminhos-de-ferro entre Técnica, Estratégia, Economia e Política (1845-1892). «XXX Encontro da APHES». Available in www.iseg.utl.pt/ aphes30/ docs/progdocs/Hugo Silveira Pereira.pdf (1-12-2010).

PEREIRA, Hugo Silveira (2011) – A construção da rede ferroviária do Minho (1845-1892).

«Cultura, Espaço e Memória» (forthcoming).

PINHEIRO, Magda (1983) – Reflexões sobre a história das finanças públicas portuguesas no séc. XIX. «Ler História», n.er 1. Lisbon: A Regra do Jogo, p. 47-67.

92 •

RAMOS, Paulo (1996) – Os caminhos de ferro e o cais da Europa. In GOMES, Gilberto; SERRÃO, Joel, coord. – O Caminho de ferro em Portugal de 1856 a 1996. O Caminho de ferro revisitado. [S.l.]: CP, p. 24-33.

SAMODÃES, Conde de (1873) – Exame retrospectivo dos actos financeiros do ministerio de 22 de Julho de 1868. Porto: Tipografia da Palavra.

SILVA, José Ribeiro da (2004) – Os comboios em Portugal: do vapor à electricidade.

Queluz: Mensagem.

SILVA, José Ribeiro da; RIBEIRO Manuel (2004) – Os comboios em Portugal: do vapor à electricidade. Queluz: Mensagem.

SILVA, José Ribeiro da; RIBEIRO, Manuel (2007-2009) – Os comboios em Portugal.

Lisbon: Terramar. 3 vols.

SOUSA, Fernando de (1978) – A Salamancada e a Crise Bancária do Porto. «Nummus», 2ª série, vol. 1. Porto: [s.n.], p. 131-160.

SOUSA, Fernando de (1995) – História da estatística em Portugal. Lisbon: INE. SOUSA, Fernando de; MARQUES, A. H. de Oliveira, coord. (2004) – Portugal e a Regeneração (1851-1900). In SERRÃO, Joel; MARQUES, A. H. de Oliveira, dir. – Nova História de Portugal, vol. 10. Lisbon: Presença.

SOUSA, Fernando de; PEREIRA, Gaspar Martins (1988) – Alto Douro. Douro Superior.

Lisbon: Presença.

TORRES, Carlos Manitto (1936) – Caminhos de ferro. Lisbon: [s.n.].

Hugo Silveira Pereira

VIEIRA, António Lopes (1983) – The role of Britain and France in the finance of portuguese railways 1850-1890. A comparative study in speculation, corruption and inefficiency. Leicester: Leicester University.

• 93

Luís Espinha da Silveira • Nuno Miguel Lima • Ana Alcântara

THE IMPACT OF RAILROAD ACCESSIBILITY ON THE POPULATION OF PORTUGAL’S INLAND NORTH REGION (1878-1930).

THE TUA AND THE BEIRA BAIXA LINES. O IMPACTO DA ACESSIBILIDADE AO CAMINHO DE FERRO NA POPULAÇÃO DA REGIÃO NORTE INTERIOR DE PORTUGAL (1878-1930). AS LINHAS DO TUA E DA BEIRA BAIXA Luís Espinha da Silveira, Nuno Miguel Lima, Ana Alcântara (Nova U., Lisbon, Portugal) Luís Espinha da Silveira is Associate Professor, Faculdade de Ciências Sociais e Humanas, Universidade Nova de Lisboa and a researcher at the Instituto de História Contemporânea, at the same University, where he coordinates the Territory and Society research group. He was the principal researcher of the Portuguese team involved in the international project “The Development of European Waterways, Road and Rail Infrastructures: A Geographical Information System for the History of European Integration (1825–2005)” of the European Science Foundation. Funded by the Fundação para a Ciência e a Tecnologia.

Nuno Miguel Lima is a PhD student, Faculdade de Ciências Sociais e Humanas - Universidade Nova de Lisboa, working on Iberian transnational railways, and a researcher at Instituto de História Contemporânea, at the same University. He benefits from a PhD scholarship awarded by the Fundação para a Ciência e a Tecnologia.

Ana Alcântara is a researcher at Instituto de História Contemporânea, Faculdade de Ciências Sociais e Humanas - Universidade Nova de Lisboa. She did a MSc dissertation in Geographic Information Systems and Science] about railways and population in Cova da Beira (1878-1930).

Luis Espinha da Silveira é professor associado, Faculdade de Ciencias Sociais e Humanas da Universidade Nova de Lisboa e investigador do Instituto de História Contemporânea da mesma Universidade, onde coordena o grupo de investigação sobre Território e Sociedade. É o principal investigador da equipe portuguesa envolvida no projeto internacional “The Development of European Waterways, Road and Rail Infrastructures: A Geographical Information System for the History of European Integration (1825–2005)”, da European Science Foundation. Este trabalho foi financiado pela Fundação da Ciencia e Tecnologia.

Nuno Miguel Lima é um aluno de doutoramento na Faculdade de Ciências Sociais e Humanas da Universidade Nova de Lisboa, onde trabalha sobre caminhos de ferro transnacionais no espaço ibérico, sendo também investigador do Instituto de História Contemporânea da mesma Universidade. Bolseiro da Fundação para a Ciência e a Tecnologia.

Ana Alcântara is investigadora no Instituto de História Contemporânea, Faculdade de Ciências Sociais

• 95

Railroads in historical context: construction, costs and consequences

e Humanas da Universidade Nova de Lisboa. Apresentou uma dissertação de mestrado em sistemas de informação geográfica sobre os caminhos de ferro na Cova da Beira (1878-1930).

Abstract Resumo

96 •

Historiography on Portuguese railroads has not paid much attention neither to their regional effects, nor to the relationship between the operation of this form of transportation and population evolution. In fact, available studies usually adopt a national approach and analyze the contribution of railroads to the overall development of the country.

The main argument of this paper is that in the Portuguese case, access to railroads reinforced pre-existing territorial inequalities and promoted different regional dynamics, mainly with regard to population growth, urban development and population mobility.

In reality, in the more developed regions, railroad access helped increase population concentration in the areas served by this infrastructure. The railways also favored the growth of pre-existing urban centers and the emergence of new ones. They also encouraged migration into towns, thus contributing to their growth. In the Inland North, where the Tua line is integrated, traditionally affected by greater transportation difficulties, railroads seem to have operated in the opposite direction, contributing to a decline in population relative to the other regions of Portugal. Moreover, this area continued to be characterized by a predominance of modest-sized cities, unable to match the dynamism of the urban centers in the coastal regions or to attract a migrant population to aid in their development. Instead, since the end of the nineteenth century, this region suffered from an increasing emigration that railroads seem to have facilitated. To explain this evolution we must also take into account the economic crisis that affected the agriculture of this part of the country, but the presence of the railroad seems to have been a significant factor.

In this paper, we will try to put the Tua line in this context, comparing its effects on population with those caused by the Beira Baixa line in another Portuguese mountainous region around the city of Covilhã.

A historiografia sobre os caminhos de ferro portugueses não tem prestado muita atenção aos seus efeitos regionais ou à relação entre esta forma de transporte e a evolução da população. De facto, os estudos disponíveis adoptam geralmente uma perspetiva nacional e analisam apenas a contribuição dos caminhos de ferro para o desenvolvimento global do país.

O principal argumento deste trabalho é que, no caso de Portugal, o acesso aos caminhos de ferro reforçou as desigualdades territoriais anteriores e promoveu diferentes dinâmicas regionais, especialmente relacionadas com o crescimento da

Luís Espinha da Silveira • Nuno Miguel Lima • Ana Alcântara

população, o desenvolvimento urbano e a mobilidade das populações.

Na realidade, nas regiões mais desenvolvidas, o acesso aos caminhos de ferro contribuiu para o aumento da concentração da população nas áreas servidas pela infraestrutura; os caminhos de ferro favoreceram ainda o crescimento dos centros urbanos pré-existentes e a emergência de novos centros. Encorajou também a migração para as cidades e vilas, contribuindo para o seu crescimento. No interior Norte, onde a linha do Tua estava integrada, região tradicionalmente afectada por maiores dificuldades de transporte, o caminho de ferro parece ter operado na direção oposta, contribuindo para um declínio da população relativamente às outras regiões de Portugal. Além disso, esta área continuou a ser caracterizada pelo predomínio de cidades de dimensões modestas, incapazes de concorrer com o dinamismo dos centros urbanos das regiões costeiras ou de atrair uma população migrante capaz de promover o seu desenvolvimento. Ao invés, desde o final do século XIX, esta região sofreu um aumento da emigração que o caminho de ferro parece ter facilitado. Para explicar esta evolução devemos também ter em consideração a crise económica que afetou então a agricultura desta região, embora a presença da linha de caminho de ferro pareça ter sido um fator significativo.

Neste trabalho tenta-se pôr a linha do Tua neste contexto, comparando os seus efeitos na população com aqueles causados pela linha da Beira Baixa noutra região montanhosa portuguesa em torno da cidade da Covilhã.

• 97

Railroads in historical context: construction, costs and consequences

98 •

Luís Espinha da Silveira • Nuno Miguel Lima • Ana Alcântara

The impact of railroad accessibility on the population of Portugal’s Inland North Region (1878-1930).

The Tua and the Beira Baixa lines Luís Espinha da Silveira, Nuno Miguel Lima e Ana Alcântara Instituto de História Contemporânea, Faculdade de Ciências Sociais e Humanas, Universidade Nova de Lisboa. This article was produced within the research project “The Development of European Waterways, Road and Rail Infrastructures: A Geographical Information System for the History of European Integration (1825-2005),” as part of the EUROCORES program of the European Science Foundation. The Portuguese team was funded by Fundação para a Ciência e a Tecnologia (INVENT/0001/2007).

Historiography on Portuguese railroads has not paid much attention neither to the regional effects of the construction of this infrastructure, nor to the relationship between the operation of this form of transportation and population evolution. The exceptions are represented by the classic work by Maria Fernanda Alegria, who studied the formation of the national market, and by two recent studies, one on the role of railroads in the relocation of industry during the first half of the twentieth century and the other on their link with the urban network, focusing primarily on Lisbon and Oporto1. Other studies focus on the country at large and analyze the contribution of railroads to national development. Usually, they adopt a pessimistic view and emphasize the prevalence of private and foreign interests, as opposed to those of the nation. They also underline the financial burden of railroad construction for the state and the corruption associated with it, as well as the incapacity of the Portuguese economy to supply the capital 1 Maria Fernanda Alegria, A organização dos transportes em Portugal, 1850-1910: as vias e o tráfego, Lisbon, Centro de Estudos Geográficos, 1990; Maria Eugenia Mata, “As bees attracted to honey. Transport and job mobility in Portugal, 1890-1950”, Journal of Transport History, vol.

29, 2, 2008; Magda Pinheiro, Cidade e caminhos-de-ferro, Lisbon, Centro de Estudos de História Contemporânea, 2008.

• 99

Railroads in historical context: construction, costs and consequences

100 •

and industrial goods necessary for the development of the railroad network, limiting its contribution to the supply of manpower2.

In previous studies, we tried to show that in Portugal, in a period similar to the one examined here, accessibility to railroads reinforced pre-existing territorial inequalities and promoted different regional dynamics, with respect to population growth, urban development, internal migration and emigration. We particularly suggested that in the Inland North, railroads seem to have contributed to a decline in population in comparison with the other regions of Portugal and, in contrast to what happened in the rest of the country, they did not stimulate urban growth. This area, whose boundaries we will explain later, was still unable to attract a migrant population to aid in its development and most likely fueled migration to other regions. Furthermore, it was gradually affected by emigration that the railroad would also have facilitated3.

This paper deepens the evolution of the Inland North and examines two areas included in its territory: Tua, served by the line with the same name, and Cova da Beira, served by the Beira Baixa line.

The research is focused on the period between 1878 and 1930, to cover the arrival of the railroad in the Inland North and to exclude the effect of bus services.

In some analyses, however, a longer timeframe was considered.

This research team has been particularly interested in exploring the link between space and History and has been applying Geographic Information Systems (GIS) to Portugal’s history. This effort began years ago with the construction of a cartographic infrastructure supported by a GIS, which now includes parish maps of mainland Portugal from 1758 up to 2001 and maps from all the existing administrative divisions above this level4. At the same time, a consid2 António Lopes Vieira, The role of Britain and France in the finance of Portuguese railways, 18501890: a comparative study in speculation, corruption and inefficiency, Leicester, University of Leicester, 1983; Magda Pinheiro, Chemins de fer, structure financiere de I’Etat et dependance exterieure au Portugal (1850-1890), Paris, Univ. Paris 1, 1986; Álvaro Ferreira da Silva and Lara Tavares, “Transport History in Portugal: a bibliographical overview”, in Michele Merger e MarieNoelle Polino (eds.), COST 340, towards a European intermodal transport network lessons from history: a critical bibliography, Paris, AHICF, 2004; Maria Eugénia Mata and Lara Tavares, “The value of Portuguese railways for consumers on the eve of the First World War”, TST: Transportes, Servicios y Telecomunicaciones, 7, 2003, pp.80-100; Angela Salgueiro, A Companhia Real dos Caminhos de Ferro Portugueses, 1859-1860, Lisbon, Faculdade de Ciências Sociais e Humanas, 2008.

3 Luis Espinha da Silveira, Daniel Alves, Nuno Miguel Lima, Ana Alcântara, and Josep Puig, “Population and railways in Portugal, 1801-1930”, Journal of Interdisciplinary History, vol. 42, 1, 2011, pp.29-52; idem,”Caminhos de ferro, população e desigualdades territoriais em Portugal, 1801-1930”, Ler História 61, 2011, pp.7-37.

4 The maps used in this article have been developed by several teams since the early 1990s. Initial results were presented at the Ninth International Conference of the Association for History and Computing in 1994, giving rise afterwards to several printed publications - among them, Luis Espinha da Silveira, Margarida Lopes, and Cristina Joanaz de Melo, “Mapping Portuguese Historical Boundaries with a GIS” in Onno Boonstra, Geurt Collenteur, and Bart van Elderen (eds.), Structures and contingencies in computerized historical research, Hilversum, 1995, pp.245252; Luis Espinha da Silveira, Território e poder: Nas origens do Estado contemporâneo em

Luís Espinha da Silveira • Nuno Miguel Lima • Ana Alcântara

erable body of data on Portuguese population in the nineteenth and twentieth centuries was collected. Recently, the geographic research scope was extended to the entire Iberian Peninsula.

To study the effect of accessibility to railroads, continental Portugal was divided into three regions (Atlantic North, Inland North and Mediterranean South), whose boundaries were defined mainly based on topography and slope, factors that influence the movement of people and goods. In any case, other criteria were considered, such as climate, settlement patterns, landowning patterns and social structure (figure 1).

Figure 1 - Portugal’s Three Regions

• 101

In this research we tried to define a railroad accessibility criterion relatively simple to implement and able to adapt to a country with very different topographic characteristics. Since parishes are the smallest territorial units for which population data can be provided over a long period of time, they would seem to be a good place to start when exploring railroad accessibility. In the original outline for this study, a parish had access to a railroad only when it had a station5.

This criterion proved to be too restrictive, especially in flat-land areas, since it tended to exclude neighboring parishes with a station not far away. Therefore, the understanding of accessibility was broadened to include parishes with a centroid no more than 5 km away from a station. The application of these two conditions turned out to be well suited to the characteristics of the different regions.

Portugal, Cascais, 1997. The results have also been available on the Internet since 2001 (http:// www.fcsh.unl.pt/atlas/) and more recently on a website based on open source software (http://atlas. fcsh.unl.pt/).

5 This criterion was also followed by Ian N. Gregory and Jordi Marti-Henneberg, “The railways, urbanization, and local demography in England and Wales, 1825-1911”, Social Science History, vol.

34, 2, 2010, pp.199-228.

Railroads in historical context: construction, costs and consequences

The criteria chosen apply equally to areas where the parishes are small and to others where they are larger, as well as to flat-land and mountainous regions6.

THE CONSTRUCTION OF THE RAILROADS

102 •

As Maria Fernanda Alegria showed, Portugal arrived late to the railroad era and built a small network in relation to its population and area of the territory7.

Bearing in mind the censuses years, it appears that the pace of construction was irregular, alternating between periods of rapid growth (1856-1864, 1878-1890, 1900-1911) and phases of downturn, corresponding to moments of economic and financial hardship8.

As in many other countries, the capital city played a pivotal role. The first priority was to connect Lisbon with Oporto, Spain and the South. The lines around the second biggest city, Oporto, were already in operation in 1878, moment in which the crossing of the Atlantic North was completed. During the great construction period (1878-1890), railroads arrived in the Inland North, in Algarve and in the coastal region north of Lisbon, a major supply area of the capital. During these years four international connections were also established.

At that time, the first section of the Tua line was opened between the station of the same name in the Douro line and Mirandela (1887). From that period on, the focus was on the construction of the regional lines, among others, the Beira Baixa one, opened until Covilhã in 1891 and ended in 1893 when it reached Guarda. The lines in the extreme northeast progressed considerably in the early twentieth century, among them the Tua, which reached Bragança in 1906. As for the regional distribution of the infrastructure, the Atlantic North, with 69.1 km of railroad per 1,000 km2 in 1930 was greatly benefited by comparison with the South (31.9 km) and the Inland North (28.9 km).

THE TUA AND COVA DA BEIRA AREAS The Tua and Cova da Beira areas were defined by selecting the municipalities crossed by the Tua and Beira Baixa lines. In the first case, the entire line was 6 Data published in the article by Magda Pinheiro, Nuno Miguel Lima and Joana Paulino, “Espaço, tempo e preço dos transportes: a utilização da rede ferroviária no final do século XIX”, Ler História, 61, 2011, pp.39-64, indicate that the cost of passenger and freight transportation substantially increases beyond 10km away from a station. This result suggests a reconsideration of the distance used so far to define accessibility.

7 Maria Fernanda Alegria, A organização dos transportes em Portugal, p.214.

8 Magda Pinheiro, “Transportes”, in Nuno Valerio (ed.), Estatísticas históricas portuguesas, Lisbon, Instituto Nacional de Estatistica, 2001, pp.372-376.

Luís Espinha da Silveira • Nuno Miguel Lima • Ana Alcântara

included, but in the second one only the section between the stations of Castelo Novo e Belmonte-Manteigas was included9.

The Tua region, with a total area of 3,561 km2, is located in the extreme northeast of Portugal and includes in its territory the Nogueira and Bornes mountains. It consists of seven municipalities and 182 parishes. Cova da Beira, between the Estrela and Gardunha mountains, has a much smaller area (1,375 km2) divided by three municipalities and 59 parishes (figure 2).

Figure 2 - Case Study Areas

• 103

In accordance with the mountainous characteristics of the Inland North, in altitudes in both areas are usually high, although the two zones present important topographical differences. In Tua, the lands between 600 and 800 meters predominate and the terrain is very rugged, with slopes between 0% and 88.6% (figure 3).

Figure 3 - Tua – Altitude and Slope

9 The effect of railroads on population in the Cova da Beira area is the subject of Ana Alcantara’s MSc dissertation in Geographic Information Systems and Science [Railway and Population in Cova da Beira (1878-1930). An Accessibility Model, ISEGI - UNL (http://hdl.handle. net/10362/6595)]. In this research, she takes into consideration population distribution by settlements, rather than by parishes, allowing the application of more refined methodologies of spatial analysis. So far, it has not been possible to follow this approach to the Tua area.

Railroads in historical context: construction, costs and consequences

Cova da Beira is mostly a plateau region, where most lands lay between 400 and 600 meters, with slopes generally less pronounced (figure 4).

Figure 4 - Cova da Beira – Altitude and Slope

104 •

The density and importance of rivers that flow through these zones also differ. In the Tua area there are three important rivers: Douro, Tua and Sabor, which run through deep valleys. Only the first was navigable, which explains its strategic significance to the region. However, many other smaller watercourses flow in this territory and mark the landscape of the region.

In the Cova da Beira area there is only one main river, Zêzere, non-navigable in this section. Secondary and seasonal streams, in smaller numbers than in the Tua area, had a fundamental interest for the development of the woolen industry established here.

Both zones are characterized by a concentrated settlement pattern and by the presence of two urban centers of certain relevance in the context of the Inland North: Bragança in Tua and Covilhã in Cova da Beira. However, while Bragança only surpassed the 5,000 inhabitants threshold in 1878, arriving in 1930 with only 6,089 inhabitants, Covilhã had already reached 10,809 inhabitants in 1878, a figure that rose to 15,640 in 1930. As a result, the weight of the urban population in each area was very different: it averaged 4.7% in Tua and 16.9% in Cova da Beira10.

Regarding economic activities, agriculture and grazing dominated in Tua, while in Cova da Beira, even if the primary sector had a significant presence, the woollen industry traditionally played a major role. The weight of the industrial activity in Cova da Beira, concentrated in the municipality of Covilhã is 10 The definition of urban center herein is an agglomeration with a de facto population of 5,000 or more inhabitants in its total number of parishes. The parish can stand as the fundamental unit because it is the most detailed one that is common to every Portuguese census and the only one that can support a comparative historical, urban spatial analysis. The adoption of the minimum limit of 5,000 inhabitants was borne of the need to take into account the small size of Portuguese urban centers throughout the period covered by the database. To select the urban agglomerations existing throughout the different years and to avoid the inclusion of parishes with large but scattered populations that were not part of any urban center, the censuses from 1911 and 1940 were used as ancillary data. In fact, these censuses identify not only the total population of each parish, but also the people living in each of its settled areas.

Luís Espinha da Silveira • Nuno Miguel Lima • Ana Alcântara

confirmed by the industrial survey of 1890 which identified facilities with more than five workers. This source indicates that in Covilhã, there were 100 factories almost exclusively engaged in wool processing, which employed a total of 3,676 workers. These figures compare with the incredibly low number of plants (4) in all the Tua area, employing 36 workers. In addition, large modern industrial units had already been established in Covilhã, using steam energy. This survey, however, leaves out much of craft and domestic industry which was a reality in the two areas in question.

POPULATION After a long period of crisis, as the economy recovered and efforts to modernize the country produced some results, the annual growth rate (CAGR) of the population of mainland Portugal, between 1849 and 1911, improved and reached a moderate level (0.79%). Between 1911 and 1920, the population almost stagnated, mainly because of high rates of emigration, the effects of World War I, and the influenza epidemic of 1918; thereafter, its growth resumed, values reaching more than 1 percent per year. This population-growth pattern is the result of a late demographic transition and is similar to that of other southern European countries11.

Unlike the Atlantic North and the South, between 1849 and 1911, the population of the Inland North increased at a rate below the national average (0.54%) and even decreased between 1911 and 1920 (-0.42%). The figures recovered in subsequent years, but again the growth rate (0.64%) was half the value calculated for the whole country and was below the value of the Atlantic North and of the South.

The differences in population evolution among the various regions are in line with the development of their economies. In particular, the slow growth rate in the number of inhabitants of the Inland North is linked to the difficulties of the regional agriculture, which from the last decades of the nineteenth century gradually lost weight in the national context12. The explanation of these facts is 11 In the first half of the nineteenth century, Portugal’s CAGR was lower than Italy’s (0.6%) or Spain’s (0.6%), but thereafter, until the end of the century, Portugal’s growth was similar to the one observed in these countries, though still far from that of other states in northern Europe, such as Belgium (1.2%), the Netherlands (1.5%), and Britain (1.6%). Neil Tranter, Population since the Industrial Revolution: The case of England and Wales, London, 1973, p.43; Massimo Livi Bacci, The Population of Europe: A History, New York, 2000, p. 132; David Reher, “The Demographic Transition Revisited as a Global Process,” Population, Space and Place, X, 2004, pp.22, 35, 37; Teresa Rodrigues Veiga, A população portuguesa no século XIX, Oporto, 2004, pp.20-22; idem, “A transição demográfica,” in Pedro Lains and Ferreira da Silva (eds.), História Económica de Portugal, 1700-200: O século XIX, Lisbon, 2005, pp.37-63; idem (ed.), História da população portuguesa: das longas permanencias a conquista da modernidade, Oporto, CEPESE - Centro de Estudos da População, Economia e Sociedade, 2008.

12 Data published by David Justino indicate that compared with the whole country, the percentage

• 105

Railroads in historical context: construction, costs and consequences

complex and does not correspond to the purpose of this paper, except in what concerns the possible contribution of railroads to the development of population, cities and migration.

Up to 1911, growth in the population of Cova da Beira was faster (1.01%) and far exceeded that of the country and of the Inland North. In the Tua area, the growth rate was higher than the latter (0.72%). In the years 1911-1920, the areas in question followed the downward trend of the Inland North, but the contraction was particularly important in Tua (-1.36%). The next decade was more positive everywhere and both Tua and Cova da Beira increased at a rate closer to the national average (respectively 1.05% and 1.04%) (figure 5).

Figure 5 - Population Growth (CAGR) 1,5 1,0 CAGR

0,5

106 •

0 -0,5 -1,0 -1,5

1849-1911 Portugal

1911-1920 Tua Inland North

1920-1930 Cova da Beira

The case study areas were defined according to the railroad presence. Therefore, the percentage of population with access to such transportation was naturally high, compared to the Inland North (up to 32.5% in 1930). In Tua, it reached 21.65% at the opening of the Mirandela section (1890) and 39.82% in 1930; in Cova da Beira accessibility was much higher, reaching 59.15% of the population in 1900, shortly after the completion of the line. This difference may be explained by the smaller area encompassed by Cova da Beira and by the presence of a medium-size urban center.

Across the country, the effect of railroad accessibility on population growth is reflected in the fact that parishes with access grew constantly faster than those of the agricultural production of Vila Real, Bragança, Viseu e Guarda districts, which correspond, roughly to the Inland North region, decreased from 1880 until the 1920s. This evolution applies to grain and wine production, and to sheep and swine breeding from 1870 up to 1925. Moreover, Oporto wine exports also experienced a decline in their share in Portugal’s exports from 1880 until the break of World War I. David Justino, A formação do espaço económico nacional, Lisbon, 1989, II, pp. 136-144. As for the exports, see Pedro Lains, A economia portuguesa no século XIX: crescimento económico e comércio externo, 1851-1913, Lisbon, Imprensa Nacional-Casa da Moeda, 1995, p.92.

Luís Espinha da Silveira • Nuno Miguel Lima • Ana Alcântara

without it, throughout the period from 1878 to 1930. This trend was also found in the Atlantic North and somehow in the South, but not in the Inland North. In fact, in this latter region, between 1878 and 1920, the difference in the growth rates of two groups of parishes (with and without access) was very low and in the decade of 1890-1900 was even unfavorable to locations with access. The growth of these parishes only took off at the end of the period under analysis (1920-1930), following a national trend (figure 6).

Difference to the global CAGR

Figure 6 - Railroads and Inland North Population Growth (CAGR) 0,4 0,3 0,2 0,1 0,0 -0,1 -0,2 -0,3 -0,4 With

1878 - 1890 Without

1890 - 1900 1900 - 1911 Polynominal (with Rrd)

1911 - 1920 1920 - 1930 Polynominal (without Rrd)

In the Tua area, the pattern is characterized by a fluctuation in the growth trends of both groups of parishes, sometimes favorable and sometimes unfavorable to those that were accessible. Compared to the Inland North, the differences between these two groups were higher and sometimes were countercyclical (figure 7).

Difference to the global CAGR

Figure 7 - Railroads and Tua’s Population Growth (CAGR) 0,4 0,3 0,2 0,1 0,0 -0,1 -0,2 -0,3 -0,4 With

1878 - 1890 Without

1890 - 1900 1900 - 1911 Polynominal (with Rrd)

1911 - 1920 1920 - 1930 Polynominal (without Rrd)

• 107

Railroads in historical context: construction, costs and consequences

The parishes of the Cova da Beira area follow a pattern different from both the Inland North and the national one. In effect, the parishes with access to railroads grew persistently less than the others until 1920. In any case, after 1900, trends in this area have similarities with those observed in the Tua area (figure 8).

Difference to the global CAGR

Figure 8 - Railroads and Cova da Beira’s Population Growth (CAGR) 0,4 0,3 0,2 0,1 0,0 -0,1 -0,2 -0,3 -0,4 With 108 •

1890 - 1900 1900 - 1911 Polynominal (with Rrd) Without

1911 - 1920 1920 - 1930 Polynominal (without Rrd)

In short, with the exception of the 1890-1900 decade in the Tua area and of the 1920-1930 period in both areas, the existence of railroads did not favor population growth in parishes with accessibility over those without it. In both areas, this result is clearer than in the Inland North.

By spatially analyzing gains and losses of population in the Tua area it is not possible to identify territories with constant patterns throughout the entire period. However, between 1878 and 1890 the positive evolution in of the parishes is in the northern zone, close to Sabor river, which by then were very far from the Tua line. Between 1911 and 1920 losses were generalized, but this period was followed by a decade of strong growth in a large part of the territory, contrasting with the losses up to the north (figure 9).

Figure 9 - Tua’s Population Growth (CAGR)

Luís Espinha da Silveira • Nuno Miguel Lima • Ana Alcântara

To deepen the analysis on the effect of railroad proximity, we defined a multiple ring buffer of accessibility up to 20 kilometers, with intervals of 5 kilometers. The Spearman’s Rank Correlation Coefficient shows that this relation was only significant in the years 1878-1890 and 1911-1920, with a positive value in both cases (0.164 and 0.255, respectively). This means that, in the first period, as the distance to railroads increased, the higher the population growth was, and in the second period, the lower were its losses. In this sense, proximity to railroad was not a guarantee of population growth during these periods.

In the Cova da Beira area it is not also possible to identify territories with constant patterns throughout the entire period under analysis and the correlation between proximity to railroad and population growth is never significant.

URBANIZATION The rate of European urbanization (excluding Russia) changed drastically from 19% in 1850 to 48% in 1930. In the case of Portugal, following the criterion adopted herein, the increase in the rate of urbanization was more modest, registering values between around 13% in 1864 and 24% in 1930, which was half of the value reported by Bairoch for the whole of Europe in 1930. Considering the relationship between urbanization rate and national wealth, this result is a clear sign of the difficulties of the country’s economic and social modernization13.

The Atlantic North, which was already the most urbanized region in 1878, not only kept its position, but reinforced it; as to the South’s rate, it was also relatively high and grew. The difference between these two regions and the Inland North is outstanding. In the latter case, the urbanization rate, although 13 Paul Bairoch, Cities and economic development: from the dawn of History to the present, Chicago, University of Chicago Press, 1988, p.216. By comparison with Spain, this limitation in urban growth was significant. In 1860, 22.5% of Spain’s population resided in urban agglomerations of more than 5,000 inhabitants; by 1930, the figure rose to 37%. David S. Reher, “Ciudades, procesos de urbanization y sistemas urbanos en la Peninsula Iberica, 1550-1991”, in Manuel Guardia i Bassols, Francisso Xavier Monclus, and Jose Luis Oyon (eds.), Atlas histórico de ciudades europeas: Peninsula Iberica, Barcelona, Salvat, 1994, p.25. See also Gabriel Tortella, El desarrollo de la Espana Contemporanea . Historia Economica de los siglos XIX y XX, Madrid, 2006, p.37.

• 109

Railroads in historical context: construction, costs and consequences

Table 1: Urban Population by Region (%) Regions

1878

1890

Atlantic North

22.1

25.9

27.8

29.5

33.1

34.9

Mediterranean South

15.7

16.5

18.2

17.7

22.7

23.6

2.9

3.4

3.2

3.1

3.5

3.6

14.1

16.5

17.9

19.0

22.2

23.8

Inland North Total

110 •

1900

1911

1920

1930

growing, maintained an incredible low level throughout the period (table 1); moreover, this region had few urban centers (seven) and almost all of them were small (less than 10,000 inhabitants)14. The exception was the city of Covilhã, which joined the medium-sized ranks in 1878. In this context, the Inland North projects an image of true urban lethargy, which is added to the scant growth in population.

The weakness of the urban centers at the time of the arrival of railroads resulted, hypothetically, from the region’s isolation and a poor agricultural economy, with the exception of wine production, namely Port wine. However, the arrival of railroads did not stimulate urbanization as it did in the Atlantic North and in the South, where a significant number of new urban centers appeared close to the area of influence of the railroad. In fact, in the Inland North only Guarda joined the pre-existing six urban centers.

In addition, the growth of the urban population in the Inland North between 1878 and 1930 (51%), was three times lower than the value for the whole country. Covilhã’s and Bragança’s growth (45% and 20%, respectively) was even more modest than that of its region. The near stagnation of Inland North’s cities might be related with the blocking of this region’s agriculture, as previously referred. The long-term evolution of Covilhã hides, nevertheless, the contraction of its population in the years 1890-1900 and 1911-1920. The beginning of Covilhã’s crisis coincides with the arrival of the railroad and with the start of a depressive economic cycle affecting the industry throughout the country.

INTERNAL MIGRATIONS Due to the absence of statistical information, it is difficult to study internal migrations. However, population censuses from 1890 to 1930, indicate for each par14 The urban centers of this region were Bragança, Chaves, Covilhã, Guarda (from 1920 onwards), Lamego, Vila Real and Viseu.

Luís Espinha da Silveira • Nuno Miguel Lima • Ana Alcântara

ish, the number of people born outside the municipality where they were at the time of the survey. By calculating the percentage of these individuals it is possible to estimate the parishes attractiveness and the mobility of the population.

The number of people registered under this category varied between 512,000 in 1890 and 844,000 in 1930, representing between 11% and 13% of the entire population, a number relatively modest considering the standards of some European countries, even though they were in line with those of Spain. However, the growth registered points out to the reinforcement of mobility of the Portuguese population, coincident with the extension of the railroad network to the entire national territory15.

The geographical analysis highlights a clear dichotomy between North and South, with the province of Alentejo, part of Ribatejo and the metropolitan arFigure 10 - Percentage of Migrant Population per Parish (Difference to the overall average percentage)

• 111

15 For the period between 1890 and 1930, the figures for mobility at the provincial level in Spain, and at the district level in Portugal, are relatively close, 9.8% and 7.8% respectively. See Javier Silvestre Rodriguez, “Las migraciones interiores en España durante los siglos XIX y XX: una revision bibliografica”, Cuadernos económicos de ICE, LXX, 2005, pp. 164-166; Álvaro Ferreira da Silva, “Padrões de mobilidade interna em Portugal na segunda metade do século XIX,” in José Vicente Serrão, Magda Pinheiro, and Maria de Fátima Sá e Melo Ferreira (eds.), Desenvolvimento económico e mudança social, Lisbon, 2009, pp.375-392; Joaquim Costa Leite, “População e crescimento económico”, in Pedro Lains and Álvaro Ferreira da Silva (eds.), História Económica de Portugal, 1700-2000: O século XX, Lisboa, Imprensa de Ciências Sociais, 2005, pp.73-74.

Railroads in historical context: construction, costs and consequences

eas of Lisbon and Oporto appearing as the zones with higher attractiveness.

Moreover, the trend observed throughout the period of 1890-1930 shows a rising geographic concentration of migrant population, particularly in urban areas (figure 10).

A more detailed statistical analysis stresses out that a significant portion of this population - on average, nearly 67% throughout the period studied - was registered in the Atlantic North, 20% in the South and the remaining 13% in the Inland North, the least attractive region from this point of view.

Beyond its low attractiveness, in the Inland North migrant population at most accounted for 7.2% of the region’s inhabitants, a value reached in 1900. This score represented half of the South value (15.1% in 1911) and scarcely more than one third of that of the Atlantic North (20.6% in 1911).

During the years 1890-1930, the relative weight of Tua’s migrant population was always higher than that of the Inland North, fluctuating between 12.3% in 1900 and 6.9% in 1930. At first, the values of Cova da Beira exceeded those of the Inland North region (8.3% in 1900), but then onwards they started decreasing, and by 1920 they were lower than those of the Inland North (figure 11).

112 •

Figure 11 - Relative Weight of Migrant Population 14% 12% 10% 8% 6% 4% 2% 0

1890

1900 Inland North

1911 Tua

1920

1930

Cova da Beira

Over half of the migrant population (55%) lived in urban centers across the country. Confirming the dynamics of the Atlantic North, in this region that percentage reached 69% in 1890 and 75% in 1930, particularly because of Lisbon and Oporto; the South had lower percentages, but even so meaningful (30% and

Luís Espinha da Silveira • Nuno Miguel Lima • Ana Alcântara

43%, respectively). In the Inland North, the weak urban vitality is once again reflected in the considerably lower importance of urban migrant population, and in its tendency to diminish (16% and 15%).

Although Bragança was a small city, the weight of its migrant population was slightly higher than the correspondent figures for the Inland North urban centers (18.7% and 19.5%). In 1890, Covilhã had a noteworthy value (52.1%), which drastically decreased in the following periods, reaching 23.5% in 1930. To a certain degree, this reduction explains the city’s population contraction between 1911 and 1920.

The railroad seems to have favored internal migrations all over the country.

In fact, migrant population normally had higher percentages in parishes with access to railroad than in parishes without access. This national pattern was equally visible in the Atlantic North and in the South, but the influence of accessibility was much lower in the Inland North (figure 12).

Difference to the overall percentage

Figure 12 - Inland North – Railroads and Migrant Population 6,0 5,0 4,0 3,0 2,0 1,0 0 -1,0 -2,0 -3,0 -4,0 With

• 113

1890 Without

1900 1911 Polynominal (with Rrd)

1920 1930 Polynominal (without Rrd)

In the Tua area, the weight of migrants in parishes with access to railroad was equally superior to that obtained in parishes without access, particularly until 1911, date when the values were higher than those of the Inland North, (figure 13). The pattern in the Cova da Beira area was similar to that of the Inland North (figure 14).

Railroads in historical context: construction, costs and consequences

Difference to the overall percentage

Figure 13 - Tua – Railroads and Migrant Population 6,0 5,0 4,0 3,0 2,0 1,0 0 -1,0 -2,0 -3,0 -4,0 With

1890 Without

1920 1930 Polynominal (without Rrd)

1900 1911 Polynominal (with Rrd)

114 •

Difference to the overall percentage

Figure 14 - Cova da Beira – Railroads and Migrant Population 6,0 5,0 4,0 3,0 2,0 1,0 0 -1,0 -2,0 -3,0 -4,0 With

1900 Without

1911 Polynominal (with Rrd)

1920

1930 Polynominal (without Rrd)

EMIGRATION As had happened in several European countries, from the 1870s onwards, Portuguese emigration progressively grew until reaching its peak in 1912. From this year until 1932, emigration continued to be significant, but its values suffered strong variations, particularly after World War I, and showed a tendency to decrease16.

The map highlights the difference between the North and the South, clearly 16 Maria Ioannis Baganha and José Carlos Marques, “População” in Estatísticas históricas portuguesas, pp.83-85.

Luís Espinha da Silveira • Nuno Miguel Lima • Ana Alcântara

outlined by the Massif Central (figure 15)17. In most of the years, nearly all of the municipalities which had a gross emigration rate close to or higher than the national average were north of that line. In the South, there was a constant low level of emigration.

Another important aspect to point out in the North of the country, is the variation of the areas with higher intensity of emigration, which progressively spread to the interior and to the South. The Cova da Beira area continued to be little affected, with an average gross emigration rate of 3.1‰ in 1911, a year of massive emigration in Portugal. In the Tua area, rates were significantly higher, with an average of 33.1‰ in that same year.

Figure 15 - Gross Emigration Rate per Municipality (Difference to the overall average percentage)

• 115

It is interesting to notice that the spreading of emigration to the interior seems to follow the progressive penetration of the railroad network in that region, as is the case of the municipalities served by the Douro, Corgo, Tua and Sabor lines in 1911. The observation that in the Inland North, especially in 1911 and 1920, the municipalities crossed 17 Data on emigration is only available per municipality and from 1887 onwards. In addition, there is no information for 1900, the year of the population census. This circumstance explains the inclusion of 1901 in the analysis.

Railroads in historical context: construction, costs and consequences

by the railroad had higher gross emigration rates goes in favor of this idea.

However, the data available (only at municipal level) are not sufficiently detailed to support deeper and conclusive analysis on the effect of railroad on emigration, both in the Tua area and the Inland North region.

CONCLUSION

116 •

As previously mentioned, the relation between railroads and population dynamics in Portugal has not received great attention from the historiography, much less on a regional perspective. In previous works, we tried to show that, in each region, railroads might have had different effects on population growth, urban evolution and internal migrations. These effects were positive in the Atlantic North and in a more moderate way, in the South, and were negative in the Inland North, where railroads still might have favored emigration.

In this paper we have tried to deepen previous conclusions regarding the latter region, by analyzing two areas there integrated: Tua and Cova da Beira.

Both of them were served by railroad lines and were located at high altitudes, in the first case with very strong slopes in a large part of its territory. They both include urban centers of different dimensions and characteristics: a small city, Bragança, in the case of Tua; in Cova da Beira, Covilhã, a medium-sized center, with a relevant industrial activity, a unique feature in the Inland North context.

The analysis of the parishes’ demographic evolution in both areas confirms and reinforces what was already known of the Inland North region; that is, with the exception of the 1890s in Tua and of the 1920s in both areas, the railroad did not favor population growth of the territories with accessibility. This result is even clearer in the two areas, than in the Inland North region. The correlation between distance to the stations and population growth is not significant in Cova da Beira, and in Tua it is only valid in two periods, when a greater distance corresponded to a more favorable evolution.

Throughout the period under analysis, the cities of Covilhã and, particularly, Bragança experienced a slower growth than that of the Inland North urban centers, which was already very weak. Covilhã even suffered a reduction of its population during the 1890s, at the same time it received the Beira Baixa line and coinciding with the national economic crisis.

In a region with a scarce capacity to attract a migrant population, Tua and Cova da Beira initially had a positive performance. However, in both cases the percentage of migrants rapidly decreased and came close to the levels of the In-

Luís Espinha da Silveira • Nuno Miguel Lima • Ana Alcântara

land North. In this region, the railroad slightly favored the attraction of migrants to accessible areas. This idea also applies to Cova da Beira. Until 1911, parishes with accessibilty in Tua showed a greater attractiveness. Before gaining access to the railroad network, migrant population in the city of Covilhã represented an important share of the migrants in the Cova da Beira area (52%), but the reduction from 1890 onwards was drastic.

By the end of the nineteenth century, several parts of the North of Portugal were deeply affected by emigration. The Tua area was among them, contrarily to Cova da Beira. Although in certain cases the visual examination of the maps may suggest an association between emigration and railroad expansion, which is equally suggested by the limited statistical information, the data have very little detail to allow a deeper and more conclusive analysis.

• 117

Eduardo Beira

TUA VALLEY: HOW DIFFERENT IS IT NOW? AN INTRODUCTION TO POPULATION DYNAMICS (1864 – 2011) VALE DO TUA: QUAIS AS DIFERENÇAS? UMA INTRODUÇÃO À DINÂMICA POPULACIONAL (1864-2010) Eduardo Beira (MIT Portugal Program, U. Minho, Portugal) Professor (by invitation) in School of Engineering, University of Minho, from 2000. From 2008, equivalent to associated professor and EDAM (Engineering Design and Advanced Manufacturing) Professor in MIT Portugal Program.

Graduated in chemical engineering (U. Porto, 1974). For twenty years he has worked as top manager in companies, both in industry and services sectors, and has been an executive member of several boards, after a first academic career in Faculty of Engineering (University of Porto).

Scholar interests in the areas of innovation, engineering and technology, regional development.

Professor (convidado) da Escola de Engenharia da Universidade do Minho, desde 2000. Atualmente equiparado a Professor Associado. Professor EDAM (Engineering Design and Advanced Manufacturing) do programa MIT Portugal.

Engenheiro químico (1974, FEUP), foi gestor e administrador de empresas industriais e de serviços durante mais de vinte anos, depois de uma primeira carreira académica na Faculdade de Engenharia da Universidade do Porto.

Interesses académicos pelas questões de inovação, desenvolvimento, engenharia e tecnologia, indústrias “tradicionais”.

Abstract Resumo

The Tua Valley was different when a decision to build the Tua railways had been made (beginning of 1880s). We will review the basic demographic trends in the Valley from mid 19th century until now, as well as the main changes in mobility within the valley and to/from outside the valley. The Tua Valley has changed (and improved) a lot, but still continues to be one of the more backwards portuguese regions, “the interior of the interior” in a peripheral area (Alto Tras os Montes and Douro).

We present the geographic context of the Tua Valley, at county and parish levels, based on data from the first and last population census in Portugal, as well as the potential impact of the railways.

• 119

Railroads in historical context: construction, costs and consequences

O vale do Tua era diferente quando a linha do Tua começou a ser construída, na primeira metade da década de 1880. Faz-se uma revisão das alterações de população, numa base territorial, a partir dos dados do primeiro e do último censo populacional em Portugal. O Vale do Tua mudou, e melhorou, nos últimos cento e cinquenta anos, mas continua a ser ua das regiões mais atrasadas de Portugal, o “interior do interior” de uma zona já por si periférica (Alto Tras os Montes e Douro).

Apresenta-se uma análise territorial das freguesias conforma a proximidade da linha e potencial impacto desta, e a correspondente evolução populacional nestes cento e cinquenta anos.

120 •

Eduardo Beira

TUA VALLEY: How different is it now?

An introduction to population dynamics (1864-2011) Eduardo Beira WHAT IS THE TUA VALLEY? The Tua Valley was different when the decision to build the Tua railways has been made (beginning of 1880s). We will review the basic demographic trends in the Valley from mid 19th century until now, as well as the main changes in centralities within the valley. Tua Valley changed (and improved) a lot, but still continues to be one of the more backwards portuguese regions, “the interior of the interior” in a peripheral area (Alto Tras os Montes and Douro).

The Foz Tua village itself also changed during last two centuries. Railways may not have improved its local and regional importance, but instead may have degraded it. It may have resisted some of the impact of the changes from fluvial mobility of people and goods to railway junctions (Douro and Tua lines), but now the new conditions with railways has been a challenge.

The Tua line was built soon in the overall Portuguese railways development cycle, surprisingly sooner than other more important lines. And was built fast: it took less than four years for the first phase (Mirandela to Foz Tua). It was a line to “nowhere” and to the “unknown”, to the “very far interior beyond the mountains”. It was a surprising political option to reach a “non-explored”, peripheral and remote area of the country.

Different definitions for the Tua Valley can be suggested, based on different physical, social and economic criteria. But for statistical analysis we need to rely

• 121

Railroads in historical context: construction, costs and consequences

on territorial data: districts, NUTs, counties (concelhos) and parishes (freguesias).

We will consider Tua Valley as five counties around the Tua river between Foz Tua village and Mirandela city: Carrazeda de Ansiães, Mirandela, Alijó, Murça and Vila Flor. We will compare 1864 census data (the first official census in Portugal) with the last one available (2011). We will look to see how different the Valley is nowadays, relative to 150 years ago, comparing the two years on each point of this timeline (2011 versus 1864). We will also make a first attempt to deal with the dynamics of population along this period of time, based on Carrazeda county data.

At the upper level of old territorial units (districts), Tua Valley includes territories from both Bragança and Vila Real districts, but in new territorial units (NUTs), it is partly in Alto Tras os Montes (aTM) and partly in Douro (NUTS III regions). Old and new territorial units have different boundaries.

We will try to contextualize these areas in Portugal (Continental, without Atlantic islands). Before that we will discuss the relevant territorial units available in old and new statistics (districts and NUTS).

122 •

FROM TRÁS OS MONTES PROVINCE TO ALTO TRÁS OS MONTES AND DOURO NUTs The literal translation for Trás os Montes is “beyond the mountains”. The mountains are Marão and Alvão, and also a multitude of other hills around these main mountains. The traditional independence and isolation of people in Trás os Montes is well documented by the popular statement Para além do Marão, mandam os que lá estão: “beyond Marão (mountains), those who live there are those in charge”.

The old Trás os Montes and Alto Douro province used to include two districts: Bragança and Vila Real. These territories are now included in two new territorial units (NUTIII): Alto Trás os Montes and Douro. Cities like Bragança, Chaves and Mirandela are in Alto Trás os Montes, the northern region closer to the border with Spain. Douro (NUTIII) is based around the city of Vila Real and extends along Douro river margins until Foz Tua, and further up to the limits of Carrazeda county, but it also includes other large areas on the left margin of the Douro river, that are neither in Bragança nor Vila Real districts. Basically NUT III Douro is the traditional home of most Port Wine vineyards.

Alto Douro is often used, and it translates into “High Douro”. Douro is the river that begins in Spain and crosses Northern Portugal until the Atlantic Ocean, in Porto. Alto Douro has no precise territorial definition nowadays: it

Eduardo Beira

used to mean the territories between the Douro river and the plateau, part of the Vila Real district and part in Bragança district. For instance, Carrazeda de Ansiães county is usually considered “Alto Douro”, although it is also part of Bragança district (not Vila Real district).

Alto Tras os Montes means “the high areas beyond the mountains”, and it includes (among others) the counties of Mirandela, Macedo de Cavaleiros, Bragança and Chaves, the most important cities in this high north eastern region of Portugal, close to the border with Spain.

TUA VALLEY: FIVE COUNTIES, A PUZZLE OF AFFILIATIONS TO DISTRICTS AND NUTs At the county (concelho) level, Tua Valley includes five counties: two in the Vila Real district (Alijó and Murça) and three in the Bragança district (Carrazeda de Ansiães – Carrazeda for short, Mirandela and Vila Flor). These five counties do have a lot of parishes (freguesias): 22 in Carrazeda, 37 in Mirandela, 17 in Vila Flor 17 in Alijó as well, and 9 in Murça. The Tua Valley is usually seen as a “second class” Douro Valley, or at least a poor continuation of this one. There are close relationships – but only up to a certain point. Differences are also important, both at physical, human and economic geography.

Tua Valley runs along territories in both Alto Trás os Montes and in the Douro territorial units, now called NUTs, the territorial units of the EU statistical system. Continental Portugal is the first level NUT, Northern Portugal region is a second level NUT, and both Alto Trás os Montes and Douro are third level NUTs. Tua Valley was part of the so called Trás os Montes and Alto Douro (TMAD), that used to be a “province”, now an abandoned administrative unit. But TMAD, or simply Trás os Montes, is still very popular in media, sometimes creating confusion about the intended territorial meaning.

But NUTs were not the territorial units in Portugal during 1864. Territorial units were “provinces”, as well as districts, counties (where municipal councils, Câmaras Municipais, are based) and parishes, the lower level administrative unit. (Note that religious parishes may be different and its territory may have no relation with the administrative parishes).

The affiliations in this north eastern region can be tricky: • Tua Valley is both in Alto Tras os Montes and Douro NUTIIIs • And it includes counties both in Bragança and Vila Real districts • Alto Tras os Montes NUT III includes only counties in Bragança and Vila Real districts

• 123

Railroads in historical context: construction, costs and consequences

Figure 1 - Tua Valley in context: Portugal and Northern Portugal, Spain and Atlantic Ocean

124 •

Figure 2 - Vila Real and Bragança districts: counties and Tua Valley (Mirandela, Vila Flor, Carrazeda de Ansiães, Alijó and Murça)

DOURO RIVER

Eduardo Beira

Figure 3 - Douro NUTIII: counties.

DOURO RIVER

TUA RIVER

• 125

But Douro NUTIII includes counties beyond Bragança and Vila Real districts (in Viseu and Guarda districts).

• And Vila Real district includes counties that are not in Alto Douro NUT III (two counties do belong to Tamega, another NUTIII, in a different NUTII) See figure 1 for the context of Tua Valley in Portugal, Northern Portugal, and (part of) Spain. Figure 2 shows Vila Real and Bragança districts, and their counties. Tua Valley counties are marked. Figure 3 shows Douro and Alto Trás os Montes NUT III territories and their counties. Also marked the counties that are part of Tua Valley and those counties that are part of the port wine demarcated region.

Table 1 shows the different affiliations of relevant counties in northeast Portugal to districts, NUTIII and Tua Valley.

•

Railroads in historical context: construction, costs and consequences

Table 1: Counties in Northeastern Portugal: affiliation with districts, NUTs and Tua Valley

Alfândega da Fé

Bragança

Guarda

Vila Real

Viseu

Douro

aTM

Tamega

Tua

district

district

district

district

NUTIII

NUTIII

NUT III

Valley

•

•

Alijó

•

Armamar

•

Boticas •

Carrazeda de Ansiães

•

• •

•

•

• •

•

Lamego

•

•

•

•

Mesão Frio

•

•

Miranda do Douro

•

•

Mirandela

•

•

Mogadouro

•

•

Moimenta da Beira

126 •

• •

Chaves

Macedo de Cavaleiros

•

•

•

Bragança

Freixo de Espada à Cinta

•

•

Mondim Basto

•

Montalegre

•

Murça

•

Penedono Peso da Régua

• •

Sabrosa

•

Santa Marta de Penaguião

•

São João da Pesqueira

• • • •

•

Ribeira Pena

• • • • •

Sernancelhe

•

•

•

•

•

•

Tarouca •

•

Valpaços Vila Flor

•

•

•

Vila Nova de Foz Côa

• •

•

Vila Pouca de Aguiar

•

Vila Real

•

• •

Vimioso

•

•

Vinhais

•

•

Source: INE

•

•

Tabuaço Torre de Moncorvo

•

Eduardo Beira

TUA VALLEY: 1864 VERSUS 2011 We will use data at the district, county and parish levels from the first official census in Portugal (1864) and we will compare it with the last available census (2011) in order to have some insight into the long-term demographic changes.

Recent data (2011) was published by NUTs, not by districts. Present district level data needed to be reconstructed from county level data in the relevant NUTs III units.

We compare old data (1864) from Bragança and Vila Real districts with present data (2011). From 1864 to 2011, a period of time close to 150 years, Portugal (Continental) population increased 2.5 times, from 4 to 10 million people, at a compounded annual growth rate of 0.63%. Its population density increased from 43 to 114 persons per km2.

Meanwhile, in the same period, the population of Trás os Montes and Alto Douro (Bragança and Vila Real districts) decreased 4%, from 372 to 323 thousand people. With 12% of the surface of Portugal (Continental), TMAD then had 10% of the national population, but now it has only 3%. Density of population (number of people per km2) decreased only from 42 to 36. This was around 72% of the national average in 1864 (not so far from the national average), but now it is less than 1/3 of the national (continental) average. The demographic importance of the region has decreased very much during the last one and half century.

Table 2 compares the population of Bragança and Vila Real districts (the old Trás os Montes and Alto Douro province). The last two censuses (2001 and 2011) do show a continuous declining trend of the population in these two districts..Bragança district was, and still is, less densely populated than Vila Real district (around half of the density).

One hundred fifty years ago Portugal was a rural country, with 88% of the population living in rural areas outside cities. But Bragança and Vila Real districts were even more rural districts in a rural country: 95% and 98% of their people lived in rural areas outside urban areas.

TUA VALLEY: FIVE COUNTIES The Tua river, from Mirandela until Foz Tua, runs along five different counties (see figure 2 or 3). Rail track goes along the left margin of the Tua river, but only in the territory of Carrazeda, Vila Flor and Mirandela counties. Murça and Alijó are two counties included in the Tua Valley, because they have territories in the

• 127

Railroads in historical context: construction, costs and consequences

Table 2: Population of Portugal and of Braganรงa and Vila Real districts Portugal Portugal C Braganรงa aTM Surface

2011

Km2

88156

2011 2011

Vila Real

Braganรงa + Vila Real

Douro

aTM+Douro

5231

3688

8919

58.6%

41.4%

100.0%

100%

5.9%

4.2%

10%

Population 1864

mil

4188

3829

159

213

372

(resident)

1864

%

109%

100%

1864

%

5.6%

10%

57.3%

100%

2001

10356

9869

149

200

349

2011

10556

10041

136

187

323

105%

100%

2011 2011

% 2011/1864

128 โ ข

cgr % hab/km2

4.2% 42.7%

2.52

2.62

0.63%

0.66%

1.4%

1.9%

3.2%

42.1%

57.9%

100.0%

0.86

0.88

0.87

-0.11% -0.09%

-0.10%

1864

43

26

49

42

1864

100%

60%

114%

97%

114

26

51

36

114

26

51

36 32%

2009

115.4

2009 2009

100%

23%

45%

1864

%

88

95

98

% female 1864

%

52%

49%

51%

52.2%

52.2%

51.9%

% rural

2011 Source: INE

right margin of the river, although no rail track. They only have territories in the right margin of the Tua river.

In 1864, the population share of those five counties was 17% of Trรกs os Montes and Alto Douro population (and 1.6% of the continental population in the country).

Eduardo Beira

Figure 4 - Population of the five counties in Tua Valley and their central parishes, 1864 and 2011.

For each county, population of the central parish and other parishes, and population difference between 1864 and 2011.

If change is negative, it appears in the left side. If change is positive, it appears in the right side.

1864 at the right side; 2011 at the left side; values for 2011, transformed to negative values (left side of axis)

• 129

In 2011, they still represent 17% of Bragança and Vila Real districts (old Trás os Montes and Alto Douro province), but only 0.55% of the national (continental) population. These five counties had around 62 thousand people in 1864, but less than 55 thousand in 2011. The five counties in the valley are less populated today than 150 years ago – in a country that meanwhile had increased its population 2.5 times.

Meanwhile, the structure of the population by county has also changed: Mirandela county share increased from 28% to 41%, Carrazeda county decreased from 18% to 12%, and all the other counties shares have also declined.

Tua Valley counties are from two different districts: Alijó and Murça belong to the Vila Real district, while Mirandela, Vila Flor and Carrazeda belong to the Bragança district. In 1864, 23% of the Bragança district population and 12% of Vila Real lived in the Tua Valley (defined as the five counties). But in 2011, these percentages were 27% and 10%. The importance of Tua Valley counties has slightly increased in the Bragança district, and slightly decreased in the Vila Real district.

Railroads in historical context: construction, costs and consequences

Figure 5 - Top ten parishes of the five counties in Tua Valley in 1864, and their population in 1864 and 2011.

130 •

TUA VALLEY: AN ANALYSIS BASED ON PARISHES An analysis based on the lower level of territorial units allows a more precise definition of the valley population and its territorial dynamics.

Main towns in each county of the Tua Valley are based on parishes with their own name (we call them “central parishes”) where the main urban centres are based. The population share of those low level units (parishes) was 9% in 1864, indicating a very dispersed population and a very low level of urbanization. But in 2011 its share was more than 37%, suggesting a strong redistribution of population in the territory, associated with stronger urbanization.

Of all the five counties of the Tua Valley, only Mirandela county now has a population larger than 150 years ago. All the other central parishes in each one of the five counties have increased in population, although increases in Mirandela and Carrazeda central parishes have been specially important (5.3 and 6.7 times). Overall, the total population of the five counties in the Tua Valley decreased more than 8 thousand people, but the population of Mirandela and Carrazeda central parishes has increased more than 13 thousand people – so the population of all the other parishes in the five counties has decreased by more than 21 thousand. See figures 4 and 5.

We can see a persistent pattern of concentration of population in the main

Eduardo Beira

Table 3: Population of the five counties in Tua Valley by type of parish, 1864 and 2011 A

A

B

1864

2011

1864

Carrazeda

1821

748

Mirandela

3349

13328

Vila Flor

2562

1302

C

C

Z

2011 1864

B

2011

1864

3359 1392

Z Total Total 2011

1864

2010

6013

4182 11193 6322

1437

737 2507 2830 10060

6131 17353 23026

Alijó

3336

797

15560

Murça

1929 1340

3970

1390

822

3529

2302

7481 4426

9761 18896 10558 4614

5899 5954

Grand Total

7732

15378

6702

2874 7256 5044 39132 26990 60822 50286

Bragança

7732

15378

1437

737 7256 5044 19602 12615 36027 33774

Vila Real

5265

2137

19530 14375 24795 16512

• 131

towns of each county (central parishes), and Mirandela city emerges as the big centrality in the region. Carrazeda central parish / town also emerged to a new centrality, from a very secondary situation in 1864, although total population of Carrazeda at the county level has decreased.

To study the change in population of each parish, in each of the five counties from 1864 to 2011, each parish was classified into four categories: A – Tua railways crosses the parish (left margin of the river) B – Parish not crossed by the rail line, but adjacent to it, on the other side of Tua river (right margin) C – Parish immediately adjacent to a B type parish, adjacent to one parish crossed by the rail line Z – All other parishes in the five counties of Tua Valley (not crossed nor adjacent to the rail line).

A more precise definition of the Tua Valley would include only A, B and C categories. Z parishes are too far from the line to receive a significant impact of the line. These are either “Douro river slope” parishes, or they are located in the top of the plateau that separates Tua and Douro valleys. Results are summarized in table 3. For the location of parishes of each category, see figure 6.

Railroads in historical context: construction, costs and consequences

Figure 6 - Tua Valley: A, B, C and Z parishes in the five counties

A B C Z

Mirandela

Murça

132 • Alijó

Vila Flor

Carrazeda de Ansiães 0

10

20 Km

We will consider A+B+C parishes as a more restrictive and realistic definition of Tua Valley as area of direct influence of Tua railways: a population of 21.5 thousand people in 1864 versus 23.3 thousand people in 2011. This makes Tua Valley a tinier, less populated region than suggested by the total population of the five counties. See figures 7 and 8.

So the “non Tua river” driven areas in the five counties of the classical definition of the Tua Valley corresponds largely to the Z type parishes: its population

Eduardo Beira

Figure 7 - Structure of the population of each category of parishes in the five counties (Tua Valley), 1864 and 2011

• 133

has decreased from 37 thousand in 1864, to 27 thousand people in 2011, a 27% relative drop. They represented 63% and 54% of the five counties population, respectively. This means that around 2/3 of the population of the five counties were too far from the reach of Tua rail line when it was built. Nowadays more than half of the five counties population is outside the direct influence of the line.

Mirandela city / parish growth is a major change in the demographics of Tua Valley. See figure 7 and 8. Population in category A parishes were the better served by Tua railways: this meant only 7700 people (13% of the population of the five counties) in 1864, versus 15400 (34%) in 2011. This may suggest a positive relationship with the line. But if we split the A type parishes between Mirandela parish and all the other parishes, the effect of Mirandela centrality and growth becomes clear, with a 5.3 times growth during 150 years. It is the opposite of all the other category A parishes, where it dropped 41% in the same period. Mirandela central parish was a town with less than two thousand people in 1864, only 3% of the five counties population. Now around a quarter of the population of the five counties in Tua Valley lives in the Mirandela parish.

Railroads in historical context: construction, costs and consequences

Figure 8 - Population of the four categories of parishes in the five counties of Tua Valley, 1864 and 2010 (top), and change in population (%) from 1864 to 2010 (bottom), including Mirandela central parish

A – Tua railways crosses the parish (left margin of the river) B – Parish not crossed by the rail line, but adjacent to it, in the other side of Tua river (right margin)

134 •

C- Parish immediately adjacent to a B one, a parish adjacent to one crossed by the rail line Z- All other parishes

THE FATE OF IMPORTANT TOWNS IN TUA VALLEY In 1864, Mirandela was a “non Douro” town with a population of the same order of magnitude of Alijó and Murça, both Douro Valley driven towns with terri-

Eduardo Beira

tory in the Tua Valley. One hundred fifty years later Mirandela has become an important city, with more than ten thousand people. And today Alijó and Murça still remain towns, a bit larger than 150 years ago, but still small, with population between two and three thousand people (figure 5).

Carrazeda town (as measured by Cararazeda central parish) also shows an impressive change during last the 150 years: its population grew 6.7 times (against 5.3 for Mirandela and 0.83 for the Valley). It had less than 300 people in 1864, then much smaller than Mirandela, Alijó and Murça. Today it still may be smaller than all of them, but it has around 1700 people, a significant “catch-up” (figure 9).

In 1864, Carrazeda parish had only 2.4% of the population of Carrazeda county. Today it has around 27%. Carrazeda county is a large territory with 21 parishes. The top five most populated parishes in Carrazeda county had more than one thousand people each in 1864 – and now all of them only have around 500 people, and all of them are geographically distant from Carrazeda town.

This means there is a long-term trend of concentration in Carrazeda, as the new local centrality, and a drop in the importance of all the other “important” towns / villages of 1864, both in Douro as well as the Tua slopes of Carrazeda county.

Figure 9 - Population of the parishes of Carrazeda de Ansiães county in 1864 and 2010, sorted by crescent order in 1864.

• 135

Railroads in historical context: construction, costs and consequences

Figure 10 - Trends in population of Carrazeda de Ansiães county and central parish, 1734 to 2010

136 •

Figure 11 - Trends in the population of some important parishes of Carrazeda de Ansiães county, 1864 to 2011 (based on data for years 1864, 1940, 2001 and 2010)

Eduardo Beira

LONG-TERM POPULATION DYNAMICS Of course, population trends from 1884 to 2011 were not linear. Long historical series for the population of parishes and counties are needed for a more complete analysis. But the case of Carrazeda, county and central parish, it can help to illustrate what has happened during last century.

Figure 10 plots data collected by a local author (Tavares, 1999) from all available censuses and other sources, also including previous information relative to 1734 and 1797. At the county level, it shows growth until the first years of the 20th century, with a drop and subsequent recovery until the ’60s, followed by a declining trend that still continues (as shown by 2011 versus 2001 census data).

At the parish level, we see a continuous positive trend in Carrazeda, even stronger after the ‘40s. Figure 11 shows the trends in Carrazeda central parish and four other important parishes (Castanheiro, Linhares, Seixo, Vilarinho) based on data for the years 1864, 1904, 2001 and 2011. These four units were chosen because they were the most populated parishes of Carrazeda county in 1940, and all of them then had more than one thousand people, at the time similar to Carrazeda central parish population, or larger (the fate of Vilarinho and Linhares parishes is indeed extraordinary: in 1940 both were 50% larger than Carrazeda, but now they are much smaller, less than 25%).

All of them show the typical upward trend until the ‘60s, followed by a downward trend – in contrast with the very atypical trend of Carrazeda central parish.

TUA RAILWAYS: FIRST YEARS OF OPERATION. Five years after starting operations, the Tua line was basically a channel for the flowing of goods (cargo) between the Douro line and Mirandela. Revenue from passengers was only 28% of total revenue and more than half was generated by the extremes of the line (junction with Douro line in Foz Tua and Mirandela).

Revenue from passengers within the Tua line (excluding Foz Tua and Mirandela) contributed less than 13% of total revenue, and revenue from cargo was only 12%. See figure 12.

Based on local traffic within the Valley, the line would never be economically feasible. Of course, this does not mean that the line was not very important for people living in the Valley – especially in category A parishes.

• 137

Railroads in historical context: construction, costs and consequences

But Mirandela and Foz Tua (Douro line) extremes contributed with an astonishing 81% of the total revenue of the line. It was a fast lane channel for traffic between Mirandela (and later Bragança) and the rest of the world – basically Regua and Porto, and from Porto, to Lisbon and the world.

Figure 12 - Structure of revenue in the Tua line (1892): first circle (interior) is passengers, middle circle is cargo and external circle is total revenue.

Source: AEP, INE

138 •

Eduardo Beira

REFERENCES Tavares, V., Conheça a nossa terra Carrazeda de Ansiães, Author edition, Carrazeda de Ansiães,1999

SOURCES OF DATA INE, Census, 1864 INE, Census, 2010 INE, Anuário Estatistico de Portugal, 1892

• 139

Maria Manuel Oliveira

NARRATIVE(S) ON THE CONSTRUCTION OF THE LANDSCAPE: THE TUA VALLEY MEMORY CENTER NARRATIVA(S) SOBRE A CONSTRUÇÃO DA PAISAGEM: O NÚCLEO DA MEMÓRIA DO VALE DO TUA Maria Manuel Oliveira (Centro de Estudos da Escola de Arquitectura, U. Minho, Portugal) Maria Manuel Oliveira, architect from the School of Fine Arts of Oporto, is a professor at the School of Architecture of Minho University, where she develops architectural practices in the framework of the Studies Center, of which she is part of the direction.

Maria Manuel Oliveira, arquitecta pela Escola Superior de Belas-Artes do Porto é docente na Escola de Arquitectura da Universidade do Minho, onde desenvolve prática arquitectónica no âmbito do seu Centro de Estudos, cuja direcção integra.

Abstract Resumo

The images presented here summarize the research by Centro de Estudos da Escola de Arquitectura, University of Minho, that lead to the proposed location of the Tua Valley Memory Center: if the scale model expresses the sublime landscape condition of the territory, then the drawings register a context where a valuable natural and constructed heritage and a socioeconomic fabric intertwine with expressive vitality and an urban condition fit to gather new experiences. It is believed that the Nucleus of Memory of the Tua Valley can accomplish a central role in the activation of this – now latent – synergy.

As imagens que aqui se apresentam sintetizam a pesquisa feita pelo Centro de Estudos da Escola de Arquitectura da Universidade do Minho e que fundamentou a proposta de localização do Núcleo de Memória do Vale do Tua: se a maqueta exprime a sublime condição paisagística do território, os desenhos registam um contexto em que a um valioso património natural e edificado se associam um tecido socio-económico com expressiva vitalidade e uma condição urbana apta a acolher novos estímulos. O Núcleo de Memória do Vale do Tua poderá, acredita-se, cumprir um papel central na activação desta - agora latente - sinergia.

• 141

Railroads in historical context: construction, costs and consequences

142 •

Maria Manuel Oliveira

Narrative(s) on the construction of the landscape: the Tua Valley Memory Center Maria Manuel Oliveira – CE.EAUM

The Tua Valley Memory Center will record the history of the Tua Valley – a stunning scenery, particularly in terms of its orographic condition – and its human settlement. It is a narrative aimed at the general public, in particular, the student population. It also aims to stimulate the local collective memory, raising their consciousness of an identity founded in the longue durée and already stated in the rock carvings near the mouth of the river. In the past few centuries, this territory has seen three major waves of human intervention that profoundly molded the landscape and determined its habitants’ way of life: the topographic transformation that accompanied, and still accompanies, the establishment of hillside vineyards since the eighteenth century, followed by the construction of railways at the end of the nineteenth century, and the completion of the dam in the twentyfirst century.

As a pretext to investigate and systematize territorial and physical nature, ethno-anthropological, artistic, and technological themes, the Tua Valley Memory Center’s main purpose is to establish itself as a device that not only allows, but encourages synchronic or diachronic, more specified or general cross-readings, drawing the attention of a wide audience with various interests.

The Tua Valley Memory Center also hopes to fulfill a modern function in the history of the Tua Valley. It hopes to contribute to the articulation of a series of urban, heritage, and landscape references, that were until now separate, and to confer the critical mass necessary for the activation of the comprehension of the

• 143

Railroads in historical context: construction, costs and consequences

144 •

territory that is latent by part of the population and the visitors. Furthermore, it is intended that the building itself become indelibly rooted in the biography of the area, constituting itself as a monument, (re)adapted to the present circumstance.

The proposal to accommodate the exposition near the goods warehouse of the Foz Tua station, intrinsic to the history of the railway and a crucial part of the local urban area, is therefore reinforced and requalified.

The Tua Valley Memory Center will hopefully be the heart of an extensive and open network of multiple endeavors in recognition of the local identity and incentive new interpretations of a territory in the making.

Maria Manuel Oliveira

• 145

Railroads in historical context: construction, costs and consequences

146 •

Maria Manuel Oliveira

• 147

PART 2 Economic impact of railroads: methods and international comparisons O impacto económico do caminho de ferro: metodologias e comparaçþes internacionais

Jeremy Atack • Robert A. Margo

LAND OWNERSHIP AND THE COMING OF THE RAILROAD TO THE AMERICAN MIDWEST, 1850-60 PROPRIEDADE DA TERRA E O IMPACTO DA CHEGADA DO CAMINHO DE FERRO AO MIDWEST AMERICANO, 1850-60 Jeremy Atack (U. Vanderbilt, USA EUA) Robert Margo (U. Boston, USA EUA) The authors are, respectively, Professor of Economics and History, Vanderbilt University, and Research Associate, National Bureau of Economic Research; Professor of Economics, Boston University, and Research Associate, National Bureau of Economic Research Jeremy Atack é Professor de Economia e História, Universidade de Vanderbilt (EUA) e investigador associado, NBER National Bureau of Economic Research (EUA).

Robert Mago é Professor de Economia, Universidade de Boston, e investigador associado, NBER National Bureau of Economic Research (EUA).

Abstract Resumo

A central question of nineteenth century American economic history which has preoccupied scholars as diverse as Simon Kuznets, Peter Lindert and Jeffrey Williamson is whether economic development was accompanied by increasing inequality.

In this paper we study a particular indicator of economic inequality – the incidence of land ownership among adult men – and its relationship to a putative causal factor in nineteenth American growth – the so-called Transportation Revolution, focusing specifically on the diffusion of the railroad.

We will revise the earlier draft and also make use of a newer version of the transportation database that contains distance-adjusted measures of rail access – for example, the share of a county’s land area that is within a specified distance of a railroad. Preliminary analysis indicates that the new measures are better indicators of rail access than the dummy variable indicator used in our earlier work.

Uma questão fundamental da história económica do século XIX, que tem preocupado investigadores tão diversos como Simon Kuznets, Peter Lindert e Jeffrey

• 151

Railroads in historical context: construction, costs and consequences

Williamson, é se o desenvolvimento económico foi acompanhado por uma desigualdade crescente. Neste trabalho estuda-se um indicador de desigualdade – a incidência da propriedade da terra entre homens adultos – e a sua relação putativa com um factor causa do crescimento americano no século dezanove – a chamada revolução dos transportes, com ênfase sobre a questão especifica da difusão dos caminhos de ferro.

Faz-se uma revisão de uma análise anterior e usa-se uma nova versão de uma base de dados sobre transportes que usa medidas de distancias ajustadas pela facilidade de acesso ao caminho de ferro - por exemplo, a fração de terrenos de um concelho que se encontram a uma certa distancia da linha de caminho de ferro.

Os resultados obtidos indicam que as novas medidas são melhores indicadores do acesso ao caminho de ferro do que os indicadores anteriormente usados.

152 •

Jeremy Atack • Robert A. Margo

Land Ownership and the Coming of the Railroad to the American Midwest, 1850-60 Jeremy Atack, Robert A. Margo

This is a revised version of a paper given at the conference on “Railroads in Historical Perspective”, Foz Tua, Portugal, October 2011. An earlier version was also given at the “New Perspectives on Wage and Wages” conference in Hackenberga, Sweden, October 2010. Comments from Stan Engerman, Tim Hatton, and participants at the Foz Tua and Hackenberga conferences are gratefully acknowledged.

A central question of nineteenth century American economic history which has preoccupied scholars as diverse as Simon Kuznets, Peter Lindert and Jeffrey Williamson is whether economic development was accompanied by increasing inequality. In this paper we study a particular indicator of economic inequality – the incidence of land ownership among adult men – and its relationship to a putative causal factor in nineteenth American growth – the so-called “Transportation Revolution”, focusing specifically on the diffusion of the railroad.1 Land in nineteenth century America was the single most important source of wealth.

In the agriculturally-based economy of nineteenth century America, access to land was a critical determinant of economic and social progress for most of the population, particularly in the Midwest where the farm family provided the bulk of the farm labor. Moreover, the ownership of that land in one form or another was the quintessential “American Dream”, the nation’s principal metaphor of upward mobility. In an economy dominated by agriculture such as the nineteenth century United States, owner-operator status, the final step on the agricultural ladder, was central to the wealth accumulation process and a variety of institutions, most notably the various nineteenth century land acts, were 1 Holding constant the distribution among those owning land, conventional measures of inequality, such as the Gini coefficient or the Theil index, are decreasing in the share of owners.

• 153

Railroads in historical context: construction, costs and consequences

154 •

structured in such a way as to facilitate the transfer of public land into a wide range of private hands. Even for non-farm households, land was an important store of wealth due to the limited variety of other types of physical and financial assets. The ease or difficulty of acquiring land may also have affected the course of development. The early onset of fertility decline in the nineteenth century, for example, has been linked to variations across space and time in the ease of acquiring land. According to Stanley Engerman and Kenneth Sokoloff, widespread land ownership facilitated a relatively equal distribution of political power, itself associated with institutions (for example, the franchise) that may have been conducive to long run growth.

Nevertheless, published census statistics on tenancy collected from 1880 onwards showed relatively high and rising tenancy rates in the Midwest where land had been settled most recently and under a public policy which increasing emphasized ease of access to this resource (Goldenweiser and Truesdell 1924).

Several scholars have also estimated tenancy rates at earlier dates (see, for example, (Bogue 1963; Cogswell 1975; Winters 1977; Atack and Bateman 1987; Atack 1988, 1989)) by comparing individual farm values with the value of real estate reported by the farm’s operator. Unfortunately, these data are available for relatively few locations and only back to 1850—the earliest date at which the census reported the value of real estate that individuals held. They are, however, informative. Tenancy rates rise over time throughout the period for which we have data. In this paper, we look at the real estate wealth that a random sample of individuals held by county and its relation to the pattern of railroad expansion..

Elsewhere, we have reported on how the coming of the railroad influenced land improvement and raised land values. Each of these different forces then plays a role in rising tenancy and reduced upward economic mobility.

Economic historians have long acknowledged that the transportation revolution was an important causal factor in economic growth in nineteenth century America. Although the revolution took many modes, arguably the most visible was the “Iron Horse” – the railroad. Previous work has suggested that the railroad lowered the cost of moving goods, information and people from one location to another, thereby positively affecting growth in a number of ways – for example, by encouraging greater division of labor in manufacturing (Atack, Margo, and Haines, 2011a); by increasing the pace of urbanization which may have raised aggregate efficiency through agglomeration effects (Atack, Bateman, Margo and Haines, 2010); and by encouraging improvements in agricultural land, thereby bringing new resources into production (Atack and Margo, 2011b).

While the coming of the railroad promoted economic development, we show

Jeremy Atack • Robert A. Margo

that it also appears to have reduced the proportion of adult men owning land.

The negative effect is strongest among men between the ages of 20 and 39 – the prime ages at which the gradient of ownership and age was steepest – particularly among those in agriculture (Cogswell 1975; Wright 1988; Atack 1989).

We also provide evidence, albeit tentative, on a plausible causal mechanism for this finding through the impact on the railroad on farm size. In those counties that acquired rail access in the 1850s, average farm size was higher in 1860 suggesting that “minimum efficient scale” in agriculture increased along with the improved trade opportunities created by the railroad. Larger farms were more expensive; therefore it took longer in these areas to move up the agricultural ladder after the railroad came.

Our analysis of the relationship between land ownership and rail access draws on a new body of data documenting the diffusion of railroads and other transportation media in nineteenth century America derived from digitized historical maps using GIS software (Atack, Bateman, Haines, and Margo 2010).

The primary database provides information on transportation access at the county level at census year intervals. For the purposes of this paper we have linked the transportation database to extracts of adult men, ages 20 to 59, from the 1850 and 1860 IPUMS samples. The IPUMS provides individual level information on ownership of real estate, occupation, and various demographic variables, which are used in our econometric analysis.

The analysis in the paper focuses on the Midwest in the 1850s. We restrict our attention to the Midwest for two reasons. First, it was in the Midwest where the bulk of the growth in the rail network occurred in the 1850s making the effects easier to identify. Second, we have developed an identification strategy that is particularly plausible and appropriate for the Midwest in this period (see below). However, analysis of the entire region is complicated by the fact that many county boundaries were redrawn during the period while other counties did not even exist until the 1840s. To minimize these complications, we have restricted our analysis to 278 counties that were in existence by 1840 and which did not change their boundaries during the 1850s.

Our primary measure of rail access is a dummy variable indicating the presence of a rail line within a county. All of the counties in our sample lacked rail access in this sense in 1850 but some would gain access by 1860. Our primary econometric approach is difference-in-differences analysis which compares the change in outcomes between 1850 and 1860 in a set of “treatment” counties – those that gain rail access between 1850 and 1860 – versus a set of “control” counties – those that did not have rail access (as we have defined it) before the Civil War. Other measures of access (such as miles of railroad in the county or

• 155

Railroads in historical context: construction, costs and consequences

156 •

proximity of the centroid of the county to a rail line) yield similar results.

Using this approach we find that rail access reduced the likelihood of land ownership. The effect is small (and insignificant) at the sample means but is much larger (and significant) when we control for individual characteristics and various county level factors associated with the coming of the railroad. It is especially strong among men ages 20-39 who worked in agriculture – controlling for various factors, such men were about 7 percentage points less likely to own real estate in the treatment counties compared with their counterparts in the control counties.

To explore the causal mechanism behind our result we use county level data from the 1860 census on the distribution of farm sizes. Conditional on various factors, counties with rail access in 1860 had relatively fewer small farms than counties in the control group. The negative effect of rail access on farm size is consistent with the hypothesis that were some economies of scale in northern agriculture before the Civil War, and that improved access to trade led to a rise in the “minimum efficient scale” in farming. Larger farms were more costly to acquire than smaller farms and, we surmise, it therefore took longer to move up the agricultural ladder.

We supplement our basic analysis with robustness checks and additional analyses. The principal robustness check addresses a key assumption in our DID analysis that, conditional on various observable determinants, the coming of the railroad to a county was a random event. If this assumption is false then the estimated treatment effects of the railroad would be biased. In particular we are unable to control for landownership in 1840 and our analysis of the farm size data is cross-sectional (because such data were not published in 1850). It is easy to imagine, therefore, that our results are contaminated by omitted variable bias.

To address this concern we re-analyze the 1860 data using an instrument variable (IV) derived from various federal government transportation surveys that were conducted in the 1820s and early 1830s—that is to say, right at the start of the railroad age and well in advance of when we observe the possible effects of the railroad on land ownership. Our IV estimates are significantly larger in absolute value suggesting that if anything, we are understating the impact of the railroad.

THE TRANSPORTATION REVOLUTION The United States experienced a “transportation revolution” in the nineteenth century (Taylor 1951; Goodrich 1961; Haites, Mak et al. 1975). The elements

Jeremy Atack • Robert A. Margo

of the revolution are well known and include dredging and other improvements to harbors and natural inland waterways, improved all-weather roads for travel by wagon, the building of canals, the marine application of steam, and last, and perhaps of greatest importance, the diffusion of the steam railroad. In this paper, we focus on the railroads rather than other elements of the transportation revolution because, as a practical matter, landownership data from the federal census is available only from 1850 onward. Consequently, landownership information only overlaps to any significant extent with the diffusion of the railroad. For example, the bulk of canal building had been completed by the early 1840s and the western river steamboat and Great Lakes shipping were also fully in place before 1850. Even the steam railroad was well established along the East Coast before 1850. However, the American Middle West was virtually untouched by rail until the late 1840s and the big expansion into the region took place in the 1850s.

American steam railroads followed fast on the heels of the British innovation by Robert Stephenson’s Stockton and Darlington Railway in 1826, emerging first as a competitive response by Atlantic port cities to the growth of the port of New York following the opening of the Erie Canal (Condit 1980). Thus, for example, by 1827, Baltimore had chartered the Baltimore and Ohio Railroad (Stover 1987) and South Carolina had chartered a line between Charleston and Hamburg, just across the Savannah River from Augusta, Georgia (Phillips 1908). Other cities such as Boston soon followed (Poor 1970). By 1840 some 2,800 miles of track were in operation (Carter, Gartner et al. 2006 Series Df874), most of it in New England, the mid-Atlantic, and South Atlantic states, and almost all of it covering short distances with minimal interconnections. Further expansion in mileage took place in the 1840s, much of it again in New England, and the Middle Atlantic states. By 1850, about 9,000 miles of railroad track were in operation (Carter, Gartner et al. 2006 Series Df874).

The 1850s witnessed a substantial wave of rail expansion (Stover 1978).

Railroad mileage in operation more than tripled, much of it being built in the Midwest. Indeed, between 1853 and 1856, more than half the track miles built in the U.S. were built in the Midwest and, in 1856, this share reached a remarkable 75 percent, of which almost 40 percent was built in just one state, Illinois (Fishlow 1965; Carter, Gartner et al. 2006 Series Df882). This railroad network was, however, less extensive than the number of miles might indicate as it was split between at least seven different gauges (see (Taylor and Neu 1956; Puffert 1991)), several of which were incompatible with one another (especially between North and South) thus preventing through-haulage. Most of the track in the Midwest was, however, standard gauge (4’ 8 ½”) while that in Ohio was somewhat wider (4’ 10” “Ohio” gauge) but still compatible with the standard

• 157

Railroads in historical context: construction, costs and consequences

158 •

gauge. Outside of Missouri, only one line, the Ohio and Mississippi Railroad which operated track on a 72” gauge from Cincinnati to St. Louis was incompatible, with locomotives and rolling stock operating in the rest of the region.

In Missouri, the railroads were divided almost 50:50 between standard gauge in the northern part of the state and a 66” gauge in the southern parts that was incompatible with gauges elsewhere (Taylor and Neu 1956).

The first railroad to operate in the Midwest was the Erie and Kalamazoo Railroad, chartered in 1833, which began steam operation in 1837 between from Port Lawrence, Michigan (now named Toledo, Ohio) at the western end of Lake Erie to a point on the Kalamazoo River which would then (theoretically) provide a link across the state to Lake Michigan (Dunbar 1969; Meints 1992). Shortly thereafter the Mad River and Lake Erie railroad began operation from Sandusky on the shores of Lake Erie in Ohio southwards towards Springfield and Dayton.

Expansion of the Midwestern railroad network, however, proceeded slowly until the late 1840s when the pace of construction began to pick up before exploding in the 1850s. Our focus on the Midwest here is explained in part by this dramatic surge in railroad construction in the region precisely at a time when we have more economic, social and demographic information available but also because, as we explain below, we have particularly detailed information about railroad construction in this region during this same period.

Economic historians have devoted considerable attention to measuring the impact of railroads on total output, or what Fogel ((1964); see also (Fishlow 1965; Williamson 1974; Kahn 1988)) called the “social savings” of the railroad.

Attention has also been paid to the “backward linkages” that railroad expansion created with other manufacturing sectors (Fogel 1964); whether railroads were “built ahead of demand” (Fishlow 1965); and the impact of transportation improvements on regional economic development (Williamson 1974). Our goal in this paper is to add to this literature by examining another “treatment effect” of the railroad, that on land ownership. To measure this treatment effect requires a data set linking the diffusion of railroads to data on wealth. After surveying the literature on nineteenth century wealth accumulation, we describe our data set and estimation strategy.

WEALTH ACCUMULATION IN NINETEENTH CENTURY AMERICA Beginning with the eighth census in 1850, the federal government began collecting information on the value of real estate held by individuals. In 1860, these

Jeremy Atack • Robert A. Margo

data were supplemented by questions concerning the value of personal estate in excess of $100 and the questions were repeated in 1870 before being dropped from the census in 1880. Here, we make use of the data on real estate ownership from 1850 and 1860 by individuals who are included in the 1 percent IPUMS sample (http://usa.ipums.org/usa/). These have then been linked to our transportation database.

Economic historians have intensively examined the antebellum wealth data (Pope 1986). With few exceptions, the focus of their work has been on establishing and measuring the correlations between wealth and various individual or household-level characteristics and examining inequality. In contrast, our primary interest in this paper involves the impact of a factor external to households – the railroad – on the incidence of land ownership.

There is no question that land was the principal asset in nineteenth century America. For example, the 1860 census reported the assessed value of real estate at almost $7 billion compared with somewhat more than $5.1 billion for personal estate, a figure which also included the value of slaves, and less than $500 million in bank notes and deposits (United States. Census Office. and Kennedy 1862 193-4) Land was both a store of value and a productive factor. As a savings vehicle individuals would invest over the life cycle as a way of smoothing consumption and to provide consumption in old age (Atack and Bateman 1981).

For farmers ownership of land was closely tied to moving up the agricultural “ladder”. Absent any inheritance individuals started on the lower rung of the ladder – wage labor – moving up to some type of tenancy followed by owneroperator status (Atack 1989). For both non-farmers and farmers, therefore, land ownership was strongly related to age. The likelihood of ownership rose with age before reaching a peak. In the data analyzed in this paper, for example, ownership was a steep function of age between age 20 and 39 (see the next section).2 In this paper we concentrate primarily on establishing an empirical link between the railroad and landownership as there are many possible causal pathways and our ability to distinguish among them is very limited. For example, land speculation certainly was common in the antebellum Midwest. From our previous work, we know that the coming of the railroad raised land values and also increased the incentive to improve land (Atack and Margo 2011b). It is possible that once an area became connected to the wider world through the railroad, speculative land acquisition increased. In our data this could register as a decrease in landownership if speculators were, on average, non-residents because, by construction, our data pertain to the real estate holdings of individu2 Previous work on nineteenth century wealth accumulation also reveals impacts of ethnicity, race, and literacy status. In our primary analyses (see below) we control for race and literacy status as these are pre-determined before the “treatment” of the railroad. We do not control for place of birth initially (e.g. ethnicity) as this arguably is endogenous – the coming of the railroad may affect the likelihood of in-migration. As we discuss later in the paper when we control for place of birth the treatment effect of the railroad is diminished slightly.

• 159

Railroads in historical context: construction, costs and consequences

160 •

als who were resident in the county at the time of the census.

That said, the primary productive use for land at the time was in agriculture and the vast majority of farms were “one farm – one owner” – that is, owneroperated. But not everyone who worked in agriculture was a farm owner – owner-operator status was the final rung, as noted above, on the agriculture ladder.

One way in which the coming of the railroad might have slowed progress up the agricultural ladder, and thus lowered the incidence of landownership, was by affecting the optimal size of land holdings. Ever since Adam Smith, economists have known that improvements in transportation can allow business enterprises to take advantage of economies of scale. Indeed, recent work with our transportation database has shown that, in the case of manufacturing, the coming of the railroad led to an increase in the share of establishments that were “factories’ – that is, firms which employed a sufficiently large number of workers that division of labor in their tasks was likely.

Although the prospects of economies of scale in northern agriculture before the Civil War may have been limited, it does not follow that there were none to be exploited. Certainly a wide range of farm sizes were observed – the census in 1860, for example, reported on farms as large as 500 -1000 acres, even in the Midwest (Anderson 1929; Gates 1945; United States. Census Office. 1990). To be sure, these figures include unimproved acreage as well as land in active production but our point is simply that some farms were large enough to routinely required hired labor alongside that provided by the farm household. Moreover, there may have been important scale effects in agricultural mechanization, which was also taking place at this time and stimulated in part by the increased ease of access to wider markets.

As we document later in the paper, counties that acquired rail access in the 1850s had relatively fewer small farms in 1860 than counties without rail access. If these farms remained under individual ownership then, in effect, some land consolidation must have taken place. A larger farm would be more productive (because of economies of scale) but also more costly. Given the imperfections of the mortgage market at the time, it is plausible that individuals would take longer therefore, to move up the agricultural ladder.

THE DATA We use our newly constructed geographic information system (GIS) database on the Midwestern transportation infrastructure in the mid-nineteenth century

Jeremy Atack • Robert A. Margo

United States which we have linked to an updated version of the well-known ICPSR county level census database. Our GIS-based methodology has substantial advantages of accuracy and consistency over earlier approaches which have generally involved matching historical transportation maps to county boundary maps by hand (and eye).3 Our Midwestern railroad transportation database covers seven states: Illinois, Indiana, Iowa, Michigan, Missouri, Ohio and Wisconsin. Data for five of these states (Indiana, Illinois, Michigan, Ohio and Wisconsin) are based upon a series of maps and a data appendix prepared over ninety years ago at the University of Wisconsin by Professor Frederick L. Paxson (1914) and his students.

They were able to use extant contemporary travel guides published between 1848 and 1860 as their principal sources of information.4 Since then, unfortunately, many of these travel guides which were cheaply printed ephemera seem to have been lost or disintegrated. This is unfortunate as there is little doubt that they are, or rather would have been, the single most valuable resource for our research. These guides served the needs of the traveling public by providing upto-date route maps and timetables. Consequently, they were published at fairly frequent and regular intervals by a number of different companies and, presumably, competition among them should have guaranteed that only the best (that is, the most accurate, complete and useful) would have continued publication for an extended period of time.5 The data for Iowa and Missouri are taken from digitized contemporary state maps from the on-line David Rumsey Map Collection (http://www.davidrumsey.com/) for 1856 (Colton 1856a; Colton 1856b), 1857 (Colton 1857) and 1859 (Mitchell 1859b, a) supplemented by national maps from the Library of Congress collection for 1858 (Sage 1858) and 1860 (Colton 1860). There were no railroads in either of these states prior to 1856.

3 See, for example, (Craig, Palmquist et al. 1998) who visually compared historical maps to county boundaries (which generally did not appear on the historical maps). For a discussion of the handmatching procedure used by Craig and the pitfalls that can arise, see footnote #2 of (Atack, Bateman et al. 2010) 4 These travel guides first appeared in the 1840s and include Disturnell’s Guide (Disturnell 1846), Doggett’s Gazetteer (Doggett 1847), Appletons’ Guide (Appleton D. and Company. 1848), Dinsmore’s Guide (Cobb 1850), Lloyd’s Guide (Lloyd 1857), Travelers’ Guide (National Railway Publication Company. 1868) and The Rand-McNally Guide (Rand-McNally 1871). Some of these were published monthly; others, semi-annually or annually. Each typically went through many editions. All of the guides that we have physically handled are fragile, especially the multipage fold-out maps, and not sturdy enough for scanning or copying, although a few guides have been digitized and are available on-line. See, for example, the June 1870 copy of the Travelers’ Official Railway Guide at http://cprr.org/Museum/Books/I_ACCEPT_the_User_Agreement/ Travellers_Guide_6-1870.pdf from the Central Pacific Railroad Museum. There are also at least two different editions of Appleton’s Guide on Google Books such as http://books.google.com/ books?vid=UOM39015016751375 as well as a number of other guides. See http://www.lib.utexas. edu/maps/map_sites/hist_guide_sites.html 5 However, as we note below, these sources do not seem to have been infallible—certainly at any moment of time.

• 161

Railroads in historical context: construction, costs and consequences

162 •

Our Midwestern transportation database has been supplemented by a more general but (as yet) temporally less detailed GIS transportation database created from geo-referenced digitized historical maps showing canals, navigable waterways and railroads at different benchmark dates.6 The canal data were taken from Poor (1860) and Goodrich (1961) as well as digitized maps from the Library of Congress “American Memory” web site such as the maps prepared

by Williams (1851) and Disturnell (Burr 1850). Information on navigable rivers was taken from U.S. Army Corp of Engineers GIS data (Vanderbilt University. Engineering Center for Transportation Operations and Research 1999) supplemented by information from Hunter (1949) and contemporary newspaper accounts regarding steamboat service on specific rivers. Since very little change occurred in the canal and river systems after the 1840s until the Corp of Engineers embarked on its ambitious navigation plans beginning with the establishment of an “Office of Western River Improvements” in 1866, we have measured access to water-borne transportation as of a single benchmark date – 1850 – rather than by change over time.

The railroad portion of our national GIS database is based on digitized statelevel maps for 1911 from The Century Atlas (Whitney and Smith 1911) since these maps appear to be accurately drawn and the rail network was largely completed by that time but had not yet begun to shrink through closures.7 This 1911 railroad network was then traced back through time to earlier dates using digitized transportation maps from the Library of Congress and other sources such as the maps created by Taylor and Neu (1956). Our implicit assumption in this approach is that most railroad investment was literally sunk in location-specific grading and other immoveable features, a presumption that is strongly supported by the available data.

To measure railroad access we use a simple county-level binary variable, ACCESS (Atack, Bateman, Haines, and Margo 2010; Atack and Margo 2011).

This variable takes on the value of 1 if a railroad operated within a county’s boundaries or formed a county’s border. This variable should be viewed with 6 Geo-referencing refers to the process of fixing specific points with known geographic coordinates between the digitized image--which was invariably drawn and printed with error (and which may also be subject to parallax error as a result of the digitization process)--and the geographically accurately rendered base boundary files (a shapefile in ESRI’s parlance). Algorithms within the GIS software then distribute the error (the difference) between the historical images and the boundary file across the space between fixed points. In essence, the process treats the historical image as if it were printed on a sheet of rubber which is then stretched over the boundary file with pins holding it in place at fixed reference points between the two. Once done, it is then possible to “trace” features from the historical image onto the geographically coordinated boundary file.

The resulting files can then be manipulated and used for computations using the GIS software.

Accurate historical county boundary files are freely available from the National Historical Geographical Information System at the University of Minnesota (http://www.nhgis.org/) along with a wide variety of historical U.S. census data.

7 These digitized images were purchased from Goldbug.com but the original source may be found in many libraries around the country including Vanderbilt’s.

Jeremy Atack • Robert A. Margo

caution because it does not “cross” county boundaries – a county could lack access in our sense yet parts of it might be accessible to a rail line in the next county over. We are currently in the process of refining our measures of rail access, for example, to estimate the share of a county area within say five or ten miles of a railroad or the area within say five miles of a railroad station or depot.8 This work is still ongoing but preliminary work with these data indicates that our results hold up regardless of the measure which we use. For the present, though, the dummy variable is still the best current measure of the historical spread of the U.S. railroad system at the county level for this period. Because counties differ in size we weight observations by land area in square miles.

Our transportation access data have then been linked with the countylevel Haines-ICPSR census data (http://www.icpsr.umich.edu/icpsrweb/ICPSR/) using the county F(ederal)I(nformation)P(rocessing)S(tandards) codes which are common to both databases. In 1840, the seven states and territorial areas in our study had 391 “counties” (these include, for example, Clayton “county” in the Iowa Territory which then comprised most of what we now know as the state of Minnesota) (Thorndale and Dollarhide 1987). By 1860, however, all of the area under consideration had been completely organized into states and was divided into 623 counties. Because of boundary changes, we have restricted our analysis to those counties that were (1) present beginning in 1840, (2) had the same county boundaries (as determined by square mileage in 1850 and 1860). Taken together these restrictions produce a balanced panel to 278 Midwestern counties. They are distributed as shown in Table 1.9

Our individual level dependent variable, landownership, is computed 8 Geographic proximity to a railroad stations or depots can only be measured beginning in the mid-1850s when maps begin to record this important information. For example, the first digitized map in the Library of Congress collection to mention stations and depots explicitly is by Ensign, Bridgman and Fanning (1856). Paradoxically, it shows many more named places along the right of way than the map from Dinsmore’s travel guide for same year (Fisher 1856) which one might have thought would show prospective travelers their station destinations. Nor does it appear that the Ensign, Bridgman and Fanning map was overly sanguine in its reporting of railroad stations. Another map published two years later (Sage 1858) shows most of the same stations as Ensign, Bridgman and Fanning plus additional ones along the new rights of way as well as a few others scattered between some of the 1856 stations and depots. Beginning in the 1870s, Rand McNally (Morgan 1969) along with other publishers began to produce “commercial” maps for the convenience of shippers, showing destinations to which freight could be consigned making the reporting of depots and stations more reliable and systematic. However, as a practical matter, however, trains could potentially stop and load almost anywhere (albeit inconveniently)—much as they do in parts of Alaska even today—and stations were built only when freight and passenger volume had passed some critical economic threshold.

9 By “balanced” we mean that the same counties appear in 1850 and 1860; no new counties enter the sample during the 1850s. Balancing ensures that county fixed effects are “differenced away” when we compute the change in economic outcomes from 1850 to 1860; this would not be the case if new counties entered the sample in the 1850s. We restrict our basic analysis to counties with fixed land area because the ICPSR census data are not adjusted for changes in land area over time. Results are qualitatively the same if we do not impose the restriction that the county not have rail access by 1850 and if we do not impose the requirement that county boundaries be the same in 1850 and 1860.

• 163

Railroads in historical context: construction, costs and consequences

Table 1: Distribution of Balanced Panel of Sample Counties State

Number of counties

Percent of Total (%)

Illinois

71

25.6

Indiana

65

23.5

Iowa

17

6.1

Michigan

12

4.3

Missouri

41

14.8

Ohio

57

20.5

Wisconsin

14

5.1

Total

277

Notes: To be included in the sample, counties must be (1) present in all three census years: 1840, 1850 and 1860; (2) have fixed county boundaries in 1850 and 1860; and (3) either not have rail access in 1850 but gained rail access by 1860 (=treatment group) or did not have rail access in either 1850 or 1860 (=control group).

164 •

from the IPUMS for 1850 and 1860. The IPUMS provide nationally representative 1 percent random samples in both census years of the free population which we have then restricted to the seven Midwestern states under consideration here and to males between the ages of 20 and 59, which covers the majority of the male labor force. The IPUMS data include county of residence so it is straightforward to link the census data, by year, to our transportation database. There are a total of approximately 12,800 men in the sample overall, about 60 percent of whom come from the 1860 IPUMS, reflecting the overall population growth and the rapid development of the Midwest in the 1850s. The censuses recorded the (self-reported) value of real estate owned by the individual; if this value is greater than zero, landownership is set equal to one, zero otherwise.

Table 2 shows the mean values of landownership by treatment status in 1850 and 1860, and the change in landownership between 1850 and 1860. The difference in differences estimator compares the change between 1850 and 1860 in the counties receiving rail access with that in the control counties. For the entire sample, the estimated effect is negative but small -- gaining rail access reduces the likelihood that an adult male owned land by 2.2 percentage points.

The standard error of the estimate is 2.1 percentage points, so the effect is not statistically significant.

However, although the overall effect is small and not statistically significant, it is clear from the remainder of the table that this overall effect masks big differences across age groups. In particular, the negative effect of the railroad is much larger among younger men, particularly those in the age group 30-39. As

Jeremy Atack • Robert A. Margo

Table 2: Difference-in Differences Estimates of the Effects of Gaining Rail Access: Landownership Among Adult Men, Ages 20-69, in the American Midwest, 1850-60 Age Group

Ages 20-59

Ages 20-29

Ages 30-39

Ages 40-59

1850 Treatment

0.450

0.228

0.549

0.671

Control

0.511

0.248

0.614

0.730

-0.061

-0.020

-0.065

-0.059

0.443

0.202

0.521

0.691

Treatment - Control 1860 Treatment Control Treatment - Control Difference in Diffrences

0.526

0.266

0.660

0.748

-0.083

-0.064

-0.139

-0.057

-0.022

-0.044

-0.074

0.002

Figures show percent owning real estate (value of real estate > 0). N =18,471. Sample consists of males, ages 20-59 from 1850 and 1850 IPUMS linked to Atack-Margo MW transportation database (278 counties). Treatment: county of residence gains rail access between 1850 and 1860. Control: county of residence does not have rail access before Civil War. County means weighted by square miles.

can be seen by examining the sample means, it is between ages 30 and 39 that the gradient between age and landownership was steepest. For men in this age group, the coming of the railroad is associated with approximately a 7 percentage point decrease in the likelihood of owning land, or about a 15 percent reduction in the likelihood of land ownership.

Ideally, in a difference-in-differences analysis, the treatment and control counties will be sufficiently “similar” or “matched” such that treatment can be (plausibly) argued to be randomly assigned. However, it is immediately evident from Table 2 that this was not the case with rail access as far as our dependent variable is concerned. Compared with the control group, counties that gained rail access in the 1850s had a lower proportion of adult men overall who owned any real estate in 1850.

This lower rate of landownership in the treatment counties is a manifestation of a broader pattern –railroads did not arrive randomly in counties. In general, railroad promoters and investors in the Midwest paid close attention to the economic characteristics of counties because the profitability of the railroad depended upon them (Gates 1934; Fishlow 1965). This is illustrated in Table 3, which shows the correlates of gaining rail access in the 1850s using a simple linear probability model applied to our 1860 sample. The dependent variable

• 165

Railroads in historical context: construction, costs and consequences

Table 3: County-Level Correlates of Gaining Rail Access in the 1850s: Adult Men, Ages 20-39, Living in 277 Midwestern Counties Coefficient

Coefficient

Rail Access = 1

Rail Access = 1

Constant

0.371* (0.159)

1.214 (0.561)

Log (Agricultural Yield in 1850)

0.292* (0.094)

0.290* (0.094)

Percent Wheat in 1840

0.121 (0.239)

0.117 (0.237)

∆ in Percent Wheat, 1850-40

0.176 (0.168)

0.175 (0.167)

Percent Urban in 1840

0.632* (0.295)

0.744* (0.295)

Ln (Population Per Square Mile in 1840)

0.037 (0.049)

0.037 (0.049)

∆ in Percent Urban, 1840-1850

0.493* (0.236)

0.491* (0.235)

∆ in Ln (Population Per Square Mile, 1840-1850)

0.034 (0.059)

0.034 (0.059)

Water Transportation Dummies Included?

Yes

Yes

State Dummies Included?

Yes

Yes

Individual Covariates

No

Yes

0.267

0.270

Dependent Variable

166 •

Adjusted R

2

Source: see text and Table 1.

Notes: Standard errors clustered at the county level. Columns 2, 3: Dependent variable = 1 if county gains rail access by 1860 (treatment group), 0 otherwise (control group). Column 4: Dependent variable = 1 if individual owns real estate, 0 otherwise. 1860 sample is used for regressions in columns 2 and 3; 1850 sample is used for regression in column 4. Agricultural Yield: value of agricultural output/(improved + unimproved acres in farms). Percent wheat: value of wheat output/value of total agricultural output. Urban and Population Density variables: percent urban in 1840, ln (population/square miles) in 1840, change in percent urban between 1840 and 1850, change in ln (population/square miles) between 1840 and 1850. Water transportation: canal = 1 if canal existed within county boundary (or part of boundary), river = 1 if navigable river existed within county boundary or part of county boundary, Great Lakes = 1 if county abutted one of the Great Lakes. Observations are weighted by the number of square miles in the county.

Standard errors are shown in parentheses. *: significant at 5 percent level.

equals 1 if the individual resided in a county which gained rail access in the 1850s – that is, the county is in the treatment group – and zero if not – the county is in the control group.10 10 Given that the 1860 IPUMS was produced by random sampling by using the individual as the unit of observation we are effectively controlling for population size.

Jeremy Atack • Robert A. Margo

In the first column, we include control variables measured at the county level as well as dummy variables for states. State dummies are included because there are numerous possible determinants of gaining rail access that are difficult or impossible to control for directly, given the available data. Three of our variables are indicators of agricultural productivity: the logarithm of the value of agricultural output per acre; the percentage of the value of total agricultural output reported in the 1840 census accounted for by wheat production; and the change in the percent wheat, as just defined, between 1840 and 1850.11 We also include the percent urban in 1840, the log of population density in 1840, and changes in these variables over the 1840s. Finally we include dummy variables for access to different modes of water transportation (river, canals, and abutting the Great Lakes).

The results show that counties that were already fairly urban in 1840 or which became more urbanized over the 1840s, as well as counties that had high agricultural productivity (as measured by yields) were more likely to gain rail access over the 1850s, other factors held constant. When we include individual controls for age, literacy status, ethnicity, race, and disability (column 2), our substantive findings with regard to the impact of county characteristics are unchanged.

Because gaining rail access clearly was not a random event, it is important that we control for these important factors in our difference-in-differences analysis (Table 4) in so far as they may have had independent effects on the change in percentage of improved acreage between 1850 and 1860. Column 1 reports our base DID estimate for all males in the IPUMS aged 20-59; by construction, this is the same as in Table 1, panel A. In columns 2 and 3 we add individual and county controls from Table 3. Adding these controls increases the magnitude and statistical significance of the treatment effect. Thus, for example, in column 4, including both county and individual controls, the coming of the railroad reduces the incidence of land ownership by 4.6 percentage points.

In the remaining columns we re-estimate the regression with the full set of controls conditional on an agricultural occupation. The negative impact of the rail access on landownership is slightly larger for farmers than non-farmers for the overall sample, but the difference in the coefficients is very small. However, when we re-estimate for the younger age group, 20-39, the effect is considerably larger for farmers than for non-farmers.

11 The 1840 census reports the value of wheat production and the total value of agricultural output.

For 1850 we use an estimate of the value of total agricultural output based on national prices multiplied by quantities; the wheat share is therefore the wheat output (in bushels) multiplied by wheat price divided by the estimated value of agricultural output. It is possible that our procedure for estimating agricultural values may overstate the growth in the percent wheat over the 1840s (because 1840 output was probably valued at local prices) but any such bias should be mitigated once we control for state fixed effects (since state-level variation arguably captures the most salient price variation). Our agricultural yield variable also uses the estimated value of agricultural output in its construction (the numerator).

• 167

-0.022 (0.021)

Gain Rail Access?

-0.043* (0.020)

Individual

Ages 20-59

-0.046 (0.022)

Individual + Conty -0.048** (0.025)

Individual + Conty

Agricultural Occupation =1

Coefficient

-0.038 (0.038)

Individual + Conty

Agricultural Occupation =0

-0.079 (0.033)

Individual + Conty

Agricultural Occupation = 1, Ages 20-39

-0.021 (0.040)

Individual + Conty

Agricultural Occupation = 0, Ages 20-39

All regressions include dummy variables for year and county. Individual: dummy variables for age group (30-39, 40-49, 50-59), Irteracy, rate, place of birth, disability status, and interactrons between age group and year, and age group and treatment status. County: (year = 1860) x (percent urban in 1840, change in percent urban between 1840 and 1850, log of pop density in 1840, change in log pop density between 1840 and 1850, percent wheat in 1840, change in percent wheat between 1840 and 1850, dummy variables for presence of canal, river, or abutting Great Lakes, state). Agricultural occupation =1. if occupation code for farmer, farm laborer, or living on a farm and laborer not otherwise classified. Observations weighted by square miles in county of residence. Standard errors clustered by county. *significant at 5 percent level; **significant at 10 percent level.

None

Covariates

Sample

168 •

Table 4: Regression Estimates

Railroads in historical context: construction, costs and consequences

Jeremy Atack • Robert A. Margo

Table 5: Instrumental Variables Estimates Own = 1

OLS, Gain Rail Access?

First Stage, coefficient of Survey IV TSLS, Gain Rail Access

Own = 1

Own = 1

-0.040* (0.015)

% Farms Under 50 Acres

% Farms Under 50 Acres

% Farms Under 50 Acres

-0.064* (0.017) 0.227* (0.058)

0.232* (0.058) -0.189* (0.086)

-0.128 (0.123)

1860 sample only (N = 11,355). Own = 1 if positive value of real estate reported. Covariates in ownership regression include indivrclual variables (age, place of birth, literacy, race, and disability status) and county variables (state, percent urban in 1840, change in percent urban 1840-50, log of population density in 1840, change in log of population density 1840-50, percent wheat in 1840, change in percent wheat 1840-50, log of agricultural yield (value per acre) in 1850, dummy variables for presence of canal, navigable river, or abutting Great Lakes). % of Farms under 50 acres: computed from farm size distribution data in 1860 census. First stage: coefficient of survey IV in regression of rail access; full set of individual and county covariates. TSB: second stage, two stage least squares coefficient of rail access. *: significant at 5 percent level.

We explore one possible explanation for the negative effect of rail access on landownership in agriculture – an increase in the “minimum efficient scale” of farms when transportation access improves due to rail access. Although the general opinion among economic historians is that there were limited opportunities for economies of scale in northern agriculture before the Civil, limited does not necessarily mean “constant returns to scale.”

If transportation access led to an increase in farm size, we would expect to see a shift towards larger farms at the county level when the county gained rail access. Unfortunately, we cannot use difference-in-differences methods to investigate this hypothesis because the 1850 census did not report the distribution of farm sizes. However we can estimate a cross-sectional regression using 1860 data, because the 1860 census did report the distribution of farm sizes at the county level.

Row 1 in Table 5 shows the coefficient of rail access in a regression of the share of farms with fewer than 50 acres. The regression is estimated over our IPUMS-linked sample and includes the full set of individual controls in the other columns along with county controls. The coefficient of rail access is -0.064, indicating that, controlling for other factors, the share of small farms decreased by about 6.4 percentage points when a county gained rail access.

The negative effect of rail access on small farms is consistent with the hy-

• 169

Railroads in historical context: construction, costs and consequences

170 •

pothesis that the coming of the railroad led to an increase in farm size, possibly because such farms could capture economies of scale as they took advantage of increased opportunities for trade. But larger farms were, by definition, more expensive farms.

An increase in optimal farm size would not necessarily result in a smaller proportion of landowners among adult men – there could be multiple owners per farm—the census schedules, for example, occasionally listed more than one name for the operator of a farm. However, while such arrangements may have existed, they appear to have been uncommon; rather, the typical organizational structure was “one farm, one owner”. Although larger farms would have likely used less labor per acre, total labor demand would not necessarily decline – and, indeed, if we estimate a difference-in-differences regression of the probability that an adult man in our sample held an agricultural occupation, we find little or no evidence of a railroad effect. Thus the likely explanation of the railroad effect is that, with the coming of the railroad, it took longer for men to move up the agricultural ladder and, thus, at a point in time, the share of men owning land declined. Indeed, the 1937 Presidential commission looking into rising tenancy placed the blame for the persistent long run trend squarely on the rising entry costs in the face of imperfect capital markets (United, Gray et al. 1937).

We have performed a robustness check on our result because, although we can control for various correlates of gaining rail access in our DID analysis, it is still possible that the analysis is invalid because the treatment – gaining rail access – is correlated with the error term even after controlling for the observable determinants of gaining rail access. This problem is of particular concern here because the 1840 census reported no statistics on land ownership making it impossible for us to control for the pre-1850 trend in the fraction of adult men owning land. If this trend is correlated with the coming of the railroad – likely, in view of the fact that the treatment and control counties clearly differed in ownership patters in 1850 – the DID treatment estimates will be biased, even when we control for the factors associated with the coming of the railroad. Further, and as just noted, we have no data on the farm size distribution in 1850 and thus are limited to the cross-sectional analysis discussed above..

One way to deal with this problem is to re-estimate the relationship using an instrumental variable—a variable that predicts gaining rail access in the 1850s when we control for other factors but which is otherwise uncorrelated with the outcomes we are examining. For our IV, wWe have used whether or not a county lay along a straight line between the two termini of federal government transportation surveys conducted during the 1820s and 1830s. Using such a variable which isolates plausibly exogenous variation in rail access such as we propose

Jeremy Atack • Robert A. Margo

here is similar to what would have been the case if rail access had been randomly assigned. We then predict rail access in 1860 using the instrumental variable and examine the effect of that predicted access on the percent of land that was improved in 1860.

The historical narrative of internal improvements in America, particularly that for canal construction but also arguably for railroads, assigns an important role to the government in promoting these advances (Goodrich 1961). One such important source of government assistance was the assignment of the Army Corp of Engineers to conduct surveys for potential transportation routes. Beginning in 1824, the President was granted authority to survey routes for “such roads and canals as he may deem of national importance, in a commercial or military point of view, or necessary for the transportation of the public mail” (U. S. Congress. 1823-4). Although railroads were not mentioned in the original act (hardly surprising since the survey law predates even the Stockton to Darlington Railway in England), it was not long before surveys conducted under this legislation also considered them. For example, in 1825 a survey to “ascertain the practicability of uniting the headwaters of the Kenawha [sic] with the James river and Roanoke river” expressly mentions railroads. Soon thereafter, railroad routes came to dominate the surveys with perhaps as many as 61 such surveys being made before the law was repealed by Andrew Jackson’s administration effective in 1838 (Haney 1908, p. 277).

Our specific instrumental variable is derived from the government surveys reported in American State Papers and compiled by Haney (1908, p. 283). For each survey, we have identified the counties at the start and endpoint of the proposed line. For example, a 1831 railway survey plotted a route from Portage Summit on the Ohio Canal (near Akron) to the Hudson River (we used Albany as the terminus is not otherwise specified). In some cases, we inferred both endpoints such as for a 1832 survey for a route between “the Mad River and Lake Erie” in Ohio (Haney 1908, p. 286). We used Springfield and Sandusky as the termini of this projected railroad. Having identified the starting and ending counties, we then drew a straight line between the center of the “start” and “end” counties. Counties that lay along this straight line received a value of one, while those that did not were coded as zero. That is, if a railroad were built, our instrument presumes that it would be built in a straight line as this is the shortest distance between the two points.12 These U.S. Army Corp of Engineers’ surveys provided valuable information to the general public and prospective railroad promoters alike regarding topog12 Our use of a “straight-line” instrument is inspired in part by Bannerjee, Duflo, and Qian (2006) who construct a similar instrument for their study of the impact of rail access in modern China on wages. Or course, many features of topography other than the shortest-distance criterion – grade, hills or mountains, and so on – influenced railroad building but these features of topography likely affected density and urbanization directly, and thus are not candidates for instrumental variables (they fail the exclusion restriction).

• 171

Railroads in historical context: construction, costs and consequences

172 •

raphy and other factors that would affect potential construction costs. Since the costs of these surveys were borne by the public purse, their existence should have raised the likelihood that a railroad would eventually be built by lowering its private costs. Indeed, Taylor (1951, p. 95) even argued “[a]s trained engineers were still very scarce … the government rendered a uniquely valuable service by making its experts available for such surveys.” Moreover, as Haney (1908, p.

284) observed “it is of some significance that in most cases the routes of these government surveys were early taken by railways … in the great majority of cases these early surveys have been closely followed” [emphasis added]. Indeed, many of the interstate highways today follow these same routes. Presumably, therefore, the surveys were found very useful by the railroad engineers charged with the construction of a line and so it would seem that our “Congressional Survey” instrument should be well suited to predicting whether or not a county gained rail access.

Table 5 reports our estimates using this instrumental variable for the full sample of men (ages 20-59). The Congressional Survey instrument does quite well in predicting treatment (gaining rail access in the 1850s) even when we control for all of the other variables included in the DID analysis (this is the “first stage” coefficient shown). We have then used this first stage regression to predict the probability of gaining rail access, and used the predicted values of treatment in the second stage of the two-stage least squares (2SLS) regression of 1860 outcomes.

The 2SLS coefficient of the effect of rail on landownership is considerably larger than the OLS estimate, and is statistically significant. We also find a larger negative effect of rail access on farm size, but in this case the IV coefficient is imprecisely estimated.

CONCLUDING REMARKS We have used a novel data set on the antebellum transportation networks in the Midwest derived from applying GIS software to digitized historical maps to estimate the impact of gaining rail access on the incidence of land ownership.

Real estate was the principal asset in antebellum America (as it is and has been in many parts of the world), and the proportion of adult men owning land is directly related to wealth inequality.

Our primary finding is that that the coming of the railroad was associated with a decrease in the proportion of adult men owning land which later studies

Jeremy Atack • Robert A. Margo

would relate to rising land costs and imperfect capital markets(United, Gray et al. 1937). A significant negative effect is apparent in a difference-in-difference analysis in which we simultaneously control for factors associated with the coming of the railroad. The negative effect is substantially larger when we allow for the coming of the railroad to be endogenous, using an instrumental variable derived from railroad surveys conducted in the 1820s and early 1830s. A plausible mechanism is that the coming of the railroad led to an increase in farm size. Larger farms were more costly than smaller farms; at the margin it took longer for men to move up the agricultural ladder, and some never made it to the top rung, In this way, our analysis shows how a fundamental and far reaching technological change – the diffusion of the railroad – was a factor behind the Kuznets curve in the American case.

• 173

Railroads in historical context: construction, costs and consequences

BIBLIOGRAPHY Anderson, Russell H. (1929). “Agriculture in Illinois During the Civil War Period, 18501870.” University of Illinois, Urbana, IL. . Ph.D Appleton D. And Company. (1848). Appletons’ Railway and Steam Navigation Guide for the United States and the Canadas. New York, D. Appleton & Co. Atack, Jeremy (1988). “Tenants and Yeomen in the Nineteenth Century.” Agricultural History 62(3): 6-32.

Atack, Jeremy (1989). “The Agricultural Ladder Revisited: A New Look at an Old Question with Some Data for 1860.” Agricultural History 63(1): 1-25.

174 •

Atack, Jeremy and Fred Bateman (1981). “Egalitarianism, Inequality, and Age: The Rural North in 1860.” The Journal of Economic History 41(1): 85-93.

Atack, Jeremy and Fred Bateman (1987). To Their Own Soil : Agriculture in the Antebellum North. Ames, Iowa State University Press.

Atack, Jeremy; Fred Bateman, et al. (2010). “Did Railroads Induce or Follow Economic Growth? Urbanization and Population Growth in the American Midwest, 1850-60.”

Social Science History 34(2): 171-197.

Banerjee, A.; Esther Duflo and Nancy Qian (2006). “The Railroad to Success: The Effect of Infrastructure on Economic Growth,”. Providence, Brown University.

Bogue, Allan G. (1963). From Prairie to Corn Belt; Farming on the Illinois and Iowa Prairies in the Nineteenth Century. Chicago,, University of Chicago Press.

Burr, Henry A. (1850). Disturnell’s New Map of the United States and Canada. New York, Disturnell’s new map of the United States and Canada.

Carter, Susan B.; Scott Sigmund Gartner, et al. (2006). Historical Statistics of the United States Millennial Edition Online. New York, Cambridge University Press.

Cobb, Charles (1850). American Railway Guide and Pocket Companion for the United States. New York, C. Dinsmore.

Cogswell, Seddie (1975). Tenure, Nativity, and Age as Factors in Iowa Agriculture, 18501880. Ames, Iowa State University Press.

Jeremy Atack • Robert A. Margo

Colton, G. W. (1856a). Missouri. (with) Vicinity of St. Louis. . New York, J.H. Colton & Co. Colton, G.W. (1857). Iowa. New York, J.H. Colton & Co. Colton, G.W. (1856b). Iowa. New York, J.H. Colton & Co. .

Colton, J. H. (1860). Colton’s New Railroad & County Map of the United States and the Canadas &C. New York, J. H. Colton.

Condit, Carl W. (1980). The Port of New York. Chicago, University of Chicago Press.

Craig, Lee A.; Raymond B. Palmquist, et al. (1998). “Transportation Improvements and Land Values in the Antebellum United States: A Hedonic Approach.” Journal of real estate finance and economics 16(2): 173-189.

Disturnell, John (1846). Disturnell’s Railroad, Steamboat, and Telegraph Book : Being a Guide through the Middle, Northern, and Eastern States and Canada : Also Giving the Great Lines of Travel South and West and the Ocean Steam Packet Arrangements, Containing Tables of Distances, &C. Telegraph Lines and Charges, List of Hotels, Express Offices, &C. &C. New York, J. Disturnell.

Doggett, John (1847). Doggett’s Railroad Guide and Gazetteer For ... With Sectional Maps of the Great Routes of Travel. New York, J. Doggett Jr. Dunbar, Willis F (1969). All Aboard! A History of Railroads in Michigan. Grand Rapids, Michigan, William B. Eerdmans.

Fishlow, Albert (1965). American Railroads and the Transformation of the Antebellum Economy. Cambridge,, Harvard University Press.

Fogel, Robert William (1964). Railroads and American Economic Growth: Essays in Econometric History. Baltimore,, Johns Hopkins Press.

Gates, Paul W. (1945). “Frontier Landlords and Pioneer Tenants.” Journal of the Illinois State Historical Society 38(June): 142-206.

Gates, Paul Wallace (1934). The Illinois Central Railroad and Its Colonization Work.

Cambridge,, Harvard University Press.

Goldenweiser, E. A. and Leon E. Truesdell (1924). Farm Tenancy in the United States.

An Analysis of the Results of the 1920 Census Relative to Farms Classified by Tenure Supplemented by Pertinent Data from Other Sources. Washington,, Govt. Print. Off. Goodrich, Carter (1961). Canals and American Economic Development. New York,, Columbia University Press.

Haites, Erik F.; James Mak, et al. (1975). Western River Transportation : The Era of Early Internal Development, 1810-1860. Baltimore, The Johns Hopkins University Press.

Haney, Lewis H. (1908). A Congressional History of Railways in the United States.

Madison, Wis.,, Democrat Printing Co. Hunter, Louis C. (1949). Steamboats on the Western Rivers; an Economic and Technological History. Cambridge,, Harvard University Press.

• 175

Railroads in historical context: construction, costs and consequences

Kahn, Charles M. (1988). “The Use of Complicated Models as Explanations: A ReExamination of Williamson’s Late 19th Century America.” Research in Economic History 11: 185-216.

Lloyd, E. (1857). Lloyd’s American Guide : Containing New Arranged Time Tables, So Simple and Correct That a Child Can Understand Them, It Being Universally Acknowledged That All Other Guide Books Are So Complicated That Not One in a Hundred Can Understand Them : The Population, States, and Distances to Every Place on All the Railroad Routes in the United States and Canadas : Photographic Portraits of All the Railroad Presidents and Superintendents--Men Controlling. Philadelphia, E. Lloyd.

Meints, Graydon M. (1992). Michigan Railroads and Railroad Companies. East Lansing, Michigan, Michigan State University Press.

Mitchell, Samuel Augustus (1859a). A New Map of the State of Iowa. Philadelphia, Charles Desilver.

Mitchell, Samuel Augustus (1859b). A New Map of the State of Missouri. Philadelphia, Charles Desilver.

National Railway Publication Company. (1868). Travelers’ Official Railway Guide of the United States and Canada. Ann Arbor, Mich., printed for the National Railway Publication Co. by University Microfilms.

176 •

Paxson, Frederic L. (1914). “The Railroads of the “Old Northwest” before the Civil War.” Transactions of the Wisconsin Academy of Sciences, Arts, and Letters 17(Part 1): 247274.

Phillips, Ulrich Bonnell (1908). A History of Transportation in the Eastern Cotton Belt to 1860. New York,, Columbia University Press.

Poor, Henry V. (1860). History of the Railroads and Canals of the United States. New York,, J.H. Schultz & co.

Poor, Henry Varnum (1970). History of the Railroads and Canals of the United States of America. New York,, A. M. Kelley.

Puffert, Douglas J. (1991). The Economics of Spatial Network Externalities and the Dynamics of Railway Gauge Standardization.

Rand-McNally (1871). The Rand-McNally Official Railway Guide and Handbook. Chicago,.

Sage, J (1858). J. Sage & Son’s New & Reliable Rail Road Map Comprising All the Railroads of the United States and Canadas with Their Stations and Distances, Compiled from the Most Acurate Statistics. Buffalo, J. Sage & Sons.

Stover, John F. (1978). Iron Road to the West : American Railroads in the 1850s. New York, Columbia University Press.

Stover, John F. (1987). History of the Baltimore and Ohio Railroad. West Lafayette, Ind., Purdue University Press.

Jeremy Atack • Robert A. Margo

Taylor, George Rogers (1951). The Transportation Revolution 1815-1860, Holt, Rinehart & Winston.

Taylor, George Rogers and Irene D. Neu (1956). The American Railroad Network, 18611890. Cambridge,, Harvard University Press.

Thorndale, William and William Dollarhide (1987). Map Guide to the U.S. Federal Censuses, 1790-1920. Baltimore, Genealogical Publishing Co. U. S. Congress. (1823-4). Laws of the U.S. Vii, 239. . Acts of 1823-24, Ch. 276.

United, States; L. C. Gray, et al. (1937). Farm Tenancy, Report of the President’s Committee.

Washington, U. S. Govt. print. off.

United States. Census Office. (1990). Statistics of the United States (Including Mortality, Property, &C.,) in 1860 : Compiled from the Original Returns and Being the Final Exhibit of the Eighth Census, under the Direction of the Secretary of the Interior. New York, N.Y., Norman Ross Publ.

United States. Census Office. and J. C. G. Kennedy (1862). Preliminary Report on the Eighth Census, 1860. Washington, Government Printing Office.

Vanderbilt University. Engineering Center for Transportation Operations and Research (1999). National Waterway Network, U.S. Army Corps of Engineers Navigation Data Center Whitney, William Dwight and Benjamin E. Smith (1911). The Century Dictionary and Cyclopedia, with a New Atlas of the World. New York,, The Century co.

Williams, W. (1851). A New Map of the United States. Upon Which Are Delineated Its Vast Works of Internal Communication. Philadelphia, Lippincott, Grambo & Co. Williamson, Jeffrey G. (1974). Late Nineteenth-Century American Development : A General Equilibrium History. London ; New York, Cambridge University Press.

Winters, Donald L. (1977). “Tenancy as an Economic Institution: The Growth and Distribution of Agricultural Tenancy in Iowa, 1850-1900.” The Journal of Economic History 37(2): 382-408.

Wright, Gavin (1988). “American Agriculture and the Labor Market: What Happened to Proletarianization?” Agricultural History 62(3): 182-209.

• 177

Marta Felis-Rota • Jordi Marti Henneberg

THE EVOLUTION OF THE ESTABLISHMENT OF THE RAILWAY NETWORK IN BRITAIN USING GIS A INSTALAÇÃO DA REDE DE CAMINHOS DE FERRO NO REINO UNIDO, COM BASE EM SIG Marta Felis-Rota (U. Autónoma de Madrid, Spain Espanha) Jordi Martí Henneberg (U. Lleida, Spain Espanha) Marta Felis-Rota is Ph.D. in Economic History by the London School of Economics and Political Science (LSE). She was formerly trained as a B.Sc. and M.Sc. economist at Pompeu Fabra University in Barcelona and École des Hautes Études Commerciales (HEC) in Lausanne, Switzerland. She has taught international economic history at the London School of Economics, Pompeu Fabra University, University Pablo de Olavide and Universidad Autónoma de Madrid. Currently, she holds an Assistant Professorship in Economic History at the Universidad Autónoma de Madrid. Her research work so far has been financed by the Economic and Social Research Council (ESRC), the Spanish Ministry of Science and Innovation, and the Regional Government of Madrid. Her research is on social capital, economic history, institutions and international trade, and transportation. Her research on social capital earned the Graduate Student Prize from the Society for the Advancement of Socio-economics. She has published at the Revista de Historia Económica: Journal of Iberian and Latin American Economic History, Cambridge University Press. http://www.uam.es/marta.felis Jordi Martí-Henneberg is Professor of Human Geography at the University of Lleida, Spain. He has lead an European Science Foundation Eurocores project (2007-2010): The Development of European Waterways, Road and Rail Infrastructures: A Geographical Information System for the History of European Integration (1825-2005), which has involved researchers from eight different European states.

Over the last ten years, he has focused the majority of his research work on European Integration. He thinks that in order to understand the process of institutional integration during the last fifty years, it is necessary to adopt a broad historical perspective that includes geohistorical aspects. He is leading three Jean Monnet projects on European Integration for the period 2011 - 2013.

Marta Felis-Rota é doutorada em história económica pela LSE (London School of Economics and Political Science). Professora Assistente da Universiade Autonoma de Madrid (Espanha). Ensinou história económica internacional na London School of Economics, Universidade Pompeu Fabra, Universidade Pablo de Olavide e agora na Universidad Autónoma de Madrid.

Jordi Martí-Henneberg é professor de geografia humana na Universidade de Leida, Espanha. Entre 2007 e 2010 dirigiu um projeto da European Science Foundation (The development of European Waterways, Road and Rail Infraestructures: a Geographical Information System for the History of European Integration, 18252005) que envolveu investigadores de oito paises europeus.

Nos últimos dez anos tem trabalhado em investigação sobre a integração europeia.com base numa perspectiva histórica que inclua os aspectos geohistóricos.

• 179

Railroads in historical context: construction, costs and consequences

Abstract Resumo

180 •

We reconstruct historical maps of the British railway network since its foundations in the 19th century using Geographical Information Systems (GIS).

We present a series of maps of the British railway network every 10 years, starting as early as 1830, and up to year 2000. This facilitates the study of the evolution of the network from a historical perspective. We have seen that there have been different construction phases. We have divided the period of study, 1830 to 2000, into 3 phases: The first one being 1830-1920, what we call the long 19th century.

This first phase is characterized by the construction boom of the 19th century. The second phase comprises the years 1920 to 1950, and signifies the stagnation of the construction process. Finally, the third and final phase we have described runs from 1950 to 2000. This period sees the more drastic line and station closures in history, accounting to up to 50 percent of constructed network. We observe that since after the massive closures of the 1950s and 1960s, the main feature of the evolution of the network has been plain stability.

Acknowledgments to Ministerio de Economía y Competitividad (Ref. CSO201016389 and ECO2010-21643), European Commission (Jean Monnet 200215-LLP2011-ES-AJM-IC) Usando SIGs, reconstruimos os mapas históricos da rede britanica de caminhos de ferro desde a sua fundação no século XIX. Apresentamos uma série de mapas da rede britanica de caminhos de ferro por decada, desde 1830 até 2000, que facilitam o estudo da evolução da rede numa perspective histórica. Dividimos o período de estudo em tres fases. A primeira, de 1830 a 1920, caracteriza-se pelo boom de construção no século XIX. A segunda fase, de 1920 a 1950, mostra uma estagnação do processo de construção. Finalmente o terceiro periodo, de 1950 a 2000, conhece o encerramento linhas mais drástica da história, que afectou até 50% da rede até aí construída. Desde os encerramentos dos anos 50 e 60 do século XX, a estabilidade da rede ferroviária parece dominar.

Agradecimentos ao Ministerio de Economía y Competitividad (Ref. CSO2010-16389 and ECO2010-21643) e Comissão Europeia (Jean Monnet 200215-LLP-2011-ESAJM-IC).

Marta Felis-Rota • Jordi Marti Henneberg

The evolution of the establishment of the railway network in Britain using GIS Marta Felis-Rota, Jordi Marti Henneberg

1. INTRODUCTION Information technologies have brought the information revolution in the last decades, changing the way we process and assimilate information altogether.

History is not an exception. Geographical Information Systems (GIS) have been a revolution in the way we analyse historical records. We can now assemble bits and pieces of information belonging to a certain spatial location, and then bring these bits and pieces together. This will result into a historical map, which will help us best interpreting historical records. This is the case for the building of the railway network in England and Wales during the 19th century, leading places in the world as for railway technology at the time, and key to understanding all railways’ building that came subsequently in other parts of the world.

Historical Geographical Information Systems Europe (HGISe) is a multidisciplinary project gathering engineers, geographers and economists altogether with the aim of assembling, diffusing and interpreting information on historical transportation networks across Europe (http://www.europa.udl.cat/). This project gathers historical information on population, urbanization, borders, and transportation networks. At the time of assembling all historical records regarding the British case, we realized that the British industrialization process starting in the last decades of the 18th century and running all the way through 19th century was articulated around a network

• 181

Railroads in historical context: construction, costs and consequences

of existing and new means of transportation, most notably the railways. The railways started as a way of transporting coal in the context of the just-born manufacturing centers and turned into a means of transporting heavy goods, fresh products, postal service and people indiscriminately. The richness and historical importance of the British case makes it worth it a more detailed look.

We have produced a GIS that combines a system for validating information about railways and stations with the population data at the civil parish level.

This opens a new window on studies of the historical geography of Britain and one that will enable conducting a detailed analysis of a series of factors, such as population changes at the parish level or changes in the work force, which will help to reveal the important role that railways played in determining the configuration of the British landscape in contemporary history. This set of key factors was already well-known, but it is only now that it will be possible to measure their respective influences. In the rest of this chapter we provide an overview survey of the evolution in the construction of the British railway network with the support of the GIS-constructed historical maps.

182 •

2. DATA SOURCES The GIS software provides the technical platform we need, but we still need to compile detailed geographical information on the exact location of new and existing train stations and railway lines. “The Railways of Great Britain” (Cobb, 2005) constitutes an important empirical advance in existing knowledge concerning the expansion and subsequent decline of the British railway network.

The Atlas provides very detailed information about the dates of opening and closure of both railway lines and stations and also about the companies that were most influential in these processes. This book shows the railways against the Ordonance Survey (OS) grid and contains additional detailed information in the appendices that has turned valuable for our mapping purposes.

Other contributions examine the importance of railway history at the national level. The Ian Allan series, for example, constitutes a very relevant example of this in Britain. Still, one of the reasons why Colonel Cobb’s contribution is so relevant is because the author has devoted great amounts of time to establishing an empirical infrastructure for further research. Thanks to its accuracy and the OS grid, this Atlas makes it possible to transfer information to a Geographical Information System (GIS). The methodology we use takes into account the principal options for the validation of the digitalisation process. This is an im-

Marta Felis-Rota • Jordi Marti Henneberg

portant step, as there are more than 8,500 stations in the complete structure of the British railway network. Technically, these constitute points at which people and goods have access to a regular service that provides transport connections for the whole country.

An example of what type of result can we obtain using this technique is in the publication by Gregory and Schwartz (2009), who have combined an analysis of the evolution of population at the regional level with the evolution of the railway network in Wales.

3. PERIODIZATION OF THE GENERAL EVOLUTION OF THE RAILWAY CONSTRUCTION IN BRITAIN We have divided the long period into 2 historical periods coinciding with what we call the long 19th century and the 20th century; and 3 construction phases.

The break that divides the 2 historical periods is the First World War. All the way up to the Great War railways were seen as the future of freight transportation, and were increasingly considering the transportation of passengers as well.

During the War, the government made an effort to keep the system running, given the importance of good communications in an emergency situation. By the 1920s, the automobile was the new and revolutionary form of transportation, and was spreading quickly in the United States, by then already the wealthiest nation in the world. Economically speaking, the end of the war in 1918 would mean the beginning of the next époque. From then onwards, 20th century Britain would be much better pictured by decadence of the railways than by the preceding construction boom. Parallel to this historical periodization, we make 3 clearly differentiated sub-periods, coinciding with the 3 phases in the evolution of the network. The first phase corresponds to what we call the long nineteenth century. This reaches up to the end of the First World War and is characterized by the railway construction mania. The second phase runs from the end of the Great War to 1948 (year of nationalization), and can be understood as a plateau phase. In British railway history we establish the break point at the time of the change in the ownership approach in 1948. This is a turning point from which closures of lines and stations are more characteristic of the coming decades than new openings. This brings a third period running from the beginning of the 1950s to the ending year in this study, the year 2000, and so comprises the second half of the 20th century. These 3 phases are well differentiated and are different in nature.

• 183

Railroads in historical context: construction, costs and consequences

4. THE LONG NINETEENTH CENTURY: CONSTRUCTION OF THE NETWORK

184 •

Several authors have been studying the construction of the railway network in Britain from different perspectives: The physical process of formation of infrastructure (Casson, 2009), political and economic factors that conditioned it (Bogart, 2009), as well as effects in the territorial distribution of population (Schwartz, Gregory and Thevenin, 2011). Our contribution consists of a georeferenced analysis of the railways that allows both quantifying any aspect of the network and also provides a basis for combined analysis with other datasets.

In this chapter, we are going to limit ourselves to describe the evolution of the network in Britain, considering it a physical unity. We are not including the rest of the United Kingdom, due to the fact that it is not included in the Cobb’s Atlas.

The first railway was invented and implemented in Britain for the first time, as is known. Moreover, it experienced a rapid expansion during the first years, which implied gaining an advantage over other European countries. In 1860, density of the British network was fourfold that of the French, and in 1920 was still twofold. Despite posterior closures, Britain continues to enjoy the densest railway facilities in Europe. Of course, this broad service is in relation to high demand, determined by its density of population and economic activity.

Looking at the railway network in more detail, the design of the British network has certain peculiarities, the most relevant being the preeminence of private initiative in financing the railways. Private sector funding certainly conditions the global morphology of the network.

The role of the state was limited to warranting security and granting construction permissions, via previous discussion of the projects in Parliament.

This facilitated the knitting of a complex spider net of local influences and tensions among competing enterprises. For this reason, each fragment of the rail map has their own particular history, often not applicable to the rest.

We can distinguish several phases in the development of the network according to not only expected transportation demand but also investors’ profit expectations. Investors’ dominant confidence was applied to regulate free markets in the United Kingdom. Differently from other countries, the general administration did not promote public lines, nor did it offer any warranty of profitability to private firms. On the contrary, the market was already offering enough stimuli so that we observe spectacular growth in construction and concessions in the 1840s. This was so up to the point that in 1847 railway investment accounted for 6.7 percent of the national income. This investment bubble soon burst to return

Marta Felis-Rota • Jordi Marti Henneberg

• 185

Railroads in historical context: construction, costs and consequences

186 •

35000

Railwaiys lenght, km.

30000 25000 20000 15000 10000 5000 0

1830 1840 1850 1860 1870 1880 1890 1900 1910 1920

to more realistic levels. Still, the first big push was already given and made the British network the densest, as pointed by several authors:

Marta Felis-Rota • Jordi Marti Henneberg

“By the mid-1840s, investing in the railways had become an attractive proposition once again and schemes for new lines began to be drawn up in every region of the country. That financial climate had changed and there was optimism in the air with an upturn in the economic cycle.

Interest rates had plummeted, encouraging people to look for a better rate of return and their savings (…) and by the spring of 1844 there was more money available for railway investment than ever before” (Wolmar, 2007: 87).

Some researchers have considered that “…the early railway network was by no means an aberration in the context of the national interest” (Fullerton, 1975, quoted by Turnock, 1998: 11). A qualified perspective makes us understand that such density favored rural areas (Schwartz, Gregory and Thevenin, 2011). Indeed, “... countryside distribution was also feasible for a wide range of manufacturers and enterprising firms like brewers in Burnon on Treant, marmalade producers in Dundee and sugar refiners in Greenock” (Turnock, 1998: 26).

The high density network favored then and conditioned ever since the organization of industrial production in all the country. By 1860 total railway length amounted to 15,000 kilometers, and continued growing. By the turn of the century it had doubled to more than 30,000 kilometers in a period of 40 years. The number of operative stations reached around 7,000 at the beginning of the twentieth century, which indicates wideness and permeability of the network, with numerous access points. Later on it had been stabilized below 3,000 stations, not only due to line closures but also to closure of stations within operative lines with the aim of raising commercial speed and diminishing maintenance costs.

We will see this in the next section.

5. THE TWENTIETH CENTURY: DECADENCE AND NEW MEANS OF TRANSPORTATION Generally speaking, we can argue that railway infrastructure had three big construction phases. The first phase (1830-1910) corresponds to the dawn and later sustained rise of the network. Expansion was however slower in the last part of the period. This phase shaped the geography of the network, with two differential features in comparison to that of other European countries. The first difference is high density of the network and the second one is the strong variation of this agglomeration in favor of urban areas, such as London or Manchester-

• 187

Railroads in historical context: construction, costs and consequences

188 •

Liverpool, among others. The provision of the railway infrastructure facilitated the processes of dispersion of urban areas departing from their central nuclei.

This feature was maintained all the way during the 20th century.

The 20th century implied the decadence of the railways due to new means of transportation. i. e. the automobile from the 1920s onwards. Although the automobile was not generalized until much later, the expenses of maintaining such a high density network were starting to reveal during this period, and were already heavy to sustain before that, especially during the Great War. Precisely because of this saturation process, the second period of division of the evolution of the railway network (1918-1948) is characterized by the stability of the network. The railway network length was maintained well above 30,000 kilometers of line until the 1950s. Still, some lines were opened during the first years, but the beginning of line closures was already starting to show towards the end of this period. In the 1940s, some areas of Scotland and Wales were already affected by line closures. Nevertheless, it is remarkable the fact that between 1910 and 1920 the government made an effort to keep the whole network functioning during the Great War, in detriment of good maintenance though.

The problem of high density remained until what we consider to be the third phase of construction, 1948 to 2000. Or maybe should we say de-construction in this case, since this period is characterized by massive line and station closures and posterior maintenance of the status quo of the layout until present day. After nationalization in 1948, in the 1950s the alternatives were either making a great investment effort or closing down a great deal of the network. The main result of the official studies conducted at the time was the Beeching Act of 1963, which opted for the second option. The result was that around 50 percent of the railway layout was closed down. This was the most drastic line closure in the history of European railways. The reduction was even larger in percentage terms if we take as reference the number of stations in service, since many stations in operative lines were closed too; the reason being that speed was given preference in detriment of connectivity in less populated areas.

Finally, during the last 2 decades of the study (from 1980 onwards), we can observe a new phase of stability, with isolated closures and only one significant opening: the high-speed train line from Waterloo station in London to the French coast. Year 2000 displays similar railway network length as the mid1860s. But beyond crude numbers that stabilize the total length of the network above 16,000 kilometers at the end of the 20th century, we need to emphasize that this was a move towards efficiency. Although isolated, the high-speed new line opening has required a vast amount of investment and technology update, together with international coordination, and so it has brought a new percep-

Marta Felis-Rota • Jordi Marti Henneberg

tion of travelling by train and signifies a true modernization of this 19th century means of transportation.

• 189

Railroads in historical context: construction, costs and consequences

35000

Railwaiys lenght, km.

30000 25000 190 •

20000 15000 10000 5000 0

1930

1940

1950

1960

1970

1980

1990

2000

CONCLUSIONS Geographical Information Systems (GIS) techniques and new software have brought the opportunity to do research in a new way. Mapping historical information making use of GIS software has made it possible to address the old questions in a new way. In this case, we have been able to reconstruct historical maps of the British railway network since its foundations in the 19th century and all the way up to present day, departing from a series of coordinates in a map. This facilitates the study of the evolution of the network from a historical perspective.

We present a series of maps of the British railway network every 10 years, starting as early as 1830, when we can observe something of a more significant railway construction activity; more than some isolated or anecdotal lines, and

Marta Felis-Rota • Jordi Marti Henneberg

up to year 2000, when we already have the most sophisticated high-speed train linking central London to the French coast.

Almost 200 years have elapsed since the construction of the first railway as is known today to nowadays, and we have seen that there have been different phases. We have divided the period of study, 1830 to 2000, into 3 phases: The first one being 1830-1920, what we call the long 19th century. This first phase is characterized by the construction boom of the 19th century, with some stagnation periods in between but that runs until the end of the First World War. The second phase comprises the years 1920 to 1950, and signifies the stagnation of the construction process. This does not mean that there were no new lines open, since this was not the case. Indeed, we observe new lines open, especially in the initial years of the period, but effervescence in construction stopped. Actually, in the last decade of this second phase we can already find some line closures.

Finally, the third and final phase we have described runs from 1950 to 2000.

Right after nationalization, this period sees the more drastic line and station closures in history, accounting to up to 50 percent of constructed network. This is done for efficiency reasons and the initiative lies entirely in the governments’ decision. In the last decades of this period we still find a remarkable opening: the high-speed train, connecting the British railway network to the continent through a tunnel under the English Channel. But this is clearly the exception, not the norm, since after the massive closures of the 1950s and 1960s, the main feature of the evolution of the network has been plain stability.

The network reached around 33,000 km of line during its peak years in the 1920s and 1930s, and had almost 7,000 operative stations. Nowadays, the network is half that length and less than half that number of connectivity points. It has evolved towards a more speedy and efficient configuration.

• 191

Railroads in historical context: construction, costs and consequences

BIBLIOGRAPHY Beeching, R., 1963. The Reshaping of the British Railway. London: Her Majesty’s Stationery Office.

Bogart, D., 2009, Nationalizations and the Development of Transport Systems: CrossCountry Evidence from Railroad Networks, 1860-1912, The Journal of Economic History, Cambridge University Press, 69 (01):202-237, March.

Cobb, M. H., 2007 (2nd ed). The Railways of Great Britain: A Historical Atlas. Shepperton: Ian Allan Publishing.

Casson, M., 2009. The World’s First Railway System. Enterprise, Competition, and Regulation on the Railway Network in Victorian Britain. Oxford: Oxford University Press.

192 •

Casson, M., 2009, The Efficiency of the Victorian British Railway Network: A Counterfactual Analysis, Networks and Spatial Economics, Springer, 9: 339–378.

Fullerton, B., 1975. The Development of British Transport Networks. Oxford University Press, London.

Gregory I.N. and Schwartz R.M., 2009. National Historical Geographical Information Systems as a tool for historical research: Population and railways in Wales, 1841-1911.

International Journal of Humanities and Arts Computing, 3, pp. 143-161.

Haywood, R., 2007. Britain’s national railway network: fit for purpose in the 21st century?.

Journal of Transport Geography 15, 198–216.

Kunz, A., Buiter, H. 2010. Digital Atlas on European Communications and Transport European Communications and Transport Infrastructures: Performance and Potentials, 1825 – 2000, a digital Atlas. Available at www.atlas-infra.eu Livet, G., 2003. Histoire des Routes et des Transports en Europe. Strasbourg: Presses Universitaires de Strasbourg, in Merger, M., and Polino, M.-N. (eds.), 2004.

COST 340. Towards a European intermodal transport network: lessons from History. A Critical Bibliography. Paris: AHICF. Schwartz, R. M., Gregory, I. N., Thevenin, T., 2011. Spatial History: Railways, Uneven Development, and Population Change in France and Great Britain, 1850-1914. The Journal of Interdisciplinary History, XLII(I) 6/2011, 53-88.

Turnock, D., 1998. An Historical geography of railways in Great Britain and Ireland, Aldershot : Ashgate.

Wolmar, Ch., 2007, Fire and Steam, London: Atlantic books.

Richard Healey • Michael Johns

DEVELOPMENT OF AN HISTORICAL GIS OF RAILROADS IN THE NORTHEAST USA 1826-1900: PHASE II DESENVOLVIMENTO DE UM SIG HISTÓRICO DAS LINHAS DE CAMINHOS DE FERRO NO NORDESTE DOS ESTADOS UNIDOS, 1826-1900: FASE II Richard Healey, Michael Johns (U. Portsmouth, UK Reino Unido) Richard Healey is professor in Department of Geography, University of Portsmouth, U.K. and researcher in the School of Computing, University of Portsmouth, in the area of multimedia databases / data warehousing.

Michael Johns studied Computer Network Design and Management at University of Portmouth and he is Technical Consultant at GlassHouse Technologies (Oxfordshire, U.K.).

Richard Healey é Professor no Departamento de Geografia, Universidade de Portsmouth, e investigador na School of Computing da mesma Universidade, sobre o desenvolvimento de bases de dados multimédia e arquivo de dados.

Michael Johns estudou projeto e gestão de redes informáticas na Universidade de Portsmouth e é agora consultor técnico na empresa GlassHouse Technologies (Oxfordshire, Reino Unido)

Abstract Resumo

This project was originally conceived in the mid-1990’s as an outgrowth of research on the anthracite mining industry of Pennsylvania, where both mining and transportation interests were prominently represented. Initial work was focused on the industrial states of Pennsylvania, New York and New Jersey, at a time when GIS software tools were rudimentary by present day standards, scanned historical maps were a rarity and on-line bibliographical sources were virtually unknown. Of more recent years, in the context firstly, of funded research on the migration of heavy industrial workers in the 19th century and secondly, of visualization of railroad development, as part of the JISC/NEH ‘Digging into Data’ Challenge (in collaboration with the University of Nebraska, Lincoln), the project has moved forward substantially, taking advantage of all the developments in both software technology and on-line resources that have taken place over the last decade. The geographical

• 193

Railroads in historical context: construction, costs and consequences

194 •

coverage of the project has also expanded to cover Delaware, Maryland, Ohio, Virginia and West Virginia, with a planned extension into New England.

This paper begins with a brief examination of the wide range of on-line and off-line bibliographic and cartographic sources that are utilized in the project. Aspects of the design and implementation of the data capture, management and analysis system are then addressed, outlining both the GIS and database components.

The GIS is obviously concerned with the location of the railroad lines, while the database holds the organizational chronology of railroad companies, including takeover, mergers and leasing activity. Linkages between the GIS and database components are crucial for tracing the highly complex relationships between the evolving organizational structure of the railroad industry and the changing pattern of ownership of specific track segments on the ground.

The final section of the paper explores current research uses of the railroad GIS, with particular reference to several of the major ‘Eastern Trunk Lines’, such as the Baltimore and Ohio RR, the Erie Railway and the Pennsylvania Railroad. Future potential applications in relation to broader work on regional industrial development will also be considered, as time permits.

Este projeto nasceu em meados da década de 1990 a partir de uma investigação sobre a industria mineira de extração de antracite na Pensilvânia, em que tanto os interesses mineiros como de transportes estavam fortemente representados. O trabalho inicial focou-se nos estados industriais da Pensilvânia, Nova Iorque e Nova Jersey, quando os softwares aplicativos de SIG eram ainda rudimentares pelos padrões atuais, a digitalização de mapas históricos era uma raridade e as fontes bibliográficas on line eram virtualmente desconhecidas. Nos anos mais recentes, e primeiro no contexto de uma investigação sobre a migração de trabalhadores da industria pesada no século XIX, e depois na visualização do desenvolvimento de caminhos de ferro, como parte do JISC/NEH Challenge “Diging into data” (em colaboração coma Universidade de Nebraska, Lincoln), o projeto tem avançado substancialmente, tirando partido dos desenvolvimentos quer na tecnologia do software como dos recursos online ocorridos na última década. Por outro lado o âmbito geográfico do projeto aumentou e passou a incluir Delaware, Maryland, Ohio, Virgínia e Virgínia Ocidental, estando planeada a sua extensão para a Nova Inglaterra.

O trabalho começa com uma análise breve das múltiplas fontes bibliográficas online e offline que são utilizadas no projeto. São também tratados aspectos do planeamento e implementação da captura de dados, da gestão e análise dos sistemas, nas suas componentes de SIG e de base de dados. O SIG está

Richard Healey • Michael Johns

naturalmente interessado na localização das linhas de caminho de ferro, enquanto que a base de dados regista a cronologia organizacional das empresas de caminhos de ferro, incluindo as aquisições, fusões e atividades de arrendamento. As ligações entre o SIG e os componentes da base de dados são cruciais para traçar as relações altamente complexas entre a evolução da estrutura da industria dos caminhos de ferro e o padrão variável da propriedade de segmentos específicos da linha no terreno.A parte final explora o uso atual de SIGs de caminhos de ferro em investigação, com uma referencia em particular às principais linhas da “Eastern Trunk Lines”, como os caminhos de ferro de Baltimore e Ohio, a Erie Railway e a Pennsylvania Railroad. Potenciais aplicações futuras em relação ao trabalho mais vasto sobre o desenvolvimento industrial regional são ainda consideradas.

• 195

Railroads in historical context: construction, costs and consequences

196 •

Richard Healey • Michael Johns

DEVELOPMENT OF AN HISTORICAL GIS OF RAILROADS IN THE NORTHEAST USA 1826-1900: PHASE II Richard Healey and Michael Johns INTRODUCTION This paper aims to provide an update on the railroad component of a long-term project to develop an historical geographical information system (GIS) of industrial development in the northeast USA in the period up to 1900. There are several reasons why such a railroad sub-project is worthwhile. Firstly, in historical and geographical terms, from very small beginnings, the railroad industry became enormous in scale and scope, with a network spanning the United States and extending to nearly 200,000 miles of track by the end of the period (Stover 1999). As early as the mid 1850’s, estimate suggest 224,000 men were employed in railroad construction (Fishlow 1965, 410) and by the time of the 1880 census, more than 418,000 men were in regular employment on the railroads and the figure grew rapidly thereafter (U.S. Census Office 1883). Secondly, in terms of digital data resources, the US National Historical GIS now provides both a wealth of decennial census data by county and state from 1790 onwards and a comprehensive set of county boundaries files in GIS format (Minnesota Population Center 2004). However, equivalent publicly available data sets for industrial activity and transportation infrastructure (roads, railroads and canals) across the country are noticeable by their absence. This limits the ability of scholars to address long-standing questions about the relationships between transportation and economic/urban development (Atack et al. 2010, Atack and Margo, this vol-

• 197

Railroads in historical context: construction, costs and consequences

ume). It also limits the opportunity for examining whether these relationships are similar in different locations or whether varied factors have an impact on the trajectories of regional growth. Thirdly, a very accurate and detailed GIS of railroad development can be used for both static digital cartographic representations and dynamic visualizations at a range of scales, assuming appropriate data generalization is undertaken. In future, as more digital datasets become available at both national and regional levels, the combination of these data layers with information on expanding transportation infrastructure will provide greatly enhanced opportunities for new ‘views’ of economic development and geographical change.

In the first part of this paper both general project aims and specific objectives are examined, before a brief chronology of project developments is given. This reflects the changing capabilities of software technology and the growing availability of scanned map resources over the period since the project commenced.

Later sections examine the range of data resources required for the project and provide an overview of the methodology currently being employed in the project.

198 •

GENERAL PROJECT AIMS The overall aim is to develop a comprehensive GIS of the development of a key portion of the US railroad system in the 19th century, key because it covers the main industrial areas and many of the important coal mining regions during the period. The sequence of coverage is to focus initially on the Northeast outside New England, starting with many of the Middle Atlantic States (Delaware (DE), Maryland (MD), New Jersey (NJ), New York (NY) and Pennsylvania (PA)), followed by Ohio (OH), Virginia (VA) and West Virginia (WV). These last two were both part of the earlier, larger state of Virginia until 1863. Later stages will expand coverage into New England and the remainder of the Manufacturing Belt (De Geer 1927). The latter region extended as far west as St. Louis and Minneapolis, encompassing the major railroad centre of Chicago.

This GIS should be a very high quality dataset that can be used for a wide variety of research purposes, initially by our own research group and our collaborators, but will subsequently by made publicly available to bona fide researchers in a staged fashion.

The GIS should provide data of sufficient accuracy to be usable from the national down to the sub-township scale (though ensuring railroad tracks are

Richard Healey • Michael Johns

always on the ‘right’ side of the river is not feasible in a dataset covering such a large area).

The dataset should meet the most rigorous scholarly standards, in terms of trace-ability of data back to the most reliable cartographic and textual sources, while recognizing that 19th century source materials are not always complete and sometimes are inconsistent or inaccurate in the information they provide (especially printed cartographic sources).

The coverage of the dataset should have three components : • The evolving physical network ‘on the ground’ • The chronologies of ownership of railroad lines • The chronologies of railroad corporate takeover, merger and leasing The system should also provide a well-defined means of tracing the complex interrelationships between the three types of data.

SPECIFIC PROJECT OBJECTIVES • To undertake digital mapping at the most detailed scale available for comprehensive historical mapping (i.e. approximately 1 inch to the mile – the same as the UK Cobb railway atlas (Cobb 2006)) • To include both railroad lines and depots on those lines • To utilize and cross-check a wide variety of data sources about railroad construction, ownership and control, with preference given to legal and company sources, such as incorporation records, charters and annual reports • To facilitate eventual web dissemination of data via shapefiles and specialized querying services, either from the University of Portsmouth or collaborators’ websites or both.

HISTORY OF THE PROJECT: PHASE I OVERVIEW Phase I of the railroad component of the overall project was undertaken 19961999, with funding support from the University of Portsmouth, although many

• 199

Railroads in historical context: construction, costs and consequences

200 •

of the research materials had previously been collected by one of the present authors, as part of his PhD work. The original methodology and preliminary findings from Phase I were subsequently reported in the relevant literature (Healey and Stamp 2000).

It should be remembered that at this time, there were very few web resources in terms of scanned printed works or maps and the user interfaces of both GIS and database software tools were relatively primitive compared to those available now (in 2012).

Nevertheless, the required analytical functionality was available in these software tools, as was the basic interfacing technology between them, though web-based capabilities were largely non-existent.

From this initial resource base, several key stages of the work were successfully completed. These included the following: • Establishment of an extensive archive of copies of paper maps and bibliographic sources • Design and implementation of an ORACLE database to hold information on railroad companies, their corporate control ‘family trees’ and their ownership of specific lines.

• Implementation of an ORACLE screen forms data entry system for the above database • Creation of digital map files, in both coverage and shapefile format at 1;100,000 scale for rail lines in the Middle Atlantic states. This work was undertaken using ARC/INFO. • The chosen scale was the only feasible scale to use at the time, based on the US Geological Survey Digital Line Graph (DLG) files, which contained surviving railroad lines as of about 1970. Earlier railroads, identified on paper sources as being in operation prior to 1900 were manually added back into these files, using the DLG and populated places data as a digital backdrop. This approach was more than adequate for regional scale mapping purposes, but rather less so for detailed work at the sub-county level.

In the period 2000-2007 there was little new development of the resource, but it was utilised to varying degrees as part of studies on other areas of the overall project, namely the anthracite coal industry, the Pennsylvania iron industry 1825-1875 (with Anne Knowles of Middlebury College) and the early US oil industry 1859-1879.

Richard Healey • Michael Johns

HISTORY OF THE PROJECT: PHASE I OVERVIEW In the course of these other studies, it became clear that there were a number of limitations to the data and variability in data quality, both over space and time, as a result of lack of access to large-scale historical mapping in scanned form.

The availability of funding on a railroad-related topic from the UK Economic and Social Research Council (ESRC) in 2008-2009, followed immediately by further funding from the first round of the JISC/NEH Digging into Data Challenge 2010-2012 (in collaboration with Will Thomas and Ian Cottingham of the University of Nebraska, Lincoln) provided an important opportunity to revisit and develop the methodology and the GIS itself, in the light of new technology and the rapid expansion of on-line digital resources. The ESRC Project was entitled ‘Railroads and Migration of Heavy Industrial Workers in the Northeast USA 1850-1900’, while the JISC/NEH project was entitled ‘Railroads and the Making of Modern America - Tools for Spatio-Temporal Correlation, Analysis, and Visualization’.

As a result of these two projects, the following methodological developments became possible: • Scanned USGS historical 15 and 30 minute topographical sheets at approximately 1 inch to mile were assembled from a variety of sources, georeferenced and used as raster backdrops for both new digitizing of selected lines and ‘positional accuracy improvement’, to borrow a phrase for the UK Ordnance Survey, of the shapefiles derived from Phase I work.

• ORACLE screen data entry forms were enhanced and converted to allow web-based access to the database • Additional database infrastructure was added to allow for systematic treatment of depot chronologies, as this had not formed part of Phase I. • Accuracy and traceability of the data were given the highest possible priority, at the expense of geographical and temporal coverage, if need be, as both projects required that data would eventually be destined for wider use within the UK research community (in the case of the ESRC project) or by the public at large, in the case of the JISC/NEH project.

• To this end systematic procedures involving multiple cross-checks for both database chronologies and GIS data were put in place.

From a substantive viewpoint, it was decided to analyse several of the major eastern trunk line systems, both because of their importance to the US railroad system as a whole and because their expanded network territories by 1900 covered major portions of the region of interest. The three trunk line systems chosen were the Baltimore and Ohio, the Erie and the

• 201

Railroads in historical context: construction, costs and consequences

Pennsylvania. These systems have been the focus of development work since 2008.

DATA SOURCES

202 •

Since the new millennium dawned, there has been a veritable explosion of online resources, both in terms of scanned books and periodicals, and scanned maps. Book scanning has obviously been led by the Google Books initiative (books.google.com) and has been greatly enhanced since 2008 by the determined efforts of the Hathi Trust (www.hathitrust.org). In terms of map scanning, an early and immensely valuable collection of scanned USGS historical topographic maps of states in the Northeast USA was provided by Mytopo (http://historical. mytopo.com/), without which Phase II of this project could not have been taken forward at the time funding became available. This resource has recently been superseded by the historical topographic map collection of the USGS National Map Project (http://nationalmap.gov/historical/), whose downloaded maps will form the basis of future extensions to Phase II work. Another key resource is the extensive collection of scanned railroad route maps available from the American Memory Project at the Library of Congress (http://memory.loc.gov/ammem/index.html). However, it should not be assumed that problems of access to material have now been solved, as very large quantities of printed material relating to 19th century railroads have not yet been scanned, especially early publications and reports, and the same applies to thousands of railroad maps at a variety of scales.

For the foreseeable future, therefore, access to major collections of railroad publications in the United States remains essential. These include Northwestern University; the University of Missouri, St. Louis; the Hagley Library (Delaware); University of Nebraska, Lincoln; New York Public Library and the Library of Congress. In the UK, the London School of Economics, the British Library and the University of Portsmouth now have some of the largest holdings of U.S. railroad publications outside the USA. A wide range of textual sources are therefore being used in the project, and one of its most important features is that all types of information included in the database, down to individual dates in many cases, are referenced back to the specific source materials from which they were taken. This provides the required level of traceability, which is particularly important in resolving the numerous cases where multiple sources disagree or contain overlapping but incomplete data. Inconsistency is not limited to textual materials, but is also common in

Richard Healey • Michael Johns

cartographic sources. For example, compilers may draw railroad lines under construction as though they were already completed, short lines may be omitted and selective inclusion of depots may be undertaken, owing to shortage of space for labelling when the map is small scale.

The major types of textual sources are listed below, to give an idea of the breadth of evidence utilised in the project: • Railroad Annual Reports (hundreds of companies, up to 70+ years of record in the 19th century) • State Railroad Commission Reports • Archival Company Records (some of the largest manuscript collections in the USA are railroad-related) • Railroad Corporate Histories –some of these can be voluminous • State and Federal Government Reports • Reports of Stockholders’ Investigation Committees etc.

• Newspapers and trade journals (e.g. American Railroad Journal; Iron Age; Pottsville Miners’ Journal) • ‘Official Railway Guides’ and Poor’s Manuals • Timetables and Pamphlets • The very large secondary literature on general railroad history and the history of specific lines – this literature has to be used with caution as not all information found therein is properly referenced • Census reports on railroads • Maps and historic photographs

PROJECT PROGRESS TO DATE (JULY 2012) It was always known that development work to the levels of accuracy and completeness required for the project would be extremely resource intensive, but the intention was (and is) that the railroad network GIS should not therefore need to be revisited by future scholars, except where a ground survey level of accuracy is required in specific locations (e.g. for legal purposes). Depots present a slightly different problem, however, and these will be examined briefly below.

It is also necessary to be realistic about available resources and the need to provide data for wider use on a reasonable timescale. To this end, two levels of data provision have been identified, basic and comprehensive. Basic or level 1 provision implies tracing of network development for a large system backwards from 1900 to the point when earlier short lines came under the control, by dif-

• 203

Railroads in historical context: construction, costs and consequences

ferent means, of the larger corporation. Comprehensive or level 2 provision implies tracing of all aspects of all predecessor lines including their construction and takeover chronologies prior to absorption into the larger system. Further to this, there are two levels of depot provision. Level 1 refers to identification of all depots as of 1900. This implies that depots abandoned before that time will not be included. Level 2 means that a detailed chronology of depots along the system in question has been undertaken (but see below).

At the time of writing, the Baltimore and Ohio system dataset has reached level 2 on both criteria, the Erie is on levels 2 and 1 respectively, while the highly complex Pennsylvania system is almost complete to levels 1 and 1. In all cases, the datasets currently extend from the Atlantic coast to the western border of Ohio. Further extension in the direction of Chicago and St. Louis is planned now the USGS historical map coverage is largely complete. Examples of the resulting shapefiles of Baltimore and Ohio Railroad network expansion for decade intervals can be seen at http://railroads.unl.edu/views/item/borr.

THE PROBLEM OF DEPOTS 204 •

It is appropriate to draw specific attention to the problem of identifying and tracking depots across a country as large as the United States. The exact number ever constructed is unknown but is likely to be well in excess of 50,000. For example, just on the three systems described above, a total of approximately 6,000 depots have already been located.

Aside from the difficulties associated with the sheer number of depots, there are a number of further problems to their incorporation in the GIS, as follows: • They do not appear on the USGS historical 15 or 30-minute quad maps (with hindsight, this was a serious error of judgement on the part of the early cartographers!)

• They do not always appear even on the town insets in historical county atlas maps (and it would be an enormous undertaking to go through all these sources) • Name changes are surprisingly common • Their construction/rebuilding/change of function is not systematically or comprehensively recorded in company annual reports • Neither any historical maps nor even the official railway guides can be relied on to report all depots in use at specific dates Under these circumstances, the best approach is to undertake systematic com-

Richard Healey • Michael Johns

parison of multiple dated textual and map sources, but this is inordinately time consuming and still does not always yield findings that are accurate to more than +/- 2 years for dates, in our experience. Further to this, there are often difficulties in identifying the exact location on the ground of early depots to within +/- 1 mile. It may be an example of a problem where the local knowledge and enthusiasm of the very large rail fan community could usefully be engaged to provide ‘volunteered’ or ‘cloud-sourced’ information. That said, there would remain a very substantial amount of work in quality control of incoming information, before a usable dataset could result.

CONCLUSIONS This project indicates that while enormous strides have been made in terms of making library resources available electronically over the internet, this does not make difficult tasks, such as the construction of an accurate railroad GIS, straightforward. However, these developments do make work that was previously infeasible on logistical grounds, potentially feasible, provided it is conducted in a staged manner and the necessary resources are available.

Building on earlier work, the present project has made initial progress towards longer term aims for a railroad GIS of the American Manufacturing Belt.

Several datasets for eastern trunk line systems are approaching the point where data can be made publicly available. They are also in the process of being utilised in spatio-temporal visualisation work, which forms part of the results of Round 1 of the Digging into Data Initiative.

ACKNOWLEDGEMENTS The authors would like to thank the ESRC for grant RES-000-22-2420 ‘Migration, Economic Opportunity and the Railroads: Movement of Heavy Industrial Workers in the North-East USA 1850-1900’ and the NEH/JISC Digging into Data Initiative grant ‘Railroads and the Making of Modern America: Tools for Spatio-Temporal Correlation, Analysis, and Visualization’ that supported part of the work described in this paper. Additional research support was provided by the University of Portsmouth. We are indebted to the Pennsylvania State University for providing a set of scanned historical quad maps of the state.

• 205

Railroads in historical context: construction, costs and consequences

REFERENCES Atack, J., Bateman, F., Haines, M. and Margo, R.A. (2010) Did railroads induce or follow economic growth?: urbanization and population growth in the American Midwest, 1850– 1860. Social Science History, 34: 171-197.

Cobb, M.H. (2006) The Railways of Great Britain, a Historical Atlas : At a Scale of 1 Inch to 1 Mile. 2nd Ed. Shepperton, Surrey : Ian Allan Publishing.

De Geer, S. (1927) The American Manufacturing Belt. Geografiska Annaler, 9: 233-359.

Fishlow, A. (1965) American Railroads and the Transformation of the Ante-Bellum Economy.

Cambridge, MA: Harvard University Press.

206 •

Healey, R. and Stamp, T. (2000) Historical GIS as a foundation for the analysis of regional economic growth: theoretical, methodological and practical issues. Social Science History, 24 (3): 575-612.

Minnesota Population Center (2004) National Historical Geographic Information System (NHGIS): Pre-release Version 0.1. Minneapolis, Minnesota: University of Minnesota. (http://www.nhgis.org).

Stover, J.F. (1999). The Routledge Historical Atlas of the American Railroads. New York: Routledge.

U.S. Census Office (1883) Vol. 4, Report on the Agencies of Transportation in the United States including the Statistics of Railroads, Steam Navigation, Canals, Telegraphs, and Telephones. Washington: GPO.

Robert M. Schwartz

GLOBALIZATION, RAILWAYS, AND AGRICULTURE IN FRANCE AND GREAT BRITAIN, 1850 TO 1914 GLOBALIZAÇÃO, CAMINHOS DE FERRO E AGRICULTURA EM FRANÇA E NO REINO UNIDO, 1850 A 1914 Robert M. Schwartz (U. Vanderbilt, USA EUA) Thomas Thevenin (U. Boston, USA EUA) Robert M. Schwartz is E. Nevius Rodman Professor of History Mount Holyoke College, South Hadley, Massachusetts. U.S.A. Thomas Thevenin (University of Boston; previously University of Burgundy and THEMA / CNRS, France) research is about transportation systems with an historical perspective.

Robert Schwartz é E. Nevius Rodman Professor de História, no Mount Holyoke College, Massachusetts, EUA. Especialista em história ambiental da Europa, desde o século XVII até agora.

Thomas Thevenin (Universidade de Boston, E.U.A.; anteriormente Universidade de Burgundy e THEMA / CNRS, França) investiga sistemas de transportes numa perpsetiva historica.

Abstract Resumo

Powered by steamships, railways, and telegraphy, the pace and extent of globalization grew dramatically from the 1850s to the Great Depression. This was especially true in the realm of agriculture, when a huge increase in the production and export of American wheat and beef generated a long agrarian crisis in Europe (1876-1896) as prices fell sharply, forcing farmers, landlords, and governments to adjust to intensifying competition in foodstuffs. Meanwhile, however, railways helped create commercial markets on the industrial scale, sometimes by investing in port facilities, sometimes merely by offering rapid transport to the huge consumer markets of London and Paris. GIS and spatial analysis, will help demonstrate the geography and growth of some commercial markets in Great Britain and France.

O ritmo e a dimensão da globalização cresceram dramaticamente entre 1850 e a Grande Depressão, especialmente no que respeita à agricultura, em que um enorme aumento das exportações de carne e de trigo Americanos criaram uma longa crise agraria na Europa (1876-1896). Or preços cairam e obrigaram os

• 207

Railroads in historical context: construction, costs and consequences

lavradores e os proprietaries a ajustarem-se a uma intense competição na produção alimentar. Entretanto os caminhos de ferro ajudaram a criar mercados comerciais com escala industrial, por vezes investindo em facilidades portuárias, outras vezes pela oferta de transporter rápido para os enormes mercados de Londres e Paris. Com a ajuda de SIG e de análise especial, demonstra-se a geografua e o crescimento de alguns mercados comerciais no Reino Unido e em França

208 •

Robert M. Schwartz

Globalization, Railways, and Agriculture in France and Great Britain, 1850 to 1914 Robert M. Schwartz and Thomas Thevenin This article is based on research supported by a Collaborative Research Grant from the National Endowment for the Humanities (RA-50577-060). We would like to thank the following for their support in producing the GIS data on British railways from M. H. Cobb’s remarkable atlas ofThe Railways of Great Britain, a Historical Atlas, 2 vols. (Shepperton: Ian Allan, 2005): Meritxell Gallart, Mateu Morillas and J. Marti-Dominguez. This work was part of a European Science Foundation initiative (Eurocores) and the project within it, “Water, Road and Rail: The Development of European Waterways, Road and Rail Infrastructures: A Geographical Information System for the History of European Integration (18252005), directed by Jordi Marti Henneberg, Inventing Europe grant FP-005. The GIS data on the French rail system have been constructed with the help of Mathilde Pizzuto with funds of the NEH Grant, and Loic Sapet with grant funds from the French National Agency of Research (ANR 07 Corp 019). Much of the data for Britain were taken from the Great Britain Historical GIS (GBHGIS, http://www.gbhgis.org), and the cantonal boundaries were kindly supplied by G. William Skinner of the University of California, Davis. Schwartz thanks students for their assistance: Jacinta Edebeli, and Ayla Ben-Chaim, Kirsten Hansen, and Morgan Wilson for help in preparing the GIS data on Dorset County agriculture. Special thanks go to our collaborator, Ian Gregory and for his indispensible help over the years. For this part of our project, Ian provided the standardized parish units and associated population data (1861-1911) and the digital photographs he made of the Agricultural Census Returns for Dorset County at the National Archives, Kew; Thomas Thevenin provided the georeferenced communal boundaries and settlement centers and supervised the work of Mathilde Pizzuto and Loic Sapet who digitized the rail lines and stations for France.

Losses year after year and increasing competition indicate that the crops now grown are not sufficient to support the farmer. When he endeavours, however, to vary his method of culture, and to introduce something new, he is met at the outset by two great difficulties. ... The first [is] the extraordinary tithe ...; the second is really even more important—it is the deficiency of transit.1

1 Richard Jefferies, “Steam on Country Roads,” in Field and Hedgerow, being the last essays of Richard Jefferies (London: Longmans, Green, and Co., 1889): 238, 231 respectively. This was a posthumously published collection of his essays of the 1880s. Jefferies died in 1887. See Eric L. Jones, “The Land that Richard Jefferies Inherited,” Rural History 16, no. 1 (2005); P.J. Perry, “An Agricultural Journalist on the “Great Depression”: Richard Jefferies,” The Journal of British Studies 9, no. 2 (1970).

• 209

Railroads in historical context: construction, costs and consequences

It is not too much to say that three parts of England are quite as much in need of opening up as the backwoods of America. When a new railroad track is pushed over [American] prairie and through primeval woods, settlements spring up beside it. When road trains [in Britain] run through remote hamlets, those remote hamlets will awake to a new life.

Richard Jefferies, “Steam on Country Roads,” 1884

210 •

After reflecting on American agriculture and railroads, Richard Jefferies, an agricultural journalist, saw one thing clearly: Britain must catch up. Goods trains in agrarian American, he wrote, stop not merely at stations but virtually anywhere along the line where there was grain and produce to pick up. The British farmer, alas, enjoys no such convenience. To get crops and produce to market is a struggle. First, he must cart them to a railway station—a slow journey of up to 10 miles; then, at the station, he faced a long wait, eventually surrendering “to the middleman to get his goods to market.”2 British trains went from town to town, but they needed to go to the farms and the crops.

Road trains, he argued, were the solution. These redesigned steam-powered trains would run not along rails but on country roads, stopping at each farm and “loading at the gate of the field”3 Railways, he granted, would still be essential for long-haul shipments, but the road trains would bring much desired change. With speedy transit at hand, farmers, he continued, will plant perishable fruits and vegetables on unused plots, the rural population will grow, and British farmers will recapture revenue now going to the Continent and America for imports. To break open rural isolation, daily road trains for passengers would connect villages with market towns. Remote hamlets would spring to life.

Casting his eye across the Channel at old rival France was no consolation.

France was moving ahead of Britain, too.

We have lately seen the French devote an enormous sum to the laying down of rails in agricultural districts, to the making of canals, and generally to the improvement of internal communication in provinces but thinly populated. The industrious French have recognized that old countries, whose area is limited, can only compete with America, whose area is almost unlimited, by rendering transit easy and cheap. We in England shall ultimately have to apply the same fact.4 Jefferies’ lament takes us back to a period of accelerating globalization in food stuffs, to crisis and adjustment in the international division of food produc2 Ibid, 231.

3 Ibid, 236.

4 Ibid., 239.

Robert M. Schwartz

tion, and it sets the scene for something new: a comparative spatial history that bridges the gap between two research areas typically treated in isolation from one another, one on railways and the other on agriculture. What we discover is a better understanding change over space and time between rail transport and agricultural production. Although rural rail service was a boon to farming by opening distant urban markets, it also pinched farmers where it hurt, bringing intensifying international competition in foodstuffs to the farm gate. Still, even as competition grew and the agrarian depression of 1880s and 1890s struck agrarian economies, accessible rail transport often helped farmers adapt to new market conditions of the globalizing world of the late nineteenth century. Jefferies was unable to see this even though he accurately depicted the general crisis of confidence in European farming.

HISTORICAL GIS AND SPATIAL HISTORY Farmers of the period knew very well that their fortunes increasingly depended upon railways and their freight charges. Today, few scholars doubt that railways and agriculture were linked and interdependent, and yet historians concern themselves, almost exclusively, with one or the other subject. Rare exceptions to this offer valuable insights, which we can improve upon in several ways. GIS and spatial analysis make it possible to study larger and more complex bodies of evidence at different scales and over time. Here, our geo-referenced evidence comes from large databases on railways, population, and agriculture for Great Britain and France from the 1830s to 1930s. Another improvement is our use of a comparative approach to investigate patterns of change within and between states, the better to identify and explain both similarities and differences in countries that had differing political economies, a difference reflected in agricultural policy by British free trade and French protectionism. In this period of globalizing markets comparative history is all but indispensible for understanding the position of any geographical area and its producers in its relation to the shifting international division of labor -a need underscored by its absence in much of the literature on the agrarian depression of the late nineteenth.5 Among historians of British agriculture there is a consensus that the depression in Britain was not a “general crisis” in agricultural output but one that 5 There are several notable exceptions that specifically apply a comparative approach in the study of agriculture and agricultural policies. Niek Koning, The failure of agrarian capitalism : agrarian politics in the UK, Germany, the Netherlands and the USA, 1846-1919 (London: Routledge, 1994); Michael Tracy, Government and agriculture in Western Europe, 1880-1988, 3rd ed. (New York ; London: Harvester Wheatsheaf, 1989); Kevin H. O’Rourke, “The European Grain Invasion, 1870-1913,” The Journal of Economic History 57, no. 4 (1997), and J. L. van Zanden “The Growth of Production and Productivity in European Agriculture, 1870-1914,” The Economic History Review 44, no. 2 (1991): 215-39.

• 211

Railroads in historical context: construction, costs and consequences

varied by region and that struck the cereal growing regions of the south and southeast much harder than elsewhere in England and Wales. Debate continues, however, as to whether or not British agriculture “failed” to meet the challenges of intensifying foreign competition. “Pessimists” point to the demise of large, more productive farms, a lack of innovation and entrepreneurial savvy, and the government’s complacent dependence on imports from the bountiful agricultural resources of the U.S. and Britain’s colonies.6 As more regional research is undertaken, “optimists” argue that resilience, not failure, characterized English farming in difficult circumstances.7 Whether optimist or not, the role of rural rail transport in response to the agrarian depression is in this literature usually absent or mentioned only in passing.8 The same is true in research on French agriculture in the second half of the 19th century. By and large, studies of agricultural performance and the depression in particular concern themselves with the national level alone, and studies of specific regions are only beginning to appear.9 Meanwhile, debate over French agriculture echoes that over British farming. French “pessimists” marshal evidence old and new to demonstrate that French agriculture lagged behind Britain and most of Western Europe.10 Optimists respond with new data and argument 212 •

6 E. J. T. Collins, “The Great Depression, 1875-1896,” in The Agrarian History of England and Wales.

Volume VII1850-1914, ed. E. J. T. Collins (Cambridge: Cambridge University Press, 2000), 138207; Richard Perren, Agriculture in depression, 1870-1940, New studies in economic and social history (Cambridge: University Press, 1995).

7 E.H Hunt and S. J. Pam, “Prices and Structural Response in English Agriculture, 1873-1896,” The Economic History Review 50, no. 3 (1997): 477-505; Bethanie Afton, “The Great Agricultural Depression on the English Chalklands: The Hampshire Experience,” Agricultural History Review 44(1996): 191-205; Michael Turner, “Output and Prices in UK Agriculture, 1867-1914, and the Great Agricultural Depression Reconsidered,” The Agricultural History Review 40, no. 1 (1992): 38-51; Michael E. Turner, “Agricultural Output, Income, and Productivity,” in The Agrarian History of England and Wales. Volume VII1850-1914, ed. E. J. T. Collins, The Agrarian History of England and Wales (Cambridge: Cambridge University Press, 2000); van Zanden “The Growth of Production and Productivity in European Agriculture, 1870-1914.”

8 See for example the limited treatment of railways in E. J. T. Collins, ed. The Agrarian History of England and Wales. Volume VII1850-1914, The Agrarian History of England and Wales (Cambridge: Cambridge University Press,2000). Important exceptions include Christine Hallas, The Wensleydale Railway (Clapham [Bedfordshire]: Dalesman Books, 1984); D.W. Howell, “The impact of railways on agricultural development in nineteenth-century Wales,” Welsh History Review 7(1974-5); Christine Hallas, “The Social and Economic Impact of Rural Railway: The Wensleydale Line,” The Agricultural History Review 34(1986); and David Turnock, An historical geography of railways in Great Britain and Ireland. Aldershot, Hants, England, Brookfield, Vt: Ashgate, 1998.

9 Christophe Bouneau, “Chemins de fer et developpement rural en France de 1852 a 1937 : la contribution de la Compagnie du Midi “ Histoire, Economie et Societe 1(1990); Hugh Clout, “The Pays de Bray: A Vale of Dairies in Northern France,” Agricultural History Review 51, no. 2 (2003) : 1990-208.

10 Jean-Pierre Dormois, “La ‘vocation agricole de la France’. L’agriculture francaise face a la concurrence britannique avant la guerre de 1914,” Histoire etMesure 11, no. 3-4 (1996): 329-66; Albert Broder, “La longue stagnation francaise: panorama general,” in La Longue stagnation en France. L’autre grande depression 1873-1897, ed. Yves; Broder Breton, Alber; Lutfalla, Michel, eds. (Paris: Economica, 1997); Jean-Francois Vidal, Depression etretour de la prosperity. Les

Robert M. Schwartz

that the French system of small-farming was more rational and productive than commonly thought.11 Within France itself, a long-held generalization holds that in agriculture—as in industry—the country was divided between the developed north and the less developed south. On the issue of regional disparities new opportunities for comparative spatial history abound, thanks in part Jean-Claude Toutain’s work on regional variations in productivity growth from 1810-1990.12 One major finding was that north-south disparities narrowed after 1860s and that growth rates in the two regions converged at the end of the ninetieth and early twentieth centuries, owing in large part to the increased productivity of the wine and market-gardening sectors in the south. Toutain’s data and argument bring welcome attention to the issue of agricultural restructuring after 1860 and renews debate. One recent article, for example argues, rather unpersuasively, that regional specialization of the kind that developed in Britain was largely absent in France from 1870 to 1914.13 In fact, the issue calls out for further research. In our larger work we answer the call, showing that the geographic restructuring of French agriculture was much facilitated by railway expansion. Although we do not pursue the broader patterns here, the analysis below of the Cote d’Or Department in Burgundy illustrates our approach.

With our problem in its historiographical frame, we can now consider tools and methods. How can our questions about spatial relationships and changes over time be systematically addressed? Time was when studying the influence of proximity in social relations was a hard row to hoe, and one had to limit either the size of the study area or the sample of data. Today, GIS and geographicallyreferenced data reduce these previous constraints and open new possibilities in spatial analysis using visualization, cartography, and spatial statistics.14 In this economieseuropeenes a la fin du XIXe siecle (Paris: L’Harmattan, 2000); J. L. van Zanden “The Growth of Production and Productivity in European Agriculture, 1870-1914.”

11 Juan Carmona, “Sharecropping and livestock specialization in France, 1830-1930,” Continuity and Change 21(2006): 235-59. Andre; Verley Straus, Patrick, “L’economie francaise au XIXe siecle.

Analyse macro-economique, une oeuvre isolee ou une ouverture vers des recherches novatrices ?”, Revue d’histoire du XIXe siecle 23(2001): 14; Vidal, Depression et retour de la prosperity. Les economieseuropeenes a la fin du XIXe siecle ; Michael Hau, “La resistance des regions d’agriculture intensive aux crises de la fin du XIXe siecle : Les cas de l’Alsace, du Vaucluse et du BasLanguedoc,” in Un siecle d’histoire agricole firanqaise. Actes du colloque de la Societe frangaise d’economie rurale, ed. Chalmin, Philippe and Gueslin, Andre (Economie rurale, 1988): 31-41; Giinther Schmitt, “Agriculture in Nineteenth Century France and Britain: Another Explanation of International and Intersectoral Productivity Differences,” The Journal of European Economic History 19, no. 1 (1990).

12 Jean-Claude Toutain, La production agricole de la France de 1810 a 1990: Departements et Regions. Croissance, productivity, structures, 3 vols., Histoire quantitative de l’economie francaise (Grenoble: Cahiers de 1’ISMEA [Economies et societes], 1992-3); “La Croissance inegales des regions francaises: l’agriculture de 1810-1990,” Revue historique 590(1994): 315-59.

13 Dormois, “La ‘vocation agricole de la France’. L’agriculture francaise face a la concurrence britannique avant la guerre de 1914,” 358.

14 A good, brief introduction to historical GIS is Ian N. Gregory, A Place in History: A Guide to Using GIS in Historical Research (Oxbow: Oxford 2003). For a fuller account see Ian Gregory and Paul Ell, Historical GIS: Technology, methodology and scholarship (Cambridge, England: Cambridge University Press, 2008). The most convenient survey of the GIS field in its variety and techniques

• 213

Railroads in historical context: construction, costs and consequences

case encoding the geographic coordinates in each unit of analysis makes it possible to calculate many different aspects of distance, proximity, accessibility, and transport cost when joined with GIS data on the development of railways and rail stations from the 1830s to 1930s. Using geo-referenced information on agricultural production and land use attached to British counties, registrations districts, and parishes and to the corresponding units of French administration— departments, cantons, and communes—gives us comparable data at these several scales of geographic resolution.

Now we turn to specific questions. Which communities in a given rural area were 10 miles or further from a railway station, the condition Jefferies characterized as lamentable? Over the years which villages continued to fall into the “distant” category as opposed to those that, with rail expansion, came to be “near” a station, having five miles or fewer to get their crops to a shipping point? Further, how was proximity to rail transport related to change in the use of agricultural land, to the shift from arable farming to livestock and dairy farming? The combination of GIS and spatial analysis bring the examination of these complexities within reach.15

214 •

BACK TO THE STORY: THE AGRARIAN DEPRESSION AND THE RAILWAY SYSTEMS OF BRITAIN AND FRANCE The Depression Many of his contemporaries agreed with Jefferies’s concern with the inadequacies of Britain’s rural rail transport services. British services were woefully outmatched by those in the United States and might be overtaken by those in France as well. This insufficiency seriously undermined the British farmer’s ability to survive the agricultural depression and withstand intensifying international competition in food stuffs from America.16 The signs of difficulties is the collection of articles by leading specialists in A. Stewart Fotheringham and John P. Wilson, The handbook of geographic information science, Blackwell companions to geography (Maiden, MA: Blackwell Pub., 2008).

15 The following discussion draws on research I’ve presented in previous conference papers and several recent articles, including “New Tools for Clio: GIS, Railways, and Change over Time and Space in France and Great Britain, 1840-1914,” 2007, a digital publication (University of Nebraska and University of Illinois Press). http://www.mtholyoke.edu/courses/rschwart/railways/; “Rail Transport, Agrarian Crisis, and the Restructuring of Agriculture: France and Great Britain Confront Globalization, 1860-1900, Social Science History 34.2 (2010): 229-55; and “Spatial History: Railways, Uneven Development, and a Crisis of Globalization in France and Great Britain,” Journal of Interdisciplinary History (under review), with co-authors Ian N. Gregory and Thomas Thevenin.

16 A brief introduction to the subject is Peter A. Coclanis, “Back to the Future: The Globalization of Agriculture in Historical Context,” SAIS Review 23, no. 1 (2003): 71-84. For Britain, see Collins, “The Great Depression, 1875-1896.” For France, Jean Lhomme, “La crise agricole a la fin du XIXe

Robert M. Schwartz

emerged in the mid 1870s when a series of cool and rainy summers led to bad harvests and cattle diseases that reached a crisis point in 1879. In the same period the first wave of American grain exports arrived in Britain and other European counties, forcing the price of wheat in particular to lower and lower levels until a mild recovery began in the mid-1890s. From 1873 to 1882, American exports of wheat rose from 40 to 150 million bushels, displacing Russia as the chief exporter of cereal grains while the largest share came to Britain.17 Well before then English interest in American agriculture produced an outpouring of articles and reports, a fair number having been written by authors who had observed American farming first hand. The best known of them was James Caird, a Member of Parliament and the main force behind the establishment in 1866 of the annual collection of British agricultural statistics.18 Touring America in 1858, he described the Middle West “as the greatest track of fertile land on the globe.”19 In 1881 a Royal Commission was set up to study the agricultural depression in England and Wales. Recognizing the importance of American imports as one of the causes, it charged one of its members, John Clay, to gather evidence in the United States and report his findings. His report lauded the workings of American wheat production and American rail, calling it at one point “miraculous.”20 Other Europeans from France, Germany, Austria-Hungary, and Russia who came to study the American system agreed.21 In France the Ministry of Agriculture’s interest rose to new heights in 1889, when the agricultural displays at the Universal Exposition in Paris caused astonishment at the prodigious agrarian capacities of the United States and other New World counties. In an 1891 report, the enviable efficiency of the American system was described in some detail. In the wheat trade good yields on enormous acreages, cheap transport, the American system of grain elevators—all worked harmoniously like a gigantic, well-designed machine. The rail system alone was as huge as the country, and its growth was remarkable. With 259,000 kilometers siecle en France. Essai d’interpretation economique et sociale,” Revue economique 21, no. 4 (1970): 521-553. For a comparative perspective, see Koning, The failure of agrarian capitalism; Tracy, Government and agriculture in Western Europe, 1880-1988; and O’Rourke, “The European Grain Invasion, 1870-1913.”

17 Morton Rothstein, “America in the International Rivalry for the British Wheat Market, 18601914,” The Mississippi Valley Historical Review, Vol. 47.3 (1960): 401-418. Harry J. Carman, “English Views of Middle Western Agriculture, 1850-1870,” Agricultural History 8.1 (1934): 3-19.

18 G.E. Fussel, “The Collection of Agricultural Statistics in Great Britain: Its Origin and Evolution Agricultural History, Vol. 18.4(1944): 161-167, p. 162.

19 Quoted in Harry J. Carman, “English Views of Middle Western Agriculture, 1850-1870,” Agricultural History, Vol. 8.1 (1934): 3-19; quotation, p. 4.

20 Supplementary report by Mr. John Clay, Jun., on American Agriculture, showing its influence on that of Great Britain, pp. 705-718, Royal Commission on Agriculture. Reports of the Assistant Commissioners. Southern district of England. Report by Mr. Little on Devon, Cornwall, Dorset and Somerset, (with summary of previous reports.), 1882.

21 Rothstein, “America and the British Wheat Market,” pp.412-16.

• 215

Railroads in historical context: construction, costs and consequences

in operation in 1890, “the United States has more than 19 times as much railway line today as it did 30 years ago.”22 In 1890, the figures for the much smaller countries of France and Great Britain were about 38,000 kilometers and 30,000 kilometers, respectively.23

216 •

Railways and Rural Transport By 1890, railway expansion in England and Wales had preceded further than Jefferies was willing to admit. At the end of that the 1880s there were few rural registration districts—a market town and surrounding parishes—that lacked a station and some connection, however indirect, with the national system. Indeed, rail service began to reach the countryside in the latel850s and 1860s, 20 years before Jefferies wrote “Steam on Country Roads.” Using the historical GIS on British railways and population yields a more precise description in graphical and cartographical displays. After calculating the distance (kilometers) from the center of each parish to the nearest railway station at a given date, a mean of the parish scores is calculated for each of 633 registration districts; then the district means for each date are classified by different levels of district population density and presented in a Figure 1A. The figure shows the pattern of increasing accessibility over the decades: except in the least populated districts, proximity to a railway station continued to increase until the turn of the 20th century, especially so for communities of modest population density (25 to 100 persons per square kilometer). Interestingly, Jefferies took this history so much for granted that he ignored it in his writings.

In his assessment of the British system, a major deficiency was the long distance between the farmer’s field and the railway station, citing a journey of up to 10 miles as not uncommon but regrettably inconvenient and outdated. Among the farmers he consulted, there were, no doubt, a goodly number who complained of this inconvenience. Still, had he travelled through French villages during the same period, he would learn that the complaints of British farmers were small potatoes indeed. In fact, a comparative study of British and French rail networks suggests a more positive story of rural railway development in England and Wales than Jefferies would have us believe. (See Figure IB) In railway development France was a decade or more behind Britain. A county four times larger than England and Wales, France had a good deal more territory over which to lay down rails, to connect major cities and ports, and to reach 22 Ministry of Agriculture, Exposition universelle, Paris, 1889. Rapports du jury international.

Groupe VIII. Agriculture, viticulture et pisciculture (Paris: Imprimerie nationale, 1892) : 99.

The figure on American railroads in 1890 seems to be the figure for 1885, according to a later compilation by the U.S. publication. See Association of American Railroads, and Bureau of Railway Economics, Comparative Railway Statistics of the United States, the United Kingdom, France and Germany for 1900 and 1909 (Washington, D.C, 1911).

23 Figures for Britain and France : Jack Simmons, The Railway in England and Wales, 1830-1914 (Leicester: Leicester University Press , 1978), Appendix 2; Ministere des Travaux Public, Statistique centrale des chemins de ferres frangais au 31 decembre 1932.

Robert M. Schwartz

Figure 1 - Average accessibility of rail transport in Britain and France, 1850 to 1939, A. Great Britain; B. France

5.5

A. Average accessibility in the registration districts of England and Wales.

Distance from District Center to Nearest Station, 1850-1920by District Population Density

5 4.5

Kilometers

4 3.5 3 2.5 2 1.5 1

1850

1860

1870

1880

1890

1900

1910

1920

Decade Sources: Parish boundaries and associated population data from Ian Gregory; rail lines and stations taken from M.H. Cobb, The Railways of Great Britain, a Historical Atlas.

2 vols. (Shepperton: Ian Allan, 2003), as digitized under the direction of Jordi Martì Henneberg, University of Lleida, Spain.

35

B. Average accessibility of rail transport in the cantons of France, 1860 to 1930.

Distance from canton center to nearest station, 1860 to 1930, by cantonal population densitypopulation density

30

Kilometers

25 20 15 10 5 0

1860

1870

1880

1890

1900

1910

1920

1930

Decade Sources: Population figures from the Bulletin des lois de la République française, 1887: 204-48; rail lines and railway stations digitized from Carte des chemins de fer français , SNCF, 1944, Bibliothèque Nationale de France, Ge BB 368.

• 217

Railroads in historical context: construction, costs and consequences

218 •

country towns and the some 30,000 rural communes in which the bulk of its population still lived and worked. Compared to Wales and the English Pennines, the uplands and mountains of the French south, the Pyrenees, and the Alps presented more formidable topographical and financial challenges. Moreover, the French pace of industrialization was relatively slow, agricultural productivity in two thirds of the country was low by British standards, and the nation’s defeat

by Prussia in the war of 1870-71 had been a costly humiliation that siphoned off tax revenues to pay the new German Empire substantial reparations.

In 1878, in the aftermath of defeat and the French state’s desire to catch up with American and Britain, the government of the new Third Republic, much as Jefferies reported and praised, launched a huge project to expand the French rail system into the countryside. Named after its chief proponent, the Minister of Public Works, Charles Freycinet, the program, in addition to the expansion of main lines, included state subsidies to promote the growth of secondary lines designed to serve rural and agrarian communities.24 A decade later, in the 1890s, the projected expansion of “lines of local interest” got underway, and the pace of construction quickened, culminating in the 1920s. Railway accessibility in the relatively vast territory of rural France lagged, accordingly, behind Britain, but the gap continued to narrow after 1870. By 1900 villages in moderately populated cantons (between 25 and 50 persons per square kilometer) were on average within five kilometers of the nearest railway station. (See Figure IB.) Table 1. The Growth of the main and secondary rail networks in France, 1870 to 1930 Years

1870

1880

1890

1900

1910

1912

1928

1930

Main lines (km)

17,707

25,759

34,878

38,261

40,214

40,696

NA

42,400

Local lines (km)

293

2,187

3,515

7,612

15,347

17,653

20,291

20,202

Sources: Ministère des Travaux Public, Statistique centrale des chemins de ferres français au 31 décembre 1932.

France, voies ferrées d’intérêt local, tramways, services subventionnés d’automobiles (Paris: Imprimerie Nationale, 1935, p. 5.); Roget Price, The Modernization of Rural France (London: Hutchinson, 1983), p 25 ; Association of American Railroads, and Bureau of Railway Economics, Comparative Railway Statistics of the United States, the United Kingdom, France and Germany for 1900 and 1909 (Washington, D.C, 1911).

Proximity to Railway Stations in rural Britain and France: Change over time and space Turning to spatial analysis, first of Britain and then of France, we use the GIS data on railways and parishes to map the distance from parish centers to the nearest stations at the end points of three different decades: the 1850s, the 1880s,

24 Francois Caron, Histoire des chemins defer en France, 1740-1883 (Paris: Fayard, 1997), 86, 361-70.

Yasuo Gonjo, “Le ‘Plan Freycinet’, 1878-1882: un aspect de la ‘Grande Depression’ economique en France,” Revue historique 248(1972) : 49-86.

Robert M. Schwartz

and the 1900s.25 As shown in Figure 2, British farmers had less reason to complain in the 1880s than before, and by the first decade of the twentieth century, even less so, for by then there were only a few clusters of parishes where the nearest station was more than five kilometers (3.1 miles) away from the parish center—a good deal closer than the 10 mile isolation point mentioned by Jefferies. In other words, the majority of parishes in 1900 or earlier fell within what one farmer thought a maximum distance: beyond three miles from a station, he remarked, is “agricultural death.”26 The high degree of accessibility existed also in Derbyshire and the midlands and in the south generally. Not surprisingly, in sparsely populated regions accessibility was more of a problem. In the peripheries of the country—in Denbighshire and Cardiganshire (Wales), the southwestern counties of Dorset, Devon, and Cornwall, and the northern county of Northumberland—there were numerous parishes where convenient access to rail stations was in doubt. And yet, even in Wales and the southwest, such inconvenience as existed in 1850 had been much reduced by the eve of the Great War. Figure 2 - Proximity of railway stations in the parishes of England and Wales, 1850s to the 1900s.

County boundaries More than 5 Km from station

1850s

• 219

1880s

NORTHUMBERLAND

DENBIGHSHIRE

DERBYSHIRE

CARDIGANSHIRE DEVON CORNWALL

DORSET WILTSHIRE

Source: ibid.

25 The GIS databases we used for this article are described in note 1.

26 Rider Haggard, Rural England, London, 1922 ,1, p. 511, cited by J. T. Coppock, “Agricultural Changes in the Chilterns 1875-1900,” Agricultural History Review 10 (1961), p. 16.

Railroads in historical context: construction, costs and consequences

Proximity to a station, of course, was only one aspect of convenient shipping and passenger travel. Poor station facilities, high shipping rates and ticket prices, infrequent trains, delays, inefficient connections from branch to trunk lines—all produced higher costs and more aggravation for the farmer. In this respect, Jefferies was on the money. But in terms of distance, the accessibility of rural rail service had improved substantially since the late 1860s when Jefferies’ career as an agricultural journalist was beginning.

In France improvement of this kind came later and at the different scale of much larger territory. In the 1860s, when “iron roads” were reaching further into the British countryside and opening remote mining and agricultural districts, the sound of a whistling locomotive was almost unknown in rural France. The major arteries of the national system were in place, but the modernizing benefits of rail transport in agricultural regions, so active in the minds of visionaries and government planners, had yet to materialize in most of the country. As shown in Figure 3, the situation had changed for the better by the 1890s. Thirty years later Figure 3 - Proximity of railway stations in the communes of France, 1860s to the 1920s.

Railways and Agricultural Change

220 •

1860s

1890s

Department boundaries Center of Commune more than 5 Km to nearest station

COTE-D’OR

1920

Source: ibid.; habitational centers of communes and departmental boundaries provided by Thomas Thevenin, Department of Geography, University of Burgundy.

Robert M. Schwartz

in the 1920s the aims of the 1878 Freycinet program of railway expansion came to fruition and the size of the main and secondary networks reached its zenith.

There were regions in the southern uplands and mountains still not well served, but in two-thirds of rural France it was no more than a half day’s walk to catch a train—less than that for horse-drawn wagons and, in the 1920s, even less for combustion-engine automobiles and trucks.

Gauging the benefits of rural railways for agriculture is a more complicated task. In Britain and France, farmers were as convinced as was Jefferies that their increasing losses resulted from the intensified international competition in agricultural products. From 1867 to 1892, wheat acreage in the United States expanded three fold, while in the whole of the United Kingdom from 1872 to 1895, it declined by more than half in response to the falling prices, caused mainly

by American imports that arrived duty-free in open markets. Comparatively speaking, wheat production remained fairly stable in France and increased in Germany—two countries in which tariffs reduced foreign competition, as shown in Table 2.27 Table 2. Wheat acreage in the United Kingdom, France, Germany, and the United States, 1867 to 1895 Years Country UK

1867

1872

1877

1882

1887

1892

1895

1,458,000 1,539,000 1,336,500 1,296,000

972,000

931,500

607,500

France

7,249,500 6,925,500 6,966,000 6,966,000 6,966,000 7,006,500 7,006,500

US

4,991,500 8,464,500 10,651,500 15,025,500 15,228,000 15,633,000 13,770,000

Germany

1,882,500 1,903,500 1,984,500 1,944,000

Source: France, Ministère de l’Agriculture. Statistique Agricole de la France. Résultats généraux de l’enquête décennale de 1892. Paris: Imprimerie nationale, 1897: 94-95. Statistics from Major P.G. Craigie, Director of Statistics of the Board of Agriculture (Great Britain), originating from his from Communication faite au Congrès de l’Institute internationale de statistique (Saint Petersburg, 3 September, 1897).

As the profitability of wheat cultivation declined, British and French cereal farmers, in regions of suitable climate and ecological conditions, looked increasingly to cattle raising and dairy farming to minimize losses, transforming crop27 The figures should be taken as estimates of orders of magnitude. Worked up for presentation at the International Institute of Statistics in 1897 by P. G. Craigie, head of the Statistical Service of the British Board of Agriculture, the figures reflect the improvement in estimates made after 1850, which occurred in step with growing global competition, state interest in agricultural policy and what we now call food security, more accurate and comprehensive statistical collections by individual nation states, and greater European collaboration in collecting and sharing statistics via the International Institute and other bodies. Craigie’s estimates were accepted by the French Ministry of Agriculture as sufficiently accurate to publish them in its report on the state of French agriculture in 1897.

• 221

Railroads in historical context: construction, costs and consequences

land to pasture and reducing their wage bills in the process. Profits from these activities were more likely in the offing because of the rising demand for meat, butter, and fresh milk in cities—a demand enlarged further as workers’ rising real incomes permitted the consumption of higher protein foods.28

222 •

Britain and Dorset County Although railways, steamships, and telegraphy powered the globalization of foodstuffs and the increased competition that struck cereal farmers particularly hard, rail transit was nonetheless a crucial factor in the expansion and intensification of livestock and dairy farming. For dairy farmers in outlying counties such as Wiltshire, Dorset, and Derbyshire, rail transport permitted the shipment of fresh milk to London, Leeds, Manchester, and other cities. Similarly, cattle farmers in outlying counties stood to benefit because they could fatten their stock on site and then ship the animals to market by train, avoiding the traditional and less profitable practice sending store cattle on foot to grazers in fattening regions or to owners of feedlots. Four Welsh counties so affected were Anglesey, Denbighshire, Flintshire, and Pembrokeshire.29 Similar patterns held true for France.30 If railways helped British and French farmers adapt themselves to difficult circumstances during the Agrarian Depression of the 1880s and 1890s, then a review of specific evidence should help confirm or refine the proposition. Selecting two cases from our existing data, we take up the County of Dorset in England, and the Department of Cote-d’Or in France.31 In Dorset we use GIS data at the parish level from the returns of the agricultural census in 1871 and 1901 to map the density of cereal production and cattle at these two dates—the first before the agrarian crisis, the second, after its abatement. In both cases, the decline of 28 P. J. Atkins, “The Growth of London’s Railway Milk Trade, c. 1845-1914,” The Journal of Transport History 4, no. 4 (1978): 208-26. D. Taylor, “London’s Milk Supply, 1850-1900: A Reinterpretation,” Agricultural History 45(1971): 33-38; Olivier Fanica, “Du lait pour la capitale.

La production laitiere autourde Paris (1700-1914),” in Acteurs et espaces de Televage (XVIIe-XXIe siecle). Evolution, structuration, specialisation., ed. Philippe; Moriceau Madeline, Jean-Marc, Bibliotheque d’Histoire Rurale (Rennes: Presses Universitaires de Rennes, 2006); Richard Perren, The meat trade in Britain, 1840-1914, Studies in economic history (London ; Boston: Routledge and Kegan Paul, 1978).

29 Robert M. Schwartz; Ian Gregory, and Jordi Marti-Henneburg, “History and GIS: Railways, Population Change, and Agricultural Development in late nineteenth century Wales,” in Geography and the Humanities, ed. Douglas Richardson (Washington, DC: Association of American Geographers, 2010).

30 For a fuller discussion, see my “Rail Transport, Agrarian Crisis, and the Restructuring of Agriculture: 234-37; Richard Perren, “Marketing of Agricultural Products,” in The Agrarian History of England and Wales. Volume VII1850-1914, ed. E. J. T. Collins (Cambridge: Cambridge University Press, 2000), 254-55; David Turnock, An historical geography of railways in Great Britain and Ireland), 254-55.

31 Although examining Wiltshire would take us to Jefferies’ back yard, the data needed for that county have not yet been added to our database.

Robert M. Schwartz

wheat production stands out clearly. The intensification of dairy and beef-cattle farming, however, is less pronounced than might be expected when the returns for those two dates are used. Nonetheless, the decline in wheat farming was dramatic. In the parishes of central and upland Dorset, wheat acreage estimated in 1871 had fallen by one half or more in 1901, well in line with the national average. In the same period, the density of beef and dairy cattle remained stable overall and increased in two clusters of parishes noted on Figure 4.

Figure 4 - Changes in Cattle Raising and Wheat Farming, Dorset County, 1871 compared to 1901 Cows in 1870

Cows in 1901

Total Cows TCows / Par100A

Total Cereals TCereals / Par100A

0-5 6-9 10 - 16 17 - 23 24 - 57

0-3 4-9 10 - 14 15 - 21 22 - 237

Cereals in 1870

Cereals in 1901 0

5

10

x

20

Sources: National Archives, Kew, MAF 68, Ministry of Agriculture, Fisheries and Food and predecessors: Statistics Divisions: Parish Summaries of Agricultural Returns, Dorset County, 1871 and 1901; Parish boundaries provided by Ian Gregory, University of Lancaster.

In Dorset, we can dig deeper into the decline of wheat growing. There at the parish level, results from a geographically weighted regression model yield estimates of the degree to which change in wheat acreage, the independent variable, are explained by the interacting effects of the mean elevation of parish terrain and the distance from its center (centroid) to the closest railway station.32 The 32 GWR is arguably the tool of choice here and in other situations when the relationships under investigation are likely to vary across a study area, as was true of wheat growing in Dorset.

• 223

Railroads in historical context: construction, costs and consequences

results show that 65 percent of the variation in wheat acreage over the decade 1881 to 1891 can be accounted for by rail accessibility and mean elevation.

The effect of proximate rail service was varied and complex, for it carried a negative or positive influence on changes in wheat acreage, depending upon location and the average elevation of the location. Taking station proximity alone, the main tendency was for wheat acreage to decline slightly as the distance of the nearest stations increased. But when joined with the effect of terrain elevation, that main tendency was inflected. As shown in Figure 5, the influence of station distance was negative in the areas shown in gray and positive in those shown black. From 1881 to 1891, in upland areas, where wheat production had been extensive, wheat acreage tended to increase or remain stable in parishes Figure 5 - Spatially varying relationship between rail-station proximity and change in wheat acreage, 1881-1891, with terrain elevation below.

Rail station Distance to station

224 •

Rail proximity & change in wheat acreage negative -0.21596 - 0.000000 positive +0.000001 - 0.022657 0 3.75 7.5 x 15Kilometers

Terrain Mean elevation (meters) 7 - 61 62 - 82 83 - 120 121 - 205

Robert M. Schwartz

farther from stations; in the lowlands, in contrast, parishes closer to a station showed an increase in acreage compared to lowland parishes farther away. The evident complexity well reminds us that we are dealing with a complicated history of short and long term decisions by farmers of the period in the face of shifting markets, weather, and leasing conditions. This complexity calls for further study and the incorporation of other factors, such as travel distance or travel time along roads—as opposed the straight-line Cartesian distance—numbers of cattle and other competing agricultural activities and crops, the varied farming ecologies of the county, the freight-handling capacity of stations, and so forth.33 What about changes in livestock? Somewhat surprisingly, our analyses—using ordinary least-squares regression, GWR, and other spatial statistics—found no significant effect of rail accessibility on changes in the size of cattle heards over three decades from 1871 to 1891. This suggests—in keeping with the patterns displayed for cows in Figure 4—that the expansion of cattle raising in Dorset intensified rather than spreading widely over the county. Further, this intensification was determined by factors other than rail proximity in the 1870s and 1880s. If the parish returns for those years had distinguished dairy cows from beef cattle, we would likely find an effect of rail accessibility and dairy farming because the fresh milk trade depended upon railways to get their product to major urban markets.

France: The Department of the Cote d-Or In France, the Cote-d’Or Department presents story with similarities and differences. There, as in France generally, the tariffs on wheat introduced in 1883 shielded cereal farming from international competition, giving cereal farmers a reprieve not enjoyed by their British counterparts. To examine this, we map the pertinent attributes: percentage change in wheat cultivation by canton (comparable to the British registration district) and the distance from the centers of communes (comparable to British parishes). In Figure 6 the gray lines connected to station nodes represent the closest distance from any given commune to a station. As for wheat acreage and cattle density, a strict comparison with Britain at the parish/commune level is not feasible because the data needed for the communes of Cote d’Or are incomplete. Our data pertain to the next higher French administrative of the canton (comparable to a British registration district).

Across the cantons of the Cote-d’Or, examining the percentage change in cereal production between 1881 and 1905 shows what one would expect. Protected by tariff, wheat production remained fairly stable from the 1880s to the mid1890s, i.e., even during the great fall in wheat prices internationally. By 1905, however, the shift from wheat to cattle raising and to other uses of agricultural land was marked in all 33 cantons of the Department except those of Dijon and 33 For our French GIS, Thomas Thevenin has recently created estimates of real travel costs from each commune in France to its nearest station, but they were not available at the time of writing.

• 225

Railroads in historical context: construction, costs and consequences

Figure 6 - Percent change in wheat cultivation (in hectares) by canton in the Department of Côte-d’Or, 1892-1905, and proximity of communes to the nearest railway station, 1900-1909.

226 • Rail lines, 1900 - 1909 Commune center to nearest station Railways stations, 1900 - 1909 -24 - -17 16 - - 9 -8 - 0 1 - 15

0

10

20

x

40 Kilometers

Sources: Archives Départementales de la Côte-d’0r, 6M 12. IIa31, IIa37, IIa50, Statisitique agricole des communes et cantons, 1881, 1892, 1905; Carte des chemins de fer français, SNCF, 1944.

Beaune where the two largest cities of the department where located. Centered on the cities of Dijon and Beaune, the two cantons saw wheat production rise in that decade. (See Figure 6.) However, beyond the Dijon region and the major vine growing districts running from Dijon to Beaune, an increase in pasture and cattle numbers attests to the expansion and intensification of livestock farming.

(See Figure 7.)

When this information is combined with data on railway accessibility, the results suggest that rail service proximity was one factor that influenced changes in wheat production, not in the 1880s, as was true in Dorset, but in the 1890s.

Robert M. Schwartz

Figure 7 - Percent change in numbers of cattle by canton in the Department of Côte-d’Or, 1892-1905

• 227 Rail lines, 1900 - 1909 Railways stations, 1900 - 1909 Percent change in number of cattle, 1892-1905 -24 - -17 16 - - 9 -8 - 0 1 - 15 0

10

20

x

40 Kilometers

Sources: ibid.

By 1905, the building of branch lines and the opening of new stations in underserved areas, much as the Freycinet program had projected, was well underway.

Consequently, more farmers in the Cote d’Or and elsewhere had rail service closer to hand. To describe the effect of improved accessibility, scatter plots depict the relationships between proximity to a station and the change in cereal production and cattle raising.34 The decline of wheat cultivation and the proximity of rail transport were inversely related: from 1892 to 1905, the percentage decline in wheat (hectares) was greater as the distance to the nearest rail station increased.35 (See Figure 34 The limited number of units (33) that comprise the Cote-d’Or cantonal database makes the application of GWR analysis unfeasible at this point in our research.

35 The average is the mean distance from each commune in a canton to the nearest station. This is a better measure than the distance from the center or seat of a canton alone because the accessibility of outlying communes would be ignored. It was a state priority to open a station in each cantonal seat {chef-lieu).

Railroads in historical context: construction, costs and consequences

Figure 8 - Rail station proximity and change in wheat cultivation in Côte-d’Or cantons, 1892 to 1905

Percentage change in hectares of wheat cultivation

20

Proximity to rail station and change in wheat cultivation in Côte-D’or, 1892 to 1905

15 10 5 0 -5 -10 -15 -20 -25 -301

2

3

4

5

6

mean distance from parish center to to station in kilometers

7

8

Sources: ibid.

228 •

8.) As in parts of lowland Dorset, farmers closer to a station tended to reduce wheat production to a lesser extent than those who were farther away. Farmers at greater distance from rail transport more likely put their land and labor to uses more profitable than growing more wheat than needed for their own consumption. Even when wheat prices were protected, they were low by earlier standards.

The additional cost of transporting wheat from remote farms was doubtless a disincentive. On the other hand, farmers close to the cities of Dijon and Beaune enjoyed a competitive advantage both in their proximity to rail service and to the largest regional markets for grain and flour. To meet the local demand for grain and flour, wheat farming in the plains around the two cities expanded.

In cattle raising, percentage changes from 1892 to 1905 ranged from positive increases to steep declines, and the relationship with mean distance from a commune to a station was inverse. (See Figure 9.) By and large, the greater accessibility (smaller distance), the greater the increase in livestock numbers; conversely, the lesser accessibility (greater distance), the smaller the smaller the density of cattle in 1905 as compared to 1891. This pattern was more pronounced in upland areas than in the plains. In contrast to the plains of Dijon and eastern Beaune where wheat farming was expanding, in the upland cantons of Chatillon-sur-Seine, Arnay-le-Duc, and Aignay-le-Duc cattle farming was

Robert M. Schwartz

Figure 9 - Proximity to rail station and change in cattle raising in Côte-d’Or, 1892 to 1905

60

Proximity to rail station and change in cattle raising in Côte-D’or, 1892 to 1905 Châtillon-sur-Seine

50

Arnay-Le-Duc

Percentage change in cattle

40 30 Aignay-Le-Duc

20 10 0 -10

Dijon

Beaune

-20 -30 -401

2

3

4

5

6

mean distance from parish center to to station in kilometers

7

8

Sources: ibid.

expanding significantly. In 1905 compared to 1892, upland farmers who were relatively close to a station (3 to 4 kilometers away) were typically raising more stock than those farther away. In so doing, they followed a national trend and depended upon rail transport to adjust the shifting conditions of markets for milk, beef cattle, and meat. In sum, in the Cote-d’Or at the turn of the century, the proximity to rail transport facilitated, to a varied degree, the shift from wheat cultivation to livestock farming.

The same advantage held more or less true in the French department of the lier and in the English county of Derbyshire.36 In Dorset, the role of railways in farmers’ adjustments to difficult market conditions was significant, if complex, in wheat production. Whether these patterns held true in other French and British regions and in the two countries generally are matters next on our agenda.

CONCLUSION: A REFLECTION ON SPATIAL HISTORY This article tries to illustrate how historical GIS, geographic thinking, and spatial statistics, in the good company of traditional forms of historical narrative and 36 Schwartz, “Rail Transport, Agrarian Crisis, and the Restructuring of Agriculture,” 239-45, 247-50.

• 229

Railroads in historical context: construction, costs and consequences

230 •

analysis, are key ingredients in the making of spatial history. The story presented here of globalizing agricultural markets, intensifying international competition, the expansion of rail transportation, and agricultural change offers a sketch of a far-reaching historical transformation. Some important features, to be sure, are there thanks to GIS and spatial analysis. In spatial history, however, HGIS (historical GIS) works best as a junior partner. Given the data, it can help identify problems and facilitate their examination. But, like other tools in the historian’s kit, it cannot frame problems worth investigation. Nor does it generate interpretations and meanings. Its complexities carry a risk. Because the preparation of geo-referenced data and the learning of the technology can take a great deal of time, practitioners of HGIS often get caught up in the methods and give too little attention in their papers and publications to substantive results. We practitioners should do a better job interpreting and communicating our discoveries to make clearer why historical GIS is worth the effort.

A related imperative is to recognize both the limits of historical GIS and the importance of complementary sources and approaches. If GIS technology were to drive the investigation, one might easily overlook records documenting the lived experience of nineteenth-century farmers, their wives, sons, and daughters, the observations and opinions of journalists like Jefferies, or the testimony of farmers before the British and French investigations into agricultural and agrarian crisis. That would be a loss. On the positive side, the analysis of census records and agricultural statistics in a GIS can provide the broader patterns needed for situating case studies and rich qualitative evidence in their proper historical context.

Spatial history ought to be more than examining questions about geographic distributions over time, as when we use shaded thematic maps to show percentage change in wheat cultivation in Dorset and the Cote-d’Or. Indentifying such spatio-temporal patters is a good first step. To go further, spatial history should concern the study of spatial relationships and of spatial interconnectivity over time, i.e., the degree to which a change in one part of an inter-related system alters other parts in turn. For example, when, where, and to what extent did railway expansion effect agricultural production and land use? How did the arrival of rail transport in remote agrarian regions affect production and land use in those districts? In districts closer to major markets, districts that heretofore enjoyed the advantage of proximity to large and growing numbers of urban consumers? Or, as so concerned Jefferies, how did the expansion of rail transport in America affect farmers in Britain?

The geographer’s concept of “scale” and historian Marc Bloch’s conception of comparative history should also be components of spatial history.37 Incorporat37 Neil Brenner, “Between Fixity and Motion: Accumulation, Territorial Organization and the Historical Geography of Spatial Scales,”’Environment and PlanningD: Society and Space 16(1998): 459-81; Marc Bloch, “A Contribution Towards a Comparative History of European Societies,” in Land and Work in Medieval Europe (New York: Harper & Row, 1967): 44-81.

Robert M. Schwartz

ing multiple scales of geographic resolution and their inter relationships brings out the interconnectivity of change— or persistence—at different levels of human activity and natural forces. In the case described here, interconnectivity was significant at the global, national, and local levels of activity. American farmers harvesting wheat in Nebraska and the Dakotas were ultimately influential in decisions about agricultural land use in Dorset County and in the Department of Cote-d’Or, just as bad weather and poor harvests in the American Great Plains could result in higher wheat prices in Liverpool.38 In the 1890s, as American wheat farmers themselves came to suffer from falling prices, they came to believe that the Liverpool market set the low prices that threatened their livelihoods.39 Another task in spatial history is to tease out links between temporal and spatial change.40 Although only briefly treated here, the expansion of rail transport and its effects on agriculture is a revealing example of change moving across territory and time. Another imperative, to repeat, is to open the conversation about spatial history to a wider audience than specialists in historical GIS. The opening offered here, imperfect as it is, starts by framing the problem through the eyes of a contemporary observer and then continues by incorporating GISrelated findings into an accessible narrative about historical change. Narrative, after all, remains the lingua franca of most of historians and readers of history.

Finally, there is the urgent task of interesting historians in spatial relationships and renewing their regard for geographical thinking. This does not require GIS, and to presume that it does may be more hindrance than aid. A salutary reminder in this regard is to recall the insistent belief shared by some geographers and historians in the 1960s and 1970s that quantitative methods were the keys to the kingdom.

As concerns the promise of interconnected spatial history, Richard Jefferies, in some sense, showed us a way forward. A keen observer of agricultural change in Britain and abroad, he recognized both the international connection between American and British farmers, and the regional and local significance of rapid and convenient transportation in the age of growing globalization. True, as an advocate for British farming interests he was apt to stretch the truth in his characterizations of American and French rivals and of the inadequacies of rural rail 38 Melanie Finn, “Effects of Local Weather on a Global Market,” unpublished seminar paper, (Mount Holyoke College, South Hadley, Massachusetts, December 2009): 12-14, demonstrates the likelihood that drought and poor wheat harvests in Nebraska in the early 1870s led to wheat price increases in Liverpool, England.

39 U.S. House of Representatives, Depression of American Agricultural Staples, Report 1899, testimony of David Lubin and Alex Wedderburn, presidents of the state Grange associations of California and Virginia respectively, March 2, 1895, pp. 31-33, 54-55.

40 Geographic information scientists are currently exploring ways to formally incorporate time in GIS. On this see May Yuan, “Adding time into Geographic Information System,” in The Handbook of Geographic Information Science, ed. A. Stewart; Wilson Fotheringham, John P., Blackwell Companions to Geography (Maiden, MA: Blackwell 2008), 169-84.

• 231

Railroads in historical context: construction, costs and consequences

service in Britain—all the better to rally ingenuity and resolve to restore British agriculture to its proper place in the world. Although he overstated the travail of moving crops from farm gate to station platform, he anticipated the day when gasoline powered trucks and automobiles would ply the roads of rural Britain.

An ardent observer of Britain’s agrarian world in international perspective, he was a visionary, too.

232 •

Robert M. Schwartz

REFERENCES Afton, Bethanie. “The Great Agricultural Depression on the English Chalklands: The Hampshire Experience.” Agricultural History Review 44 (1996): 191-205.

Atkins, P. J. “The Growth of London’s Railway Milk Trade, c. 1845-1914.” The Journal of Transport History 4, no. 4 (1978): 208-26.

Bloch, Marc. “A Contribution Towards a Comparative History of European Societies.” In Land and Work in Medieval Europe 44-81. New York: Harper & Row, 1967: 44-81 Bouneau, Christophe. “Chemins de fer et developpement rural en France de 1852 a 1937 : la contribution de la Compagnie du Midi “ Histoire, Economie et Societe 1 (1990): 95112.

Brenner, Neil. “Between Fixity and Motion: Accumulation, Territorial Organization and the Historical Geography of Spatial Scales.” Environment and Planning D: Society and Space 16 (1998): 459-81. Broder, Albert. “La longue stagnation francaise: panorama general.” la La Longue stagnation en France. L’autre grande depression 1873-1897, edited by Yves; Broder Breton, Alber; Lutfalla, Michel, eds. Paris: Economica, 1997.

Carmona, Juan. “Sharecropping and livestock specialization in France, 1830-1930.”

Continuity and Change 21 (2006): 235-59.

Caron, Francois. Histoire des chemins defer en France, 1740-1883. Paris: Fayard, 1997.

–. Histoire des chemins defer en France, 1883-1937. Paris: Fayard, 2005.

Clout, Hugh. “The Pays de Bray: A Vale of Dairies in Northern France.” Agricultural History Review 51, no. 2 (2003): 199-208. Coclanis, Peter A. “Back to the Future: The Globalization of Agriculture in Historical Context.”

SAISReview 23, no. 1 (2003): 71-84.

Collins, E. J. T., ed. The Agrarian History of England and Wales. Volume VII1850-1914.

Edited by Joan Thirsk, The Agrarian History of England and Wales. Cambridge: Cambridge University Press, 2000.

–. “The Great Depression, 1875-1896.” In The Agrarian History of England and Wales.

Volume VII1850-1914, edited by E. J. T. Collins, 138-207. Cambridge: Cambridge University Press, 2000.

• 233

Railroads in historical context: construction, costs and consequences

Dormois, Jean-Pierre. “La ‘vocation agricole de la France’. L’agriculture francaise face a la concurrence britannique avant la guerre de 1914.” Histoire etMesure 11, no. 3-4 (1996): 329-66.

Fanica, Olivier. “Du lait pour la capitale. La production laitiere autourde Paris (17001914).” In Acteurs et espaces de Televage (XVIIe-XXIe Steele). Evolution, structuration, specialisation., edited by Philippe; Moriceau Madeline, Jean-Marc. Rennes: Presses Universitaires de Rennes, 2006: 141-54 Fotheringham, A. Stewart, and John P. Wilson. The handbook of geographic information science, Blackwell companions to geography. Maiden, MA: Blackwell Pub., 2008.

Gonjo, Yasuo. “Le ‘Plan Freycinet’, 1878-1882: un aspect de la ‘Grande Depression’ economique en France.” Revue historique 248 (1972): 49-86.

Gregory, Ian N. A Place in History: A Guide to Using GIS in Historical Research. Oxbow: Oxford 2003.

Hallas, Christine. “The Social and Economic Impact of Rural Railway: The Wensleydale Line.” The Agricultural History Review 34 (1986): 29-45.

–. The Wensleydale Railway. Clapham [Bedfordshire]: Dalesman Books, 1984.

234 •

Hau, Michael. “”La resistance des regions d’agriculture intensive aux crises de la fin du XIXe siecle : Les cas de 1’Alsace, du Vaucluse et du Bas-Languedoc”.” In Un siecle d’histoire agricole frangaise. Actes du colloque de la Societe frangaise d’economie rurale, edited by Philippe Chalmin and Andre Gueslin (Economie rurale, 1988) : 31-41 Howell, D.W. “The impact of railways on agricultural development in nineteenth-century Wales.” Welsh History Review 7 (1974-5): 40-62.

Hunt, E.H; Pam, S.J. “Prices and Structural Response in English Agriculture, 1873-1896.”

The Economic History Review 50, no. 3 (1997): 477-505.

Jones, Eric L. “The Land that Richard Jefferies Inherited.” Rural History 16, no. 1 (2005): 83-93.

Koning, Niek. The failure of agrarian capitalism : agrarian politics in the UK, Germany, the Netherlands and the USA, 1846-1919. London: Routledge, 1994.

Lhomme, Jean. “La crise agricole a la fin du XIXe siecle en France. Essai d’interpretation economique et sociale.” Revue economique 21, no. 4 (1970): 521-53.

O’Rourke, Kevin H. “The European Grain Invasion, 1870-1913.” The Journal of Economic History 57, no. 4 (1997): 775-801.

Perren, Richard. Agriculture in depression, 1870-1940, New studies in economic and social history. Cambridge: University Press, 1995.

–. The meat trade in Britain, 1840-1914, Studies in economic history. London ; Boston: Routledge and Kegan Paul, 1978.

Robert M. Schwartz

Perren, Richard “Marketing of Agricultural Products.” In The Agrarian History of England and Wales. Volume VII1850-1914, edited by E. J. T. Collins, 953-98. Cambridge: Cambridge University Press, 2000.

Perry, P.J. “An Agricultural Journalist on the “Great Depression”: Richard Jefferies.” The Journal of British Studies 9, no. 2 (1970): 126-40.

Schmitt, Giinther. “Agriculture in Nineteenth Century France and Britain: Another Explanation of International and Intersectoral Productivity Differences.” The Journal of European Economic History 19, no. 1 (1990): 91-113.

Schwartz, Robert M.; Gregory, Ian; Marti-Henneburg, Jordi. “History and GIS: Railways, Population Change, and Agricultural Development in late nineteenth century Wales.” In Geography and the Humanities, edited by Douglas Richardson. Washington, DC: Association of American Geographers, 2010.

Straus, Andre; Verley, Patrick. “L’economie francaise au XIXe siecle. Analyse macroeconomique, une oeuvre isolee ou une ouverture vers des recherches novatrices?.”

Revue d’histoire du XIXe siecle 23 (2001): 210-18.

Taylor, D. “London’s Milk Supply, 1850-1900: A Remterpretation.” Agricultural History 45 (1971): 33-8.

Toutain, Jean-Claude. “La Croissance inegales des regions francaises: l’agriculture de 18101990.” Revue historique 590 (1994): 315-59.

–. La production agricole de la France de 1810 a 1990: Departements et Regions.

Croissance, productivity, structures. 3 vols, Histoire quantitative de l’economie francaise.

Grenoble: Cahiers de 1TSMEA [Economies et societes], 1992-3. Tracy, Michael.

Government and agriculture in Western Europe, 1880-1988. 3rd ed. New York; London: Harvester Wheatsheaf, 1989.

Turner, Michael. “Output and Prices in UK Agriculture, 1867-1914, and the Great Agricultural Depression Reconsidered.” The Agricultural History Review 40, no. 1 (1992): 38-51.

Turner, Michael E. “Agricultural Output, Income, and Productivity.” In The Agrarian History of England and Wales. Volume VII1850-1914, edited by E. J. T. Collins, 224320. Cambridge: Cambridge University Press, 2000.

Turnock, David. An historical geography of railways in Great Britain and Ireland. Aldershot, Hants, England, Brookfield, Vt: Ashgate, 1998. van Zanden , J. L. “The Growth of Production and Productivity in European Agriculture, 1870-1914.” The Economic History Review 44. 2 (1991): 215-39.

Vidal, Jean-Francois. Depression et retour de la prosperite. Les economieseuropeenes a la fin du XIXe siecle. Paris: L’Harmattan, 2000.

• 235

Railroads in historical context: construction, costs and consequences

Yuan, May. “Adding time into Geographic Information System.” In The Handbook of Geographic Information Science, edited by A. Stewart;Wilson Fotheringham, John P., 169-84. Maiden, MA: Blackwell 2008.

236 •

Joaquim da Costa Leite

THE ALTO DOURO AND EMIGRATION, 1855-1914 O ALTO DOURO E A EMIGRAÇÃO, 1855-1914 Joaquim da Costa Leite (U. Aveiro, Portugal) Costa Leite is professor (associate, with habilitation) in University of Aveiro, Portugal. His research concerns economic history and emigration issues.

Costa Leite é professor associado, com agregação, na Universidade de Aveiro, Portugal. As suas áreas de investigação realcionam-se com a história econonomica e a emigração.

Abstract Resumo

Since the mid-nineteenth century until the outbreak of World War I, Portugal felt the impact of globalizing forces, and emigration was simultaneously a reaction to and a driver of internationalization.

Portuguese emigration was characterized by a rising trend throughout the period, while the geographical spread of the migratory movement revealed a considerable regional variation, from regions with permanent and intense migratory flows since the beginning of that time period to regions where emigration was practically unknown.

The presentation will focus on the Alto Douro, trying to characterize its migratory flow — chronology, sex and age composition — in a comparative perspective, discussing its characteristics in relation with the region’s demographics and agrarian system.

Desde meados do século XIX até ao inicio da primeira guerra mundial, Portugal sentiu o impacto das forças da globalização, sendo que a emigração foi simultaneamente uma reação e um promotor da internacionalização.

A emigração portuguesa caracterizou-se por uma tendência crescente nesse período. A geografia dos movimentos migratórios revelou uma considerável variação regional, desde regiões com fluxos permanentes e intensos desde o principio do período, até regiões em que a emigração foi praticamente desconhecida.

Neste trabalho analisa-se o caso da emigração no Alto Douro, procurando caraterizar os seus fluxos migratórios – cronologia, composição por sexo e por idade – numa perspetiva comparativa, discutindo as suas características em relação com a demografia e os sistemas agrários da região.

• 237

Railroads in historical context: construction, costs and consequences

238 •

Joaquim da Costa Leite

The Alto Douro and emigration, 1855-1914 Joaquim da Costa Leite

Within its small territory, Portugal displays significant contrasts in terms of landscape, climate, human settlement, agrarian system and demographic regime. The migratory phenomenon interacts with agrarian and demographic factors, revealing also a significant regional diversity.

The present paper will analyse those interactions, which converge to give specific characteristics to the Alto Douro, constituting an interesting case-study. In the first section, I will consider the basic types of three agrarian and demographic regimes which characterized the Portuguese traditional society for centuries; in the second section I will draw attention to the migratory phenomenon in Portugal and some of its basic regional characteristics; in the third section I will examine the Alto Douro in order to characterize its migratory flow; in the fourth section I will make some concluding remarks.

1.

Figure 1 presents some major demographic characteristics of mainland Portugal mid-nineteenth century. Given the fact that this was before the introduction of railways, newspapers, and the modern postal system, the map is a picture of a traditional society. The fundamental regional differences had deep historical roots, and in some respects would continue to influence the economy, society and politics until the present day.

This was also before the large migratory flows of the second half of the nineteenth century, but there was already some emigration from the densely settled

• 239

Railroads in historical context: construction, costs and consequences

areas of the Portuguese Northwest, which was significant precisely because it revealed the connection between the regional demographic system and the pressure to send young men away in search for a living.

Figure 1 - Demographic Maps, 1849

density

male ratio

marriage rate

birth rate

mortality rate

Source: Maps compiled from the online edition of Silveira (2001) consulted 13-Jun-2003.

240 •

It is possible to observe a significant contrast between North and South, the first more densely settled, based on a system of intensely cultivated, small parcels of land, with small landowners constituting a sizable part of the population.

The South, based on large estates, was spotted with large villages inhabited by mostly landless laborers. These tended to marry early, gave birth to more children, and the mortality rate was higher; while the small landowners to the North, trying to contain the fragmentation of property, married later or not at all — the celibacy rate was higher — had less children, and the mortality rate was lower.

In Northern Portugal, the Northwest was further characterized by the moderating influence of the sea, with heavier rainfall and less extreme temperatures in summer and winter, favoring a diversified, mixed husbandry, which required careful work all year-round. The result was a fairly productive agriculture which provided a modest living, adapting to different economic cycles on the basis of its diversity, a subsistence nucleus with some commercial involvement, but unable to specialize, grow and become rich. Still, these modest peasants had some property which they could mortgage or sell when they wanted to send away a son to become a priest, to learn a trade, or try their luck in emigration. The sons, for their part, were encouraged to go away; unless they married a bride with a dowry, the alternative would be to stay with their parents, help the sibling who inherited the land, and grow old, single without children. In those villages, emigration was accepted, becoming a part of the regional culture; it became accept-

Joaquim da Costa Leite

able even for a married man to go away for long periods of time, while his wife took care of the house, family and land, with the help of relatives and neighbors.

In order to check some of these characteristics, it is possible to concentrate on three distritos where Braga, in the province of Minho, represents the Northwest; Bragança in Tras-os-Montes represents the Northeast; and Beja, in Alentejo, represents the Southern plains. Table 1 shows that fertility within marriage, corrected for absent husbands (fcc), was higher in Braga, but the crude birth rate was brought low by a higher age at marriage (idcm) and a higher percentage of women remaining single (cdm). Already in 1864 the population census counted 97 married men for 100 married women (rmc), while in the total population there were 80 males to 100 females (rmt). Bragança was close to Braga in terms of age at marriage, somewhat more relaxed in terms of celibacy, and close to equilibrium of the sexes in 1864. In fact, both Bragança and Beja had an excess of males, probably due to some mobility for employment and the army. In Beja access to marriage was facilitated by popular culture, revealed in a significantly lower age at marriage and lower celibacy.

Table 1 - Regional characteristics of family systems idcm

cdm

fcc

ileg

rmc

rmt

Braga

27,5

27,2

774

10,5

97

80

Bragança

26,5

18,1

738

19,4

102

103

Beja

24,2

11,5

639

9,6

101

104

Age at marriage (idcm) and percentage remaining single in age group 40-44 (cdm) of women in 1878. Fertility within marriage, corrected (fcc) in 1890. Male ratio (number of males per 100 females) in the married population (rmc) and total (rmt) in 1864.

Sources: Robert Rowland (1997b) quadro 3.4 p. 96; Massimo Livi Bacci (1971) quadro 19 p. 68; Movimento da população and population censuses.

Table 2 shows some characteristics of the agrarian system, and the consequent population density. Staple grains were an essential part of the agrarian system, adjusted to climate, rainfall and access to water, agricultural calendar, etc. Maize dominated in the fields of Braga, while rye predominated in Bragança, complemented with wheat; the plains of Beja were sown almost exclusively with wheat. In the North, landholdings are expected to have been more widespread, but this is not proven by the tax collection figures; however, the average size of land units shows a clear difference between North and South.

Climate and soil, and human effort converted into concrete agrarian practices accumulated throughout the centuries, resulting in 1890 in a population density

• 241

Railroads in historical context: construction, costs and consequences

Table 2 - Regional characteristics of agrarian systems Density

Grains maize

Braga

dmp

rye

cpf

wheat

124

96

4

0

0,44

82

Bragança

27

2

61

37

0,90

114

Beja

14

2

1

97

8,73

84

Population density (inhabitants per square kilometer), 1890. Staple grains in percentage of crop, 1903. Average size of land property (dmp) in hectares, 1890. Tax collections in 1891 per 100 households (cpf) in 1890.

Sources: Censo da população em 1890 vol. I quadro VIII p. 34 (land surface of distritos and population density); Annuario Estatistico de Portugal 1903 vol. II quadro nº 20 pp. 110-111 (grain production in hectoliters, converted in tons according to Lains (1990) p. 12); Estatistica Agricola: Resumos Estatisticos. IV (average size of land holdings); Annuario Estatistico das Contribuições Directas 1891 “Mappa geral estatistico, por districtos, sobre o serviço da contribuição predial no anno de 1891” p. 17.

242 •

of 124 inhabitants per square kilometer in Braga; 27 in Bragança; and 14 in Beja.

The regional differentiation thus sketched also had migratory consequences in the cultural acceptance of emigration, the propensity to emigrate evident in the North — especially the Northwest — and the economic groups likely to emigrate.

2.

Figure 2 shows the evolution of emigration rates, taking as a reference the five population censuses from 1864 to 1911. The maps exhibit a rising trend throughout the period — the trend is subject to fluctuations, as 1900 was less pressing than 1890, for example — while the migratory attraction, centred in Porto and its immediate region in 1864, spread to a larger area as more people emigrated, sending money and good news to the network of family and friends back home.

It is also evident that emigration spread throughout the North, but the movement did not easily penetrate the Southern distritos, where social and economic barriers made it difficult to take hold: economic polarization meant that the poorer segments of the population, who had something to gain from emigration, could not afford to, while those who could did not need to emigrate. It is also possible that cultural and social characteristics in the South countered the diffusion of information favorable to emigration.

Joaquim da Costa Leite

Figure 2 - Emigration Rates in the Mainland

Notes: Emigration rates (per thousand) are calculated from four-year averages centred on the census date, with the exception of emigration in 1866-69 relative to population in 1864 (no emigration figures are available on a distrito basis before 1866). The overall rates for the Mainland and the Islands are also shown for comparison.

Sources: Emigration and population data compiled from official statistics. For more details, see Joaquim da Costa Leite, “Portugal and Emigration, 1855-1914” (Ph. D. Dissertation, Columbia University, 1994), appendix 3.

Contemporaries explained the phenomenon as the result of the propaganda shipping agents exerted on a gullible and illiterate population, and this interpretation is still reflected in current references and studies of emigration. Shipping agents did indeed advertise their routes and the increasingly comfortable — and less risky — means of transportation to Brazil and other destinations.

But the insistence of intellectuals, politicians and priests on the greed of shipping agents and employers abroad was itself a widespread form of propaganda against emigration. Apart from that, the traditional interpretation ignored some basic facts, such as the evidence of economic opportunities and higher wages in Brazil; distritos with higher illiteracy rates were the districts where emigration was irrelevant. But above all, it is essential to acknowledge that even illiterate people could have access to personalized, concrete information about opportunities abroad through their network of relatives and friends.

Emigration requires the combination of need and opportunity. Railways, steamship lines, newspapers, a modern postal system, commercial and banking connections, facilitated the circulation of people, goods, and information, furthering the meeting of need and opportunity. Whenever data allows for the quan-

• 243

Railroads in historical context: construction, costs and consequences

tification of variables and statistical testing, those basic premises are confirmed.

Table 3 tests for factors in regional emigration based on data from the distritos of the Mainland in 1890. Emigration in 1890 is explained by the network of emigrants from a previous period — emigrants to Rio de Janeiro in 1870-74 relative to population in 1890 — as a proxy for the migratory experience, or available network of friends and relatives abroad; the percentage of taxpayers per household as a proxy for the capacity to borrow or pay for emigration; and in some areas there was also a sign of crisis in hectares of vineyards devastated by the phylloxera, relative to the active population, as an additional factor.

Table 3 — Factors in Regional Emigration, 1890

244 •

1)

EmMF = -3.040 +0.283 ImRJ +0.047 Txp +0.045 Phx

R2=0.859

2)

EmMF = +2.199 +0.235 ImRJ -0.218 Txp +0.056 Phx

R2=0.787

3)

EmMF = -3.045 +0.037 Txp +0.06 Phx

R2=0.341

EmMF = Average annual emigration rates (per thousand) in 1889-1892, both sexes combined.

ImRJ = Emigration rates calculated on the basis of immigrants entered in Rio de Janeiro in 1870-74, relative to population in 1890.

Txp = Percentage of taxpayers (1891) per household in 1890.

Phx = Hectares of vineyards ravaged by the phylloxera (defined as área de destruição, 1886), relative to the active population in 1890.

Compiled from sources in Joaquim da Costa Leite, “Portugal and Emigration, 1855-1914” (Ph. D. Dissertation, Columbia University, 1994), appendixes 3 and 5.

Table 4 refers to the situation in 1911. We find the same basic factors, except that the element of crisis is now more widespread, identified with the agrarian system based on rye as the staple grain.

The network of previous emigrants, the so-called family and friends effect, was always relevant. Access to property and the average size of land units were also relevant, showing that it was easier to find potential emigrants in areas of small landholdings, more characteristic of Northern Portugal. But then, apart from the more or less natural spread of the migratory phenomenon as more and more people had friends and relatives abroad telling them of better opportunities, while providing help and support to those who wanted to try, there was in some areas an additional factor of a crisis situation, sometimes causing a large number of people to leave the country. From the end of the nineteenth century until the outbreak of World War I, it points predominantly to the distritos of Bragança and Vila Real, comprising the region of the Alto Douro.

Joaquim da Costa Leite

Table 4 — Factors in Regional Emigration, 1911 1)

EmMF = -5.754 +1.417 Em90 +0.060 Txp +0.252 Rye

R2=0.807

2)

EmM = -8.134 +2.464 EM90 +0.106 Txp +0.166 Rye

R2=0.874

3)

EmF = -3.173 +0.306 EmM +0.247 Rye

R2=0. 795

4)

EmM = +4.202 +2.732 Em90 -0.718 Fms +0.218 Rye

R2=0. 788

5)

EmM = -7.866 +0.177 Txp +0.193 Rye

R2=0. 588

EmMF = Average annual emigration rates (per thousand) in 1910-1913, both sexes combined. Specific rates are shown as EmM (male) and EmF (female).

Em90 = Average annual emigration rates (per thousand) in 1889-1892, both sexes combined.

Txp = Percentage of taxpayers per household in 1911.

Rye = Percentage of rye in grain production (maize, wheat and rye) in 1903.

Fms = Average size of land units in 1910.

Compiled from sources mentioned in Table 3.

• 245

3.

We have seen that Vila Real and Bragança, at some distance from Oporto, did not significantly participate in emigration during the early decades of the second half of the nineteenth century. By 1890 however, Vila Real, with its vineyards attacked by the phylloxera, was already among the distritos with higher emigration rates; Bragança was still relatively immune, but its relatively low rates at the time meant that part of the population was already developing a network abroad, which would be put to effective use when the need arose.

Figure 3 shows that the spread of emigration to the Portuguese hinterland brought a fast growth of emigration rates, while differentiating the migratory flow with an increasing participation of women and children. The increasing share of women and children was evident in all areas of emigration, witnessing the diversity and spread of the migratory phenomenon, but some distritos maintained a traditional reference, with a clear predominance of males. We can see in table 5 that in Braga, emigration rates rose from just over 5 per thousand in 1890 to close to 10 per thousand in 1911, while the participation of women in-

Railroads in historical context: construction, costs and consequences

creased, but the male ratio among emigrants, which was 721 in 1890, decreased somewhat in 1911, although there were still 580 males to 100 female emigrants.

In the meantime, emigration rates rose much higher in Vila Real and Bragança, while the sex ratio was almost balanced among emigrants. These two aspects combined were very meaningful, and had great consequences for the future.

Figure 3 - Emigration Rates, Growth and Sex Composition of the Migratory Flow in Northern Portugal, 1890-1911

246 •

While in the Northwest emigration rates rose but remained relatively moderate without disrupting the traditional pattern — predominantly male emigrants in search of a career or complementary wages — emigration represented an element of flexibility and adaptation. The agrarian system and demographic regime was able to accommodate the flow, benefitting from some relief of demographic pressure and excess manpower while receiving an inflow of remittances to pay for a mortgage, buy cattle, extend the family home, or send a son to college.

In the Northeast, extremely high emigration rates with large numbers of men and women leaving together were a sign of entire families going away, many with little thought of coming back. This represented a threat of depopulation, disrupting traditional patterns of work and human settlement.

If we look into the figures at the concelho level, we verify that this was widespread. Table 6 shows that in Vila Real, with population ranging from just over 7000 inhabitants in Mesão Frio to 37000 in the concelho of Vila Real, emigration rates were practically all higher than in Braga — often much higher — and the sex ratio was not very unbalanced, the record being set by Vila Pouca de Aguiar with 252 male to 100 female emigrants.

Joaquim da Costa Leite

Table 5 - Emigration Rates, 1864-1911 Distritos

1864 te

1890 rme

te

1911 rme

te

rme

Vila Real

1,29

nd

9,70

528

23,53

Bragança

0,14

* 1350

4,27

234

36,79

164 111

Braga

2,70

6087

5,09

721

9,68

580

Beja

0,23

* 144

0,92

132

2,85

110

Mainland

1,29

1371

4,29

428

10,23

279

Islands

6,03

236

15,57

114

23,51

128

Portugal

1,70

625

5,16

294

11,15

246

* Less than 200 emigrants in four years Emigration rates per thousand inhabitants (te) and male ratio of emigrants (rme). Average values of four years centred on censos dates, except in 1864 with emigration in the period 1866-1869.

Sources: Population census; Primeiro Inquérito Parlamentar; Movimento da população.

Table 6 - Emigration Rates in the concelhos of Vila Real, 1911 Te Pop.

Rme

Males

Females

Total

Alijó

20005

36,73

22,34

29,18

148,72

Boticas

10637

29,56

10,32

19,27

249,43

Chaves

37834

30,44

15,42

22,47

174,49

Mesão Frio

7289

33,59

14,67

23,51

200,58

Mondim de Basto

7910

19,99

9,50

14,41

185,00

21820

10,38

3,85

6,94

241,35

7418

41,57

25,33

32,98

146,31

Peso da Régua

18983

18,20

9,73

13,75

169,07

Ribeira de Pena

10256

13,77

4,62

8,94

266,67

Sabrosa

12408

38,40

20,68

28,99

163,81

Santa Marta de Penaguião

11210

27,38

16,00

21,35

151,93

Valpaços

25175

34,47

21,51

27,55

140,02

Vila Pouca de Aguiar

17424

34,84

12,14

22,75

251,78

Vila Real

37178

43,26

23,00

32,29

159,06

245547

30,44

15,79

22,65

214,00

Montalegre Murça

Total

Emigration rates per thousand inhabitants (te) and male ratio of emigrants (rme). Average values emigration in the period 1910-1912.

Sources: 1911 Population census; Movimento da população.

• 247

Railroads in historical context: construction, costs and consequences

In Bragança, the situation was basically similar, with even higher emigration rates and the sex ratio ranging from 90 to 167. Population size ranged from under 7000 in Freixo de Espada-à-Cinta to 31000 in Bragança.

Table 7 - Emigration Rates in the concelhos of Bragança, 1911 Te Pop.

Males

Females

Total

9316

47,82

44,32

46,05

105,92

Bragança

31307

27,53

26,51

27,02

105,01

Carrazeda de Ansiães

13654

47,63

29,52

38,33

152,82

6934

33,84

25,79

29,71

124,73

20991

47,31

43,78

45,53

105,96

Alfândega da Fé

Freixo de Espada-à-Cinta Macedo de Cavaleiros

248 •

Rme

Miranda do Douro

11210

6,14

3,46

4,76

166,67

Mirandela

22109

31,95

23,12

27,44

132,44

Mogadouro

17063

49,46

50,37

49,91

100,24

Torre de Moncorvo

16512

48,18

32,81

40,35

141,43

Vila Flor

10352

40,87

33,25

36,93

115,20

Vimioso

11882

37,97

34,88

36,39

103,61

Vinhais

20694

27,86

29,54

28,72

90,29

192024

36,75

31,51

34,09

144,83

Total

Emigration rates per thousand inhabitants (te) and male ratio of emigrants (rme). Average values emigration in the period 1910-1912.

Sources: 1911 Population census; Movimento da população.

What did such figures mean? In short, it meant that the regional economy was breaking down, its poverty exposed before increased opportunities abroad, a trend accelerated by any sectoral crisis.

4.

In Vila Real and Bragança, the poor subsistence sector, in a population with some access to property to raise cash, generated a trend to look for better opportunities abroad, and the connections and networks required to emigrate became available at the end of the nineteenth century.

But here we need to add a brief reference to a factor too often ignored: the

Joaquim da Costa Leite

interconnection between the subsistence crops and cash crops. Subsistence and cash crops are frequently described following different, separate logics, while in fact, in a traditional economy, they are related, sometimes intimately connected.

Peasants need some cash to pay taxes, to settle accounts, to make dowries or set a son as an apprentice, or to even send him abroad. When a cash crop goes through a crisis, it also threatens the subsistence sector in diverse ways. The factors in regional emigration, already mentioned, detected the effect of phylloxera and rye: we read that as the poverty of an agrarian system, predominantly of subsistence, based on rye; and the cumulative effect of phylloxera on vineyards and the wine trade. There was also a crisis affecting the silkworm, diminishing another one of the traditional commercial opportunities. These commercial sectors were not offset by the development of new industries.

In contrast, emigration in the Northwest did not disrupt the demographic system, providing some flexibility to further some agricultural adaptation, while the region developed some urban employment.

It is clear, comparing with the situation in the Northwest, that the opening of the economy of the Northeast in the presence of commercial crisis and the absence of industrial employment stimulated high rates of emigration, involving entire families. The regional economy was undergoing a profound adjustment of its relative resource endowment, an adjustment that left the local elites apparently helpless. As for emigration, we may regret that so many went away; alternatively, we may rejoice that so many were able to try having a second opportunity across the Atlantic.

• 249

PART 3 Engineering, finance and management Engenharia, financiamento e gest達o

Álvaro Ferreira da Silva

MORE THAN A BRASS NAMEPLATE ON THE DOOR: FOREIGN OWNERSHIP AND CONTROL IN THE COMPANHIA REAL DOS CAMINHOS DE FERRO PORTUGUESES (1860s-1890s) INVESTIMENTO ESTRANGEIRO E CONTROLO DE GESTÃO NA COMPANHIA REAL DOS CAMINHOS DE FERRO PORTUGUESES (1860-1890) Álvaro Ferreira da Silva (Nova School of Business and Economics, U. Nova de Lisboa, Portugal) Associate Professor of Economic and Business History (Universidade Nova de Lisboa, Faculdade de Economia). Undergraduate degree in History (Universidade de Lisboa, 1982), Master in Historical Economics and Sociology (Universidade Nova de Lisboa, 1990), Doctor in History (European University Institute, Florence, 1998). His research interests are focused on business and economic history, urban history and the history of technology. Recent works: História Económica de Portugal, 1700-2000 (2005, edited with Pedro Lains), História da Informática em Portugal (2006), “Sanitary revolution and technology in nineteenth-century Lisbon” (2006), “The peculiar customer: conflicts of power and the modern water supply system in Lisbon (1850-1930)” (2007), “Una máquina imperfecta: tecnología sanitaria en Lisboa en la segunda mitad del siglo XIX” (2007), “A família em Portugal no século XVIII: posição familiar dos jovens e as dinâmicas dos grupos domésticos” (2008), “The ‘script’ of a new urban layout: mobility, environment and embellishment. Street’s uses in Lisbon between 1880 and 1920” (Lely Prize at the T2M Conference, International Association for the History of Transportation, Traffic and Mobility, Otawa, 2008), “Foreign capital and problems of agency: the Companhias Reunidas de Gás e Electricidade in Lisbon (1890-1920)” (2008), “Engineers and organizational behaviour: the Companhia Real dos Caminhos de Ferro Portugueses (1870-1885)” (2009), “Padrões de mobilidade interna em Portugal na segunda metade do século XIX” (2009), “In search of the urban variable: Understanding the roots of urban planning in Portugal” (2009), “Street, sanitation and beautification: urban intervention in nineteenth-century Lisbon” (2010), “A economia portuguesa na I República” (2011), “Foreign investment and multinationals in Portugal (1926-1974)” (forthcoming), “A crise orçamental e monetária portuguesa no contexto internacional (1914-1931)”, “Circulation and appropriation of management knowledge: management consulting in Portugal during the early 1970s” (forthcoming). Álvaro Ferreira da Silva é professor de História Económica e Empresarial na Universidade Nova de Lisboa (Faculdade de Economia). Doutorado pelo European University Institute, Florença (Itália) em 1998.

Os seus interesses de investigação centram-se nas temáticas de história económica e história empresarial, história urbana e história da tecnologia.

• 253

Railroads in historical context: construction, costs and consequences

Abstract Resumo

The relationship between ownership and control of distant ventures has been a major topic in the international business literature during the first global economy. It prompted the creation of a specific organizational form: the free-standing company.

The Portuguese railway Companhia Real did not have the formal characteristics of a free-standing company. However, the reconstruction of its decision-making processes reveals that it faced similar challenges to accommodate the agency problems between holders of property rights and managers in the field. This paper sheds further light on the discussion about ownership and managerial control in firms like railways, utilities or mining companies in peripheral countries.

254 •

A relação entre propriedade e controle em empreendimentos empresariais a longa distância tem sido um tópico importante na literatura sobre investimentos internacionais durante a primeira economia mundial. A proposta de um conceito específico de organização empresarial – free-standing company – foi o resultado deste esforço. A Companhia Real, a maior empresa portuguesa de caminhos de ferro, não tinha as características formais da free-standing company. A reconstrução dos processos de decisão desta empresa revela, porém, que ela enfrentava desafios semelhantes para acomodar os problemas de agência entre detentores de direitos de propriedade e gestores no terreno. Este capítulo lança uma nova luz na discussão sobre propriedade e controlo em empresas em sectores como os caminhos de ferro, as infraestruturas ou as minas em países periféricos.

Álvaro Ferreira da Silva

More than a brass nameplate on the door: foreign ownership and control in the Companhia Real dos Caminhos de Ferro Portugueses (1860s-1890s) Álvaro Ferreira da Silva 1. INTRODUCTION In 1851, a military insurrection substituted the ruling government in Portugal.

Its manifesto stated that “it was a revolution to end with all the revolutions”, promising political stability, as well as prosperity1. The conviction was that economic backwardness in Portugal did not result from a less favorable resource endowment regarding advanced European countries. The lower level of market integration blocked economic growth. Transport infrastructure would improve circulation and integrate the domestic market, unleashing the potential for growth. Railway construction was the most well-known facet of this policy2, with traits similar to what Saint-Simonians were sponsoring as a way to promote economic growth3.

A vicious circle haunted these great expectations. If the lack of transport infrastructures checked growth, the shortage of domestic financial resources limited private investment. It was believed that this vicious circle could only be broken by the combined action of foreign borrowing and public financial support. Through foreign investment, it would be possible to overcome the deficiencies in the domestic capital market, characterized by low liquidity and institu1 Manuel Vilaverde Cabral, 0 Desenvolvimento do Capitalismo em Portugal no Século XIX (Lisboa, 1976); José Miguel Sardica, A Regeneração sob o Signo do Consenso (Lisboa, 2001).

2 Maria Eugenia Mata, “As Três Fases do Fontismo: Projectos e Realizações,” in Estudos e Ensaios: Em Homenagem a Vitorino Magalhães Godlnho (Lisboa, 1988).

3 Michael Cowen and Robert Shenton, “The Invention of Development,” in Power Of Development, ed. Jonathan Crush (London, 1995) and John C. Eckalbar, “The Saint-Simonians in Industry and Economic Development,” American Journal of Economics and Sociology, 38 (1979): 83-96

• 255

Railroads in historical context: construction, costs and consequences

256 •

tional underdevelopment. However, the economic backwardness of the country implied that entrepreneurs would not find attractive any investment in transport infrastructures unless state subsidies provided some guarantee. Public incentives to attract foreign investment would eventually break the vicious circle, aligning private returns with social returns.

Similar programs were common to other countries across the world4. The rise of foreign investment flows became a typical response to the kind of problems faced by Portugal5. As a result, foreign railway companies came into being, assembling capital, technical knowledge, and other services in more advanced economies to be mobilized to less-developed ones6. This large mobilization of resources introduced a clear institutional specificity to the railway business in many parts of the world. The sources of capital, technical and managerial knowledge were located many miles away from the places where actual construction and operation were taking place, potentially introducing agency problems between the holders of property rights and the managers in the field7.

In Portugal there have been several studies on patterns of investment in railways during the nineteenth century8, as well as on their impact on economic growth9. However, an analysis of the governance and managerial structures characterizing these firms is absent from the literature. The goal of this study is to fill this gap. There is a further argument running sotto voce in the current studies on the financial links between foreign investors and the Portuguese railway companies suggesting that foreign capital was mostly a portfolio investment without the involvement of foreign shareholders on the actual management.

4 For Spain, Miguel Artola, Los Ferrocarriles en España, 1844-1943. 1. El Est ado y los Ferrocarriles (Madrid, 1977) and Miguel Mufioz Rubio et a/., Siglo y Medio del Ferrocarril en España 1848-1998: Economia, Industria y Sociedad (Madrid, 1999). For a global perspective, Ralf Roth and Günter Dinhobl, Across the Borders: Financing the World’s Railways in the Nineteenth and Twentieth Centuries (Aldershot, 2008).

5 Barry Eichengreen, “Financing Infrastructure in Developing Countries. Lesson from the Railway Age,” in Infrastructure Delivery. Private Initiative and the Public Good, ed. Ashoka Mody (Washington DC, 1996).

6 Mira Wilkins, “Multinational Enterprise to 1930,” in Leviathans: Multinational Corporations and the New Global History, eds. Alfred D. Chandler Jr. and B. Mazlish (Cambridge, Mass., 2005), 53-6.

7 M. C. Jensen, Foundations of Organizational Theory (Cambridge, Mass., 1998).

8 António Lopes Vieira, The Role of Britain and France in the Finance of Portuguese Railways, 1850 - 1890. A Comparative Study in Speculation, Corruption and Inefficiency, Ph.D. dissertation, University of Leicester, 1983; Fernanda Alegria, A Organização dos Transportes em Portugal (1850-1910). As Vias e o Tráfego (Lisboa, 1990); Magda Pinheiro, Chemins de Fer, Structure Financiere de L Etat et Dependance Exterieure au Portugal (1850-1890), PhD. dissertation, Univ. de Paris I, 1986; Magda Pinheiro, “The French Investors in Portuguese Railways from 1855 to 1884: Three Cases,” in Across the Borders.

9 David Justino, A Formação do Espaço Económico Nacional. Portugal, 1810-1913 (Lisboa, 1989), Maria Eugenia Mata and Lara Tavares, “The Value of Portuguese Railways for Consumers on the Eve of the First World War,” Transporte, Servicios y Telecomunicaciones, 7, (2004): 81-100, Luis Espinha da Silveira et a/., “Population and Railways in Portugal, 1801-1930,” Journal Of Interdisciplinary History 42 (2011): 29-52.

Álvaro Ferreira da Silva

This article is focused on the analysis of the Companhia Real dos Caminhos de Ferro Portugueses (hereinafter CRCFP), the first successful foreign venture in the railway business in Portugal, and the only firm to build a network of railway lines during the second half of the nineteenth century. As in other railway companies in peripheral Europe, a long distance existed between the sources of capital and knowledge, and the location of operations. The study of CRCFP raises the foremost research questions for a comparative assessment of ownership and control in situations when communication and managerial problems arise as a result of the long distance links aforementioned: What were the functions of foreign investors, besides capital provision? How were the managerial responsibilities shared (or not) between the capital home and the railway operational facilities? What was the nature of the governance and control mechanisms existing in the railway company?

The literature on nineteenth-century international business identified a business form to deal with long-distance investment. The free-standing company was incorporated and had its headquarters in the home country’s capital, but developed its business activity in another country. It sets apart from the classical multinational firm, which starts with domestic operations and then moves abroad. After the seminal texts10, research has progressed and the free-standing company was considered one of the few contributions business history had to the international business literature11. Besides urban utilities, natural or renewable resources, and financial services, railways were one of the fields favored by free-standing companies. However, few studies looked at this organizational form in this industry12.

The CRCFP did not have the formal characteristics of a free-standing company. The largest proportion of the capital came from abroad, but its headquarters were in Lisbon. However, the above-mentioned questions on how the relations between ownership and control shaped the railway company’s organizational structure and governance are part of the research agenda on free-standing companies. This agenda mobilizes several dimensions13. Two are particularly important to the present study. The first one deals with the complex character of international capital flows in late nineteenth century. The second focuses on the institutional characteristics assumed by foreign investment, as capital was embedded in an organizational form with specific information asymmetries and 10 Mira Wilkins, “The Free-Standing Company, 1870-1914: An Important Type of British Foreign Direct Investment,” Economic History Review, 2nd Ser., 41 (1988): 259-82.

11 Matthias Kipping and Behliil Usdiken, “Business History and Management Studies,” in The Oxford Handbook Of Business History, eds. Geoffrey Jones and Jonathan Zeitlin (Oxford, 2007), 104. For a synthesis, Mira Wilkins and Harm Schroter, The Free-Standing Company in the World Economy 1830- 1996 (Oxford, 1998) 12 Wilkins, “Multinational Enterprise to 1930,” 57; David Boughey, “British Overseas Railways as Free-Standing Companies, 1900-1915,” Business History 51 (2009): 484-500.

13 Mira Wilkins, “The Significance of the Concept and a Future Agenda,” in The Free-Standing Company, 421-39.

• 257

Railroads in historical context: construction, costs and consequences

258 •

motivations for exerting or delegating managerial responsibilities.

The period focused on is the transition between the construction phase (1860-early 1870s) to the full operation of the network during the 1880s. Several changes in the reference shareholders and conflicts involving foreign and domestic shareholders or managers characterize this period. Company sources on governance were particularly opaque. They cultivated formal and normative approaches, hiding the actual way in which authority and control was exercised.

The company statutes and internal regulations will be used to reconstruct governance rules. However, more importance were to be given to moments of conflict in order to disclose the actual mechanisms of control. This approach has some similarities to the concept of “social drama”14, used in anthropology to elucidate social practices beneath the surface of formal and social regularities.

This study progresses by reviewing the characteristics of the foreign investment in the CRCFP. Section 2 will provide the patterns of ownership characterizing the company and how they were expressed in its top administration. After establishing the extensive presence of foreign capital, section 3 investigates the model of governance in use for protecting property rights of foreign investors.

The article proceeds with sections 4 and 5 to establish the mechanisms of control available to reference shareholders, trying to understand how the company was managed and what functions the head office performed. Finally, the last section integrates the conclusions made from the analysis of the company in literature on international business during the nineteenth century.

2. FOREIGN INVESTMENT IN THE CRCFP In order to attract foreign investment, the Portuguese government granted the concession for the construction and exploitation of the railways together with some advantages, such as the guarantee of the interest on the sums invested, a subsidy for each kilometer of line constructed, various kinds of fiscal exemption, as well as other incentives15. There was ultimately nothing very different from what was being put into practice in other European countries at the time16.

The importance of free-standing companies to railway construction in Portugal is irrelevant. The Central Peninsular Railway Company (1852), the South Eastern of Portugal Railway Company (1860) and Minho District Railway Company (1874) were registered in London, but went bankrupt in a short period 14 Victor Turner, Dramas, Fields and Metaphors: Symbolic Action in Human Society (Ithaca, NY, 1974).

15 Vieira, The Role of Britain and France, 110 ff.; Alegria, A Organização dos Transportes em Portugal (18501910), 306 ff.

16 Gerald Crompton, “Railway, Governments and Management: A Comparative Approach,” in Railway Management and its Organizational Structure: Its Impact on and Diffusion into the General Economy, eds. Michele Merger and Andrea Giuntini (Sevilla, 1998), 153-167

Álvaro Ferreira da Silva

of time, not completing the lines that they had received in concession17. Railway companies assuming the free-standing form seem to conform to one of the characteristics sometimes attributed to it - short­lived, speculative investments by entrepreneurs who only wanted to secure the concession of a business venture, rapidly selling it to another company with some profit or going bankrupt after pocketing state subventions or financial fees. Hardy Hislop, an Englishman living in Portugal, exemplifies this kind of speculator, acting as a broker between the Portuguese political milieu and the British financiers. He was behind the creation of the Central Peninsular and the South Eastern, as well as two gas companies, the Oporto Gas Company and the Companhia Lisbonense de Iluminação a Gás18.

The period between 1852 and 1859 had been particularly plagued by unsuccessful railway projects, with contracts being granted to foreign companies and individuals and later revoked by the government19. The CRCFP was the first company to materialize the modernization of the transport infrastructure.

It came into being in 1859, when the contracts for the construction and exploitation of a railway line from Lisbon to Badajoz (the Eastern Line) and from Lisbon to Oporto (the Northern Line) were granted to Jose de Salamanca, the well-known Spanish financier20. In 1890 it operated more than 900 kilometers of track, by far the largest railway network in Portugal, but very distant from the largest European railway companies at the time21.

The financing of the company came mostly from French and Spanish capital, through a network of foreign investors, similar to the clusters of financiers and technicians characterizing similar ventures during the period22. On the first 17 Gaspar Fino, Legislação e Disposições Regulamentares sobre Caminhos de Ferro (Lisboa, 1883), 25-37,62-3; Vieira, The Role of Britain and France, 181 ff., 204 ff, 142 ff.

18 Vieira, The Role of Britain and France, 182, 206-7; Ana Cardoso de Matos and Álvaro Ferreira da Silva, “Foreign Capital and Problems of Agency: The Companhias Reunidas de Gás e Electricidade in Lisbon (18901920)”, TST- Transportes, Servicios e Telecomunicaciones, 14 (2008): 143-161.

19 Vieira, The Role of Britain and France.

20 Fino, Legislação e Disposições Regulamentares.

21 Albert Carreras, “Los Ferrocarriles en Europa Algunas Perspectivas Históricas,” in Siglo y Medio, 39.

22 Wilkins, “The Free-Standing Company Revisited,” in The Free-Standing Company 13-14; Geoffrey Jones, “Globalization,” in The Oxford Handbook Of Business History eds. Geoffrey Jones and Jonathan Zeitlin (Oxford: 2007), 153; Gordon M. Winder, “Webs of Enterprise 18501914: Applying a Broad Definition of fdi,” Annals of the Association of American Geographers, 96 (2006): 788-806. Spanish railways also attracted French capital through clusters of foreign investors: Javier Vidal Olivares, “Las compañias ferroviarias y la difusion de las modernas formas de gestion empresarial en España, 1850- 1914,” in La Empresa en la Historia de España eds.

Francisco Comin and Pablo Martin Acena, (Madrid, 1996), 285-301; Francisco Comin et ai, 150 Ahos de Historia de los Ferrocarriles Espaholes (Madrid, 1998). For the role of French capital in railway construction in Mediterranean Europe, Russia and Austro-Hungarian Empire see Rondo Cameron, France and the Economic Development of Europe, 1860-1914 (Princeton NJ, 1961) and Youssef Cassis, Capitals of Capital: A History of International Financial CenTrês, 1780-2005 (Cambridge, 2006), 58.

• 259

Railroads in historical context: construction, costs and consequences

Board of Directors only three out of the 17 members were Portuguese23, the remaining being made up equally of Spanish and French citizens. Besides Jose de Salamanca, the foreign directors represented banks and businessmen connected with the construction of railways, particularly active in several French companies and in the Spanish Madrid-Zaragoza-Alicante (MZA)24. Portuguese shareholders were in a clear minority when the company was launched: of the 70,000 shares issued, only 335 were held by Portuguese investors25.

Table 1 provides the distribution by nationality of the shareholders represented in the shareholders’ meetings after 1876, when the exploitation of the Northern and Eastern lines was already underway. It does not consider all the equity holders, as many of them did not attend or were not represented in the meetings. However, it presents the best possible approximation of the company’s capital structure, identifying the shareholders mobilized for having a “voice” in company decisions.

Table 1: Nationality of the CRCFPs shareholders registered in the General Meetings (1876-1884)

260 •

Nationality

1876

British

1.0

French Portuguese

1877

1878

1879

0.9

0.3

73.7

27.0

63.7

67.8

15.7

17.3

6.0

23.4

Spanish

9.6

54.8

21.9

7.5

Other

0.0

0.0

8.1

0.0

Unknown

0.0

0.0

0.0

100.0

100.0

100.0

Total

Number of shares 10435 11715 30697 represented

1.3

1880 0.9

1881

1882

1883

1884

0.9

1.1

1.5

0.5

47.6

72.7

56.1

63.6

61.3

21.3

13.9

25.9

27.5

20.6

23.9

4.5

5.6

7.4

2.7

0.0

0.0

0.0

0.0

13.5

0.0

6.3

8.0

11.3

0.0

1.4

100.0

100.0

100.0

100.0

100.0

100.0

7963 10880 11164

8874

6730 19126

Source: CNDF/CCFP, Minutes from the shareholders’ meetings (1876-1884)

French shareholders dominated. They were mainly financiers linked to Paris banks, who launched CRCFP with Jose de Salamanca: Edward Blount, a Scotsman banker, resident in Paris, was also in the Boards of Compagnie de I’Ouest, Compagnie Paris-Lyon-Mediterranee, Southern Railway Company in the Habsburg Empire and MZA, respectively the first, the sixth, the twentieth and the twenty-fourth largest European railway companies; Gustave Delahante, 23 Duke of Saldanha (politician and military chief), Baron of Paiva (Portuguese ambassador in France, broker between foreign financiers and the Portuguese political milieu), and Fortunate Chamico (banker).

24 Vieira, The Role of Britain and France, 271 ff and Pinheiro, Chemins de Fer, vol. Ill, 80.

25 Pinheiro, Chemins de Fer, 420.

Álvaro Ferreira da Silva

director at the Banque de Paris and in two other railway firms, Paris Grand Central and MZA; Joseph de Bouillerie, director of the Societe Generate du Credit Industrielle et Come rciale, the main financial partner of the CRCFP. After the 1870s the core of major shareholders evolved, with the arrival of the Societe Generate pour favo riser le developpement du commerce et de I’industrie en France, founded in 1864 and chaired by Edward Blount after 1875, and the Banque de Paris et des Pays Bas (BPPB), recently founded and particularly active in railway investments across Europe26.

The Camondo family represented another strand of French investors. Sephardic Jews, originally from Spain, they lived for a period in Venice before settling in Constantinople in the eighteenth century. In 1872 the family moved to Paris, very close to the time when it entered into the CRCFP, rapidly becoming the largest shareholder. Aside from his family bank, Abraham de Camondo was also one of the main shareholders in Banca Commerciale Italiana, BPPB, Banco Hipotecario de España, the Portuguese Sociedade Geral de Credito Agricola e Financeiro and Andaluces, the Spanish railway company27.

The presence of Spanish shareholders is linked to the creation of the CRCFP’by Jose de Salamanca. In the 1870s he maintained only a residual presence (100 shares), as he abandoned the railway business after the crisis of 186628. Spanish capital was represented mainly by La Gandara29 and La Cuadra30 families, as well as other Spanish capitalists with a lesser number of shares. The 1880s saw, nevertheless, a decrease of the Spanish share in the capital of the company, due to the loss of the Gandara family’s influence, who privileged their railway business in Spain, leaving the Board of Directors in 188231.

In the 1850s, when the CRCFP was founded, the British market was considered the most important source of investment in Portuguese railways. However, Table 1 reveals its irrelevance 20 years later. It seems to follow the trend stressed 26 Cameron, France and the Economic Development of Europe, 197-8.

27 Pinheiro, Chemins de Fer, 281 and “The French Investors in Portuguese Railways,” 142; Nora §eni and Sophie Le Tarnec, Les Camondo, ou L ‘Eclipse d’une Fortune (Paris, 1997); Comin, 150 Anos de Historia, 192; Pedro Tedde de Lorca, “La Compania de los Ferrocarriles Andaluces (1868-1920): Una Empresa de Transporte en la España de la Restauracion”, Revista de Investigaciones Economicas, 12 (1994): 27-76.

28 Raquel Sanchez Garcia, “El Marques de Salamanca y la Amortization de los Ferrocarriles,” Cuadernos de Historia Contemporanea, 25 (2003): 199-215.

29 Joaquim de la Gandara was active in railways, particularly MZA and Andaluces, as well as in Portuguese banking (Sociedade Geral de Credito Agricola e Financeiro).

30 Luis de La Cuadra, Marques de Guadalmina, was associated with Joaquim de la Gandara in several businesses: the Sociedade Geral de Credito Agricola e Financeiro, the MZA, Andaluces and a large agricultural enterprise, the Colonia Agricola de San Pedro Alcantara, in Andalucia. He also had interests in mining and housing (Francisco de los Cobos Arteaga and Tomas Martinez Vara, “Spanish Society of Secondary Networks: The Failure of a Major International Project to Create an Additional Railway Network in Spain,” in Across the Borders, 87).

31 Centro National de Documentação Ferroviaria (CNDF/CCFP), Minutes from the meetings of the Paris Committee, 22 Oct. 1882. Hereinafter this documentation will be identified by the date of the meeting.

• 261

Railroads in historical context: construction, costs and consequences

262 •

by Augusto Fuschini for Portuguese sovereign bonds32: British investors were replaced by French and German ones, signaling the declining confidence in the solvency of the Portuguese State33. The same might have occurred in the foreign investment flows to railways, which depended on decisive financial guarantees by the state.

Portuguese shareholders were always a minority. However, after the late 1870s their participation increased, systematically maintaining more than one fifth of the capital represented in the general meetings of the CRCFP. This anticipated the changes in the power relations within the company that will be addressed later (section 5).

The railway companies developed an elaborated hierarchical system of professional managers. The reasons for this development have been sufficiently emphasized: the dimension of the workforce; the complexity of railway operation, dependent upon an increased flow of information; problems of safety associated with railway exploitation34. Efficient managerial capabilities in the field were needed, explaining why the organizational structure of the CRCFP was similar to companies with identical dimension. However, the relations between local salaried managers and foreign shareholders remain obscure. After establishing the critical importance of foreign investors in the capital structure of the company, the next step lies in understanding how the governance of the firm might secure the reference shareholders property rights in a company whose operations took place thousands of kilometers away.

3. THE GOVERNANCE OF THE FIR M: THE TWO-HEADED BOARD The governance structure of the CRCFP can be established by combining different sources of information: the company Statutes approved in 185935; the directive of bureaucratic procedures (1872), containing a functional description of every managerial position36; a list of the company’s services and personnel (1882)37, also specifying the salary for each post, a useful information that discloses a hierarchy of responsibilities.

The statutes remained unchanged until 1885, which means that the governance structure created when the Company was founded proved to be adequate for a quarter of a century. They established that the company would have a 32 Augusto Fuschini, Liquidações Politicas: Vermelhos e Azues (Lisboa, 1896).

33 Rui Pedro Esteves, “Finanças Públicas,” in Historia Economica de Portugal, II, eds.

Pedro Lains and Álvaro Ferreira da Silva (Lisboa, 2005), 318-9 34 Alfred D. Chandler Jr., The Visible Hand: The Managerial Revolution in American Business (Cambridge, Mass., 1977).

35 Estatutos da Companhia Real dos Caminhos de Ferro Portuguezes (Lisboa, i860).

36 CNDF/CCFP, “Ordem de Serviços n.° 2”, 17 Aug. 1872.

37 CNDF/CCFP, “Cadre du Personnel. Compagnie Roy ale des Chemins de FerPortugais”.

Álvaro Ferreira da Silva

Board of Directors of sixteen to twenty members, where half should be Portuguese or Spanish nationals. Although not as large as the boards in other railway companies38, the top layer of CRCFP follows the French model, privileging a large number of Directors39.

The headquarters of the company were in Lisbon, where the Board of Directors had its official seat. The composition of the Board in 1882 still reflected the multinational origin of the capital: nine directors had their residence in Paris, six in Lisbon, and four in Madrid. The existence of a delegation in Paris is foreseen in the statutes, demonstrating the relevance played by French investment in the creation of the company40. The importance of foreign directors as representatives of reference shareholders was recognized and guaranteed. The Board in Lisbon had no decision power on a number of subjects, without the formal agreement of the foreign Directors. Contracts with patrimonial impact, setting tariffs, relations with the Portuguese government, the strategy for exploiting the railway lines or the establishment of the yearly budget were some of the subjects in which any decision should be preceded by consultation with all directors and their written position (Statutes, article 26). Any document circulating above the General Manager should be written in French, so that foreign directors could be up to date with the information about the company.

The governance structure that could be drafted from the statutes is very different from reality. The Chairman of the Board of Directors (the Count of Camondo in 1882) was located in Paris and headed the so-called “Paris Committee”41. This designation, absent from the statutes, points to the existence of a formal body and corresponds to its effective integration within the organic structure of the company (see Figure 1, with the reconstructed organizational chart from 1870 to 1883).

The statutory formula does not give formal existence to the Paris Committee. The Board of Directors based in Lisbon was the only top management body and the directors living abroad would participate by means of an individual consultation and vote. The “delegation” seems to have an actual role far more influential than what may be deduced from the statutes’ normative account.

Besides the presence of the Chairman of the Board in Paris, the number of its members reveals the second sign of the Committee’s importance. The Paris Committee also included one executive director, a position occupied in 1882

by Luis de la Cuadra, Marquis of Guadalmina, and earlier by his business as38 in 1879 the Boards of the Compania de los Ferrocarriles del Norte de Espanha (Norte) and MZA had, respectively, 40 and 25 directors (Javier Vidal Olivares and Pedro Pablo Ortunez, “The Internationalisation of Ownership of the Spanish Railway Companies, 1858-1936,” Business History, 44 (2002): 29-54).

39 Michele Merger and Andrea Giuntini, “L’ organisation des Chemins de Fer Italiens: Un ou Des Modeles (1839-1939)?” in Railway Management and its Organizational Structure: Its Impact on and Diffusion into the General Economy eds. J. Armstrong eta/., (Sevilla, 1998), 153-167.

40 Estatutos, articles 12 and 13.

41 CNDF/CCFP, “Cadre du Personnel...”; Minutes from the meetings of the Paris Committee; correspondence between the Paris Committee and Lisbon.

• 263

Railroads in historical context: construction, costs and consequences

Figure 1 - CRCFP’s organizational chart, 1870-1883 France

Chairman Executive Director + 8/9 directors Consulting engineer Accounting and financial staff Clerical staff

Committee of Paris Portugal Directors (Lisbon)

Executive Director 5/6 directors

General Manager

Accounting

264 •

Control

Health

Traffic

Movement

Rolling Stock

Works

Store Houses

sociate, Joaquin de la Gandara. He was in charge of the relations between the Committee and Lisbon, as well as any business abroad. Throughout the period under study, it was this director who came to Lisbon at particularly important times, where the use of the telegraph or the postal services to convey information and instructions was not appropriate42. Therefore, even if face-to-face communication was not very frequent, it was not absent, contrarily to what happened in similar ventures43. The important role performed by this executive director is recognized by his salary, the third highest in the company (15,000 French francs, 50% more than the Chairman).

The Paris Committee also included an engineering consultant (Sosthene Le François44), for dealing with the technical aspects of the construction and exploitation of the railway network. He attended the Paris Committee meetings and at times was sent to Lisbon for the implementation of decisions taken in Paris. He also received a top salary, on par with that of the executive director in Paris.

Further technical and administrative staff demonstrates the relevance of the functions concentrated in Paris. Besides providing clerical support, the eleven members of the staff have accounting expertise in three areas: general accounting, equity and debt management, and financial investment. The accounting of the company was carried out in Paris, based on data provided by Lisbon45. The reorganization of the company’s accounting system in 1883 provides a clear 42 Meetings of the Board, Lisbon, 12 Ap. 1876, 18 Jan. 1881, 6 Nov. 1883, 14 June 1884, for instance.

43 Boughey, “British Overseas Railways,” 495.

44 Also in other railway companies: Sociedade de Caminhos de Ferro Madrid-Cáceres-Portugal (MCP) and Andaluces.

45 Memorandum written to the government by the Board in Lisbon, in 1885: “the accounts are made in Paris by the bankers of the company, the Société Générate de Crédit Industrie et Commerciale (Fino, Legislação e Disposições Regulamentares, 271).

Álvaro Ferreira da Silva

sign of the leading position of Paris in financial and management accounting46.

All of the improved control system was developed there and the Lisbon office had only a passive position47. Thus, the Paris office did not have the tiny dimension and virtually non-existent activities signaled in other cases48, leading to inferences that home headquarters were “little more than a brass nameplate”49. It appears in regular correspondence with Lisbon and was fully operational.

The formal Board in Lisbon had a smaller staff (only five employees), with merely administrative duties: organization of correspondence and business records, or taking minutes at the meetings of the Board. Its functions were far less extensive and complex than those carried out by the dozen employees in Paris.

A second executive director, who also monitored local railway operation, guaranteed the connection between Lisbon and Paris. This position had the second largest stipend in the CRCFP, exceeded only by the General Manager (below), thus highlighting his executive responsibilities. It was held by Osborne Jacques de Sampaio between 1875 and 1884. He had a career in the top management of railway companies: firstly as executive director in the Compagnie du Chemin de fer du Grand Central and later in the Spanish MZA50.

The everyday operation of the company in Portugal was under the responsibility of the General Manager. According to the statutes, he was responsible for all the departments of the CRCFP, running the company’s personnel and local operations, but it was the Board that should approve every important decision51.

The revenue accompanying this position places it at the top of the salary hierarchy (25,000 French francs). It was a post always occupied by engineers: until 1871 by French engineers - Goudchaux (1865-1870) and Sosthene Le Francois (1870-1871) - and after by Manuel Afonso Espregueira (1871-1884), a Portuguese engineer graduated in the École des Ponts et Chaussées, having an important role in the reorganization of local operations during the 1870s52.

The CRCFP shows a two-headed directorship, a dual-board structure, not at all evident in the formal governance of the company. It is thus important to determine the real powers of the Paris Committee, its relations with the Board in Lisbon and with the operational management of the company, symbolized by the General Manager.

46 Meetings of the Committee, 4 and 18 Sep. 1883, when the accounting reorganization was discussed.

47 The accounting service in Lisbon revealed serious difficulties in adapting to the new rules and a clear dependence on accounting undertaken in Paris office (Meeting of the Committee, 4 Oct. 1883).

48 Boughey, “British Overseas Railways”; T. A. B. Corley, “The Free-Standing Company, in Theory and Practice” in The Free-Standing Company, 129-147.

49 Stephen Nicholas, “British multinational investment before 1939,” Journal Of European Economic History, (1982): 606; Mira Wilkins, “The Free-Standing Company”, 264.

50 Miguel A. Lopez-Morell, La Casa Rothschild en Espaha (Madrid, 2005), 164.

51 Article 28 from the Statutes and “Ordem de Servicos n.° 2”, 17 Aug. 1872 (CNDF/CCFP).

52 Álvaro Ferreira da Silva, “Engineers and Organizational Behaviour: The Companhia Real dos Caminhos de Ferro Portugueses (1870-1885),” The Quest for a Professional Identity: Engineers Between Training and Action, eds. Paula Diogo eta/. (Lisbon, 2009), 329-342.

• 265

Railroads in historical context: construction, costs and consequences

4. THE ROLE OF THE PARIS COMMITTEE

266 •

The decisions taken by the Paris Committee define the degree and the type of control exercised by the Committee over the management of the company. The frequency of the meetings were on average every three weeks, with some periods when meetings were more frequent, providing a first hint on the close monitoring that the directors seating in Paris maintained. They were not holding a sinecure position53.

The company strategy was defined entirely in Paris, and afterwards presented and ratified in Lisbon. The internationalization of the company after the second half of the 1870s is a clear hallmark. In September of 1876 several meetings of the Paris Committee decided the essential vectors of the move54: the company should promote a more rapid connection between Lisbon and Madrid through the construction of the Caceres branch-line55, securing an alliance with the small Spanish company Malpartida-Caceres, in order to construct the connection between Caceres and the Portuguese frontier. It aimed the creation of an alternative to Spain, independent from the two Eastern connections, dominated by the largest Spanish railway companies (Norte and MZA). The meeting of the Board of Directors in Lisbon that ratified this internationalization strategy was held with the presence of the executive director from the Paris Committee, de la Gandara, who travelled to Lisbon and provided the information about the decisions taken in Paris56. Another move tried to find stable sources for freight in the new line, through a contract with the Companhia Geral dos Fosfatos of Cáceres, giving the CRCFP exclusive rights to transport phosphate. The use of the new line provided a faster and cheaper connection to the European markets through the port of Lisbon. Once again, the strategy was delineated in Paris57. Once again Gandara was sent, in person, to Portugal to explain the decisions taken in Paris and to mobilize the Board in Lisbon to implement them58.

Almost all the directors in the Paris Committee were partners in the Spanish company (Malpartida-Cáceres), which was to provide the construction of the branch-line between Caceres and the Portuguese frontier, and which later would give rise to the Companhia Madrid-Cáceres-Portugal (MCP), following its merger with the Companhia de Caminhos de Ferro del Tajo59. The cross53 Corley, “The Free-Standing Company”.

54 The strategy was discussed in the meeting of 13 Sep. 1876 and decision was taken on 19 Sep. 1876.

55 The earlier connection with Madrid was made via the Elvas-Badajoz frontier, increasing the distance between the two Iberian capitals in more than 200 kilometres.

56 Meeting of the Board, Lisbon, 28 Sep. 1876.

57 Meeting of the Committee, 12 Dec. 1876, which tackles this theme for the first time. The meetings of the Committee on 14 Feb. and 7 Mar. 1877 analysed the different components of this strategy.

58 Meeting of the Board, Lisbon, 12 Ap. 1877.

59 Gandara (39.5%), Camondo (20%), La Cuadra (12.5%), Delahante (10%), de la Bouillerie (7%), Blount (7%), the engineer of the Paris Committee, Sosthene Le Francois (4%). Only two members of the Paris

Álvaro Ferreira da Silva

participation between the boards of directors of the two companies led to an uncommon situation, where the negotiations with the Companhia Geral dos Fosfatos de Cáceres were undertaken by Joaquin de la Gandara and La Cuadra, as representatives both of the CRCFP and MCP60. Technical personnel and managers also had the same interlocking, with the engineering consultant of the Paris Committee serving as executive director of the Spanish company after 187961.

The strategy was later pursued through new contracts aimed at integrating the CRCFP in a network of intermodal connections with domestic or foreign railway lines, shipping companies through the port of Lisbon (E. Grosos, Havre; John Hall Junior & Cie, London; Burrell & Son, Glasgow) or transport services (contract with Wagons-Lits)62. All these contracts were decided upon and carried out in Paris.

Critical decisions were approved of in Paris, whether it might have been a new strategy, new contracts, the issue of new bonds63, business alliances and loans to other companies64, without consulting the Board in Lisbon, which was only informed a posteriori. Other decisions relevant to railway development and operation were equally taken in Paris, but in this case there was a prior consultation with the General Manager and the directors in Lisbon. The definition of new tariffs for the transport of phosphates from Caceres is a clear example of this situation65. But even matters apparently less relevant and strictly concerned with the operational management in Portugal were taken in Paris, such as the regulation of the railway timetables for connections with MCP, or the fines to be applied to train drivers, when responsible for delays66.

The importance of the Paris Committee in the strategic management of the company is evident. Some matters easily classified within the realm of operational management were decided by the headquarters in Paris, when the relation with strategic issues was decisive, such as the operation of the line running to the Spanish border. The Board of Directors in Lisbon had a second-rate status in this two-headed top-management structure, formally approving decisions alCommittee were not shareholders of the Spanish railway company: E. Joubert (but in the Board of the Sociedade dos Fosfatos de Cáceres) and the Marquis of Scepeaux. Relations with the Companhia del Tajo became very close after the decision of the Paris Committee (8 Oct. 1878) to proceed with a loan to the Spanish company. For the evolution of the Spanish company see: Comin et al., 150 Años de Historia, 171 ff, Juan José Cendal Búrdalo, “Compañia de explotación de los ferrocarriles de Madrid a Cáceres y Portugal y del Oeste de España,” in Siglo y Medio, 563-583 and Francisco Wais, Historia de los Ferrocarriles Españoles (Madrid, 1974), 375 ff.

60 Meeting of the Committee, 18 June 1877.

61 Meetings of the Committee, 27 Feb. and 8 Ap. 1879.

62 Meetings of the Committee, 23 Nov. 1881 and 21 June 1882.

63 As an example, meeting of the Committee, 14 Oct. 1880.

64 Examples of loans to other firms: the Companhia del Tajo, as well as companhia Malpartida-Cáceres (meeting of the Committee, 18 Jan. 1881).

65 Meeting of the Committee, 2 Mar. 1881.

66 Meeting of the Committee, 7 Dec. 1883

• 267

Railroads in historical context: construction, costs and consequences

268 •

ready taken in Paris, monitoring the operational management of the railway, guaranteeing that all the relevant information was sent to Paris, and developing contacts with the Portuguese government and administration.

Changes in the ownership of the share capital were reflected in the composition of the Committee. The representatives of the financial institutions that supported the CRCFP remained permanently in the Paris Committee until 1885.

From the beginning of the 1880s, the Camondo family had a weighty presence on the Board of Directors, replacing the Gandara family, who had held the presidency and executive administration. While this was happening, the Board of Directors in Lisbon remained almost unchanged since the beginning of the 1870s and until 1884, with the position of the executive director being always occupied by Osborne Jacques de Sampaio. This was a further sign that the real center of power of the two-headed Board of Directors was located in Paris.

Similar dual-board structures existed in other industries67. They were ubiquitous in the three largest Iberian railway ventures, where French capital was well represented68, as well as in other countries69. However, the existing literature emphasizes the managerial tasks assumed by local boards. It is argued that foreign committees usually had the last word on important issues related to finance70, but control mechanisms are not studied in detail. In short, more attention has been paid to formal governance rules than to actual managerial behaviors.

5. CRISIS AND OR DER: UNVEILING CONTROL AND AUTHORITY RELATIONS WITHIN THE FIRM Periods of crisis and disruption are privileged times to look behind the norma67 In the banking industry see the case of the Imperial Ottoman Bank (Philippe L. Cottrell, “The Coalescence of a Cluster of Corporate International Banks, 1855-75,” Business History, 33 (1991): 31- 52). For electrical utilities, Matos and Silva, “Foreign Capital and Problems of Agency”.

68 Comin et al., 150Ah~0S de Historia; Cuellar Villar, “El ferrocarril en España”; Lopez-Morell, La Casa Rothschild; Miguel A. Lopez-Morell and Jose M. O’Kean, “A Stable Network as a Source of Entrepreneurial Opportunities: The Rothschilds in Spain, 1835-1931,” Business History 50 (2008): 163-184; Munoz Rubio et al., Siglo y Medio; Vidal Olivares and Pablo Ortunez, “The Internationalisation of Ownership”.

69 J. Armstrong et al., Railway Management and its Organizational Structure: Its Impact on and Diffusion into the General Economy (Sevilla, 1998); B. Barth and J. C. Whitehouse, “The Financial History of the Anatolian and Baghdad Railways, 1889-1914,” Financial History Review, 5 (1994): 115-37; Cameron, France and the Economic Development of Europe; Philippe L. Cottrell, “A survey of European investment in Turkey, 1854-1914: Banks and the finance of the state and railway construction,” in East meets West eds. P. L. Cottrell et al. (Aldershot, 2008), 59-95; Roth and Dinhobl, Across the Borders, Albert Schram, Railways and the formation of the Italian state in the nineteenth century (Cambridge, 1997).

70 Comin et al., 150Ah0S de Historia, 179; Lopez-Morell and O’Kean, “A Stable Network,” 177.

Álvaro Ferreira da Silva

tive rules in order to discover the actual control practices within organizations. In the mid-1880s the CRCFP went through a period of serious conflict for several years. The events are here presented in a condensed way71.

In 1884 the CRCFP negotiated with the Portuguese government the concession of a new line (Beira Baixa railway line). When the contract had already been drafted, the Parliament approved an additional clause ruling that a majority of Portuguese citizens should seat in the Board of Directors. It sparked a reaction from the company against the change in governance72.

What was in motion was a battle for power within the company, clearly emerging in the shareholders’ meeting convened to discuss the new law. In an unprecedented decision, only the proxies of foreign shareholders recognized by the Portuguese embassies were accepted as valid votes. This maneuver put out much of the foreign shareholders’ representation. The Portuguese shareholders got the majority of the votes, approved the change of the company statutes and elected a new Board, where Portuguese directors had the majority (12 against 10 foreign directors), and without the previous president, Camondo, who was substituted by Henrique Eugenio Macieira, a director of the Banco Lusitano. Almost all the elected foreign directors refused to seat in the new Board.

In 1885, a new revision of the statutes tried to reconcile the two groups, maintaining the guarantees to foreign shareholders and formally recognizing the Paris Committee. However, the balance of power did not revert to the previous situation and pacification was impossible. The foreign directors were systematically forgotten in decisions affecting the management of the company.

The two-headed organization that had been in place in the company for a quarter of century came to an end with the resignation of the Paris Committee in 1887.

The powers formerly vested in the foreign managers based in Paris lost their reason to be maintained after the Committee had been stripped away from its extensive and informal managerial control. The statutory review of 1887 removed any attributes assigned to the Paris Committee and moved managerial control to the Board in Lisbon. Informally, the center of decision had already moved to Lisbon in 1884.

These events can only be understood in the light of an accumulation of tensions within the organization’s two-headed Board, revealing the relations of power and control before 1884. International expansion and insertion into an intermodal network characterized the strategy of CRCFP after the late 1870s, met with reservations in Lisbon73. The subordination of the domestic operational management approved by the Paris Committee triggered the crisis. The persistent delays in the schedule of the Lisbon-Madrid train led to the creation of a new post of salaried top manager (General Inspector of Operation) at the end of 1883, 71 See also Pinheiro (Chemins de Fer) and Salgueiro (A Companhia Real) for an account of these events.

72 Meeting of the Board, Lisbon, 10 Sep. 1884.

73 CNDF/CCFP, letter 30 Jan. 1877 from Espregueira to the executive director in Paris, La Gandara.

• 269

Railroads in historical context: construction, costs and consequences

occupied by a French engineer, nominated by Paris74. This would create two top managerial positions in Portugal (see figure 2): the General Manager (Manuel Afonso Espregueira) would oversee the General Services; the new General Inspector of Operation would direct the Technical and Operational Services (Rolling Stock, Movement, Maintenance and Works, Storehouses), crucial for the multimodal operation75. It constituted the final straw in the increasingly tense relations between the Paris Committee and the operational management in Portugal. In June 1884, in the span of three days, the General Manager, the deputy engineer responsible for construction (Pedro Inácio Lopes) and two members of the Board in Lisbon (Francisco de Oliveira Chamico and Fortunato Chamico Junior) presented their resignation.

Figure 2 - CRCFP’s organizational chart, 1883-1884

Committee of Paris Directors (Lisbon)

Chief Engineer of Construction

270 • General inspector of Operation

General Manager

Accounting

Control

Health

Traffic

Movement

Rolling Stock

Works

Store Houses

Behind the strategy of expansion lay motivations related to the interests of the company’s shareholders in Spanish railway companies. The shorter connection to Madrid and the intermodal agreements allowed an appreciation of assets with limited worth when taken separately. Other motivations were present in the expansion into Spain. For reference shareholders incentives for investing in railways did not lie primarily in traffic profits, but in the earnings obtained in the construction phase76. The concession contracts with railway companies were based on a state subsidy per kilometer, difficult to monitor and determine 74 Meeting of the Board, Lisbon, 10 Nov. 1883.

75 Silva and Gomes, “A Companhia Real”.

76 See Philip Keefer, “Protection Against a Capricious State: French Investment and Spanish Railroads, 18451875,” The Journal Of Economic History, 56 (1996): 170-192, and Pedro Tedde de Lorca, “Las compañias ferroviarias en España, 1855-1935,” in Los ferrocarriles en España, 1844-1943, vol. 2, ed. M. Artola (Madrid: 1978), 118 ff.

Álvaro Ferreira da Silva

by public authorities and thus subject to tough bargaining77. The presence of financiers and engineers with experience in railway construction was an essential requirement, since cost evaluation and project design was critical in these site-specific projects with an idiosyncratic component. The construction phase could have lead to important opportunities for extra profits derived from project management or it could have been an open door to fiasco, if costs were poorly calculated and the contract with the government poorly designed (Eichengreen, 1996: 120). Finally, the banks involved in the placement of debt or equity had other means to channel revenues to reference shareholders. Underwriting fees and other commissions were an additional motivation for the involvement of banking institutions in railway investment78.

Along with the dissatisfaction regarding the strategy of expansion into Spain and the decreasing of autonomous top management by Portuguese engineers, the Companhia Real became a tempting target for those interests wanting to capture the revenues of financial services and project management. Moreover, the stabilization of railway operations after the late 1870s generated a high volume of cash flow, fuelling further ambitions to control them by Portuguese financial institutions, particularly those aiming at a rapid and leveraged growth. A more strictly political dimension could be added - a “nationalized” and more compliant company could offer additional comfort to any government, through shortterm loans to meet recurrent financial difficulties79. In 1887, the decision of the Board in Lisbon to lend 10 million francs to the Portuguese government was the most flagrant example of this use of the company funds80. It triggered the final rupture between the Board of Directors in Lisbon and the Paris Committee, who resigned in July 188781.

The identification of the protagonists supports this interpretation. The mobilization of Portuguese shareholders and the preparation of the scheme that prevented the acceptance of foreign shareholders’ proxies were carried out by financiers from the Banco Lusitano and the Sociedade Geral de Crédito Agricola e Financeiro, minor institutions in the Portuguese banking system, but aspiring to a rapid growth and having close ties to the Progressive Party. The Member of Parliament who presented the clause imposing Portuguese majority in the Board of Directors was Mariano de Carvalho, one leading figure in the Progressive

77 Frederico Pimentel {Apontamentos para a História dos Caminhos de Ferro Portugueses (Lisboa, 1892): 43), reviewing the guidelines followed by the Portuguese state for promoting railway construction.

78 See the conclusions of the report presented to the Board, when the Paris office was seized by the new administration, after the resignation of the French directors (meeting of the Board, Lisbon, 27 Nov. 1884).

79 It happened on several occasions, even before the crisis. Between 1871 and 1881 these operations accounted 1,812,000$000 (Salgueiro, A Companhia Real, 52). But after 1885 these loans became more frequent and large (meeting of the Board, Lisbon, 10 Sep. 1885 and 16 June 1886).

80 Meeting of the Board, Lisbon, 23 June 1887.

81 Meeting of the Board, Lisbon, 3 Aug. 1887.

• 271

Railroads in historical context: construction, costs and consequences

272 •

Party and future Minister of Finance82. He was elected to the company’s Board in 1884, together with representatives of the Portuguese banks just mentioned.

The external connections were done by two French engineers, with experience in public works’ contracts, namely the Suez Canal and railways: Charles Cottard and Edmond Bartissol83. The latter was actively involved in several public works in Portugal (port of Leixões, Beira Alta railway line) and had links to the Societe Financiere in Paris84. The objective was to dislodge the French banks with influence in the company to access underwriting fees and project management earnings.

The last act of this crisis occurred when the operating income decreased, following the economic recession that affected the Portuguese economy (Figure 3)85. New debt had to be issued to rollover previous financial responsibilities.

Between late and mid-1891 the CRCFP issued 65,500,000 francs in bonds, increasing the burden of debt in more than 50% when compared to 188486. The financial difficulties of the Banco Lusitano made things even worse, as it owed large sums to the company, thanks to the peculiar symbiosis between the two institutions since 1884. It was the end of was of defections in the Board of Directors and in January 1892 a financial scandal affected another protagonist of the 1884 takeover. Mariano de Carvalho had been appointed as Minister of Finance, after several years as director of the company. In 1891 the CRCFP was unable to pay the interest to bondholders. In order to prevent its financial collapse, the minister advanced a large sum to the company, without the knowledge of his government colleagues87. The political scandal precipitated the financial crisis: in February 1892, the Companhia Real suspended payments to bondholders and the protagonists of the 1884 takeover remaining in the company were removed from the Board88.

It sounded the death knell for the program of railway construction, initiated in the 1850s. Portugal’s accession to the gold standard in 1854 and the availability of capital in European markets raised the appeal for the Portuguese sovereign debt, as well as for bonds issued by utilities and concessionaire companies operating in Portugal, a country that had been far from exemplary in meeting its 82 Paulo Jorge Fernandes, 0 Poder Oculto, Mariano Cirilo de Carvalho. 0 poder oculto do liberalismo progressista (1876-1892) (Lisbon, 2010).

83 Dominique Barjot, “Les Entrepreneurs Francais de Travaux Publiques et l’Equipement du Portugal: Une Contribution Multiforme (Milieu du XIX Siecle-Milieu des Annees 1970),” Ler Historia, 26 (1994): 93-116; Jean-Louis Escudier, “Itineraire d’un Entrepreneur de Travaux Publics Eclectique: Edmond Bartissol (18411916),” Histoire, economie et societe, 14 (1995): 229-251.

84 Pinheiro, Chemins de Fer, 480 85 Figure 3 presents the operating profitability of CRCFP, before netting out capital costs, representing the ratio of operating receipts per kilometre to expenses per kilometre.

86 Meetings of the Board (Lisbon) from September 1890 to June 1891 (CNDF/CCFP). The burden of debt comes from Pinheiro, Chemins de Fer, 495.

87 Fernandes, 0 Poder Oculto.

88 Fino, Legislação e Disposições Regulamentares.

Álvaro Ferreira da Silva

Figure 3 - Companhia Real, evolution of the operating ratio (1865 - 1­ 893)

Operating ratio

3 2,5 2 1,5 1 0,5 0 1865 1867 1869 1871 1873 1875 1877 1879 1881 1883 1885 1887 1889 1891 1893

external commitments for much of the nineteenth century89. The beginning of the 1890s was dramatic for the Portuguese economy: the decline in the emigrants’ remittances due to the political situation of Brazil imposed a strong pressure on the balance of payments; a banking crisis, generated by the fall in remittances and the difficulties to get financing abroad, started a bank run, which in turn precipitated the abandonment of the fixed exchange rate regime in 1891; the internal political situation (British Ultimatum in 1890, republican revolt of 31 January 1891) and the abandonment of the seal of approval represented by the gold standard launched a wave of distrust on foreign creditors about the State’s ability to meet its financial commitments. On 13 June 1892 the Portuguese government decided on a partial default, unilaterally reducing two thirds of the foreign debt interest. The Saint-Simonian policy of the Regeneration had come to an end.

6. OWNERSHIP AND CONTROL IN FARAWAY VENTURES: DISCUSSION AND REMARKS The CRCFP did not possess the formal characteristics of the free-standing company. However, in reality, even a company registered in the country hosting foreign investment and having its headquarters there could be organized in a similar way. This demonstrates the benefits of following a methodology more attentive to the practices than to the rules, privileging the information that may reconstruct business decision-making processes. It does not matter if the foreign company was registered at the home or host country. The statutory rules regard89 Esteves, “Finanças Públicas”.

• 273

Railroads in historical context: construction, costs and consequences

274 •

ing the location and functioning of the Board are not critical. Decisive is how the decision-making processes and the governance structure were arranged, where decisions were taken and authority exercised.

The Paris Committee did not perform a passive role in the managerial structure of the CRCFP. It was not a mere financial vehicle, without further control and intervention over the activities of the firm. It was neither a mere repository of financial information to keep foreign stockholders informed about the railway venture. Management and financial accounting were used to provide the information needed to run the firm from Paris. There, the Committee coordinated the strategy, designed the concession contracts for the construction and exploitation of railway lines, decided on contracts with business partners, took all the financial decisions and preserved the control of the company. The 1884 crisis illustrates the importance of the Paris Committee: a boardroom coup was needed to strip away its largely informal competences. It was the Board of Directors, statutorily located in Lisbon, which had a passive role, with the exception of its executive director. The General Manager retained the operational authority regarding day-to-day management, but the Paris Committee closely monitored operational decisions considered critical to the overall strategy.

The Paris Committee resorts to standard procedures and hierarchical means of corporate control and information transmission, using accounting and statistical data provided by Lisbon, correspondence, reports and direct instructions. In critical times, face-to-face communication was used as the most efficient means.

It did not rely on delegating extensive responsibility to agents placed abroad or on the socialization strategies shaping the “clan orientation” identified by Geoffrey Jones, in which owners, managers and agents share common objectives and culture, a sense of trust and belonging, as substitutes to more hierarchical and bureaucratic mechanisms of control90. Lesser distance (geographic, but also political and institutional) and a more uniform business venture may explain the more standardized mechanisms of control.

Many railway companies across peripheral Europe are organized in away similar to the CRCFP, as Rondo Cameron’s analysis of French investment in the continent profusely reveals91. Large Spanish railway companies are particularly interesting for a comparative approach, as they relied on French capital and displayed dual-board structures. However, the actual mechanisms of control are difficult to trace from the existing literature. The importance of local management is emphasized against the role of the foreign committees92, even if conflicts 90 Geoffrey Jones, Merchants to Multinationals: British Trading Companies in the Nineteenth and Twentieth Centuries(Oxford, 2000).

91 Cameron, France and the Economic Development of Europe. Also Roth and Dinhobl, Across the Borders.

92 Javier Vidal Olivares, “La estructura de la propiedad, de la organization y la gestion de una gran empresa ferroviaria: la Compania de los Caminos de Hierro del Norte de España, 1858-1936,” Revista de Historia Economica, 3 (1999), 638. However, Lopez-Morell and O’Kean (“A Stable Network as a Source of Entrepreneurial Opportunities,” 177) suggest that in its early years MZA’s

Álvaro Ferreira da Silva

between foreign and local boards might reveal a more proactive stance from the foreign board93. Further research should explore a comparative perspective on these dual-board structures.

Doubts on equating place of registration to headquarters are clearly confirmed in this study94. Even more important for an analysis of international business over time is the danger of equating formal headquarters and managerial control. In the present study and in similar dual-board firms, strategic control is not paired with either the place of registration or the formal head office. If the free-standing company may stand as a theoretical innovation emerging from business history, a broader definition should be used instead of formal registration and headquarters, comprising the business ventures created “to undertake business activities (...) abroad without prior domestic business”95, specializing in a single product or service and operating in a single country.

Could the location of control evolve over time in dual-board firms? The evidence collected in to this study provides some tentative answer to this question.

In the initial period of a business venture like nineteenth-century railway companies, characterized by project management issues, it was more efficient to have the control located in the home country. Besides channeling capital, only the foreign board could assemble technological expertise, synthesize it into a project solution and manage the procurement of equipment and services. When railway network was stabilized and operation running smoothly, the efficient use of the investment could place much more weight on local management96. This evolution might run without serious conflicts through a progressive “domestication of management”, as suggested by the literature about Spain97, following the general trend to halt railway investment as an international business by the 1920s98. In other cases - like CRCFP- the “nationalization of control” was achieved overthrowing the foreign top management through dubious means. Similar processes are documented in other infrastructure ventures, intermingling economic, political and social reasons in a complex web of causes99. It may suggest another window to look at patterns of evolution in international business by industry and region. strategic decisions were taken in Paris.

93 Cobos Arteaga and Martinez Vara, “Spanish Society of Secondary Networks,” 84; Keefer, “Protection Against a Capricious State,” 180; Vidal Olivares, “La estructura de la propiedad,” 647 ff 94 Harm Schroter, “Continental European FSCs: The case of Belgium, Germany, and Switzerland,” in The FreeStanding Company, 334 ff.

95 Geoffrey Jones, Multinationals and Global Capitalism (Oxford, 2005), 21.

96 Wilkins (“The Free-Standing Company Revisited,” 23) and Casson (“An Economic Theory of the Free standing Company,” 104) mention that different stages in the development of local operations might lead to the elimination of foreign headquarters.

97 Vidal Olivares, “Las compañias ferroviarias”; Vidal Olivares, “La estructura de la propiedad”; Vidal Olivares and Pablo Ortunez, “The Internationalisation of Ownership”.

98 Wilkins, “Multinational Enterprise to 1930,” 78 99 Wilkins, Hausman and Hertner, 2008: 258 ff

• 275

Railroads in historical context: construction, costs and consequences

276 •

Should the extended managerial functions attributed to the Paris office qualify the company as a multinational enterprise? In the case of CRCFP there was a cross-border investment of capital and managerial resources through direct control over foreign operations. This characterization does not leave any doubts as to classify these flows as direct investment as opposed to portfolio investment. However, in order to be eligible as a multinational enterprise other criteria should be taken into consideration, following Casson’s suggestion that sometimes “the usual connection between foreign direct investment and the multinational enterprise breaks down”100. If the multinational enterprise can be defined as the firm that owns and controls assets and activities in more than one country, then the basic question is to know if the Paris office qualifies as a separate activity. Even if mostly undertaken by part-time engineering, accounting or clerical staff, as well as part-time and non-executive directors, the Paris office was the real headquarters of the firm, providing the expected top-management functions. In addition, and on an on-going basis, it recruited personnel, purchased equipment and services, accessed consulting services on arm’s length, when inhouse expertise was not available. Geoffrey Jones101 analyzed a similar role in the case of British banks operating overseas, where modest head offices provided not only capital, but also managerial expertise and recruitment services, benefiting from the comparative advantage European metropolises had in the intermediation of capital, goods, information and other intangible services102.

This access to sources of capital and expertise would not be possible in Portugal or other peripheral European countries and overseas regions. As a result, a firm like CRCFP is based on foreign direct investment because the Paris board exercised control over its operations and is a multinational firm because the foreign headquarters performed value-added functions.

There is a further dimension of CRCFP as a multinational. Its foreign topmanagers, financiers and technicians participated in similar firms in other European countries. Networks of promoters, financiers, engineers and consultants were involved in the creation of railway companies across Europe and overseas, as well as other business ventures characterized as free-standing companies.

In this sense these business undertakings were not the kind of atomistic firms implied by the definition of free­standing company103. These apparently isolated ventures reveal its entrepreneurial nature, taking advantage of the business opportunities opened up by the late nineteenth-century global economy and doing that through flexible organizational forms between webs of economic agents, 100 Mark Casson, “An Economic Theory of the Free-Standing Company,” in The Free-Standing Company, 106.

101 Geoffrey Jones, British Multinational Banking, 1830-1990 (Oxford, 1993) and “British Overseas Banks as Free-Standing Companies 1830-1994,” in The Free-Standing Company. See also Lynn Hollen Lees, “International Management in a Free-Standing Company: The Penang Sugar Estates, Ltd., and the Malayan Sugar Industry, 1851-1914”, Business History ReviewZl (2007): 27-57.

102 Casson, “An Economic Theory of the Free-Standing Company,” 125-126.

103 Geoffrey Jones labels the free-standing company as a misnomer (“British Overseas Banks,” 345).

Álvaro Ferreira da Silva

much in the same way as similar flexible solutions are coping with nowadays global economy104.

This study concludes that dual-board firms share the characteristics of free­ standing companies, even when registration and formal headquarters were located outside. Besides underlining the polymorphous character of the free-standing form, this conclusion comes in direct relation with the research agenda once stated by Wilkins, placing stronger emphasis on the functions of home office and problems of ownership and control105. As she stated, “capital was not exported in a (...) vacuum, but rather it was embodied in a business organization”106.

All the questions on how managerial decisions were taken, the characteristics of the governance structure, the functions of the head office and its relations to operations abroad were raised in this paper on the assumption that organization matters. This is one of the most important traditions on the historical analysis of the firm.

• 277

104 Mark Casson and Howard Cox, “International Business Networks: Theory and History,” Business and Economic History 1 (1993): 42-53.

105 This statement is pervasive in many of the contrbutions to Wilkins and Schroter, The FreeStanding Company.

106 Wilkins, “The Significance of the Concept and a Future Agenda,” 423.

José Manuel Lopes Cordeiro

THE MAN BEHIND TUA RAILWAY: CHIEF ENGINEER DINIS MOREIRA DA MOTA O HOMEM POR TRÁS DA LINHA DO TUA: O ENGENHEIRO CHEFE DINIS DA MOTA José Manuel Lopes Cordeiro (U. Minho, Portugal) José Manuel Lopes Cordeiro was born in Oporto, he is graduated and has a doctorate in Contemporary History from University of Minho, where he teaches at the Department of History, Institute of Social Sciences.

He is Director of the River Ave Valley Textile Industry Museum, situated in Vila Nova de Famalicão, TICCIH Board member (“TICCIH – The International Committee for the Conservation of the Industrial Heritage” is the UNESCO/ICOMOS consultant agency for industrial heritage) and chairman of the Portuguese Society for Industrial Heritage. He is also director of Arqueologia Industrial (Industrial Archaeology), a biannual peerreviewed academic journal on this subject. He has published numerous articles and books in the areas of industrial heritage and industrial archaeology, as well as economic history and contemporary politics. He is director of the Master in Tourism and Cultural Heritage at the University of Minho, and integrates CITCEM – Transdisciplinary Research Center “Culture, Space and Memory”.

José Manuel Lopes Cordeiro, natural do Porto, é licenciado e doutorado em História Contemporânea pela Universidade do Minho, onde exerce funções docentes, sendo Professor Auxiliar do Departamento de História, do Instituto de Ciências Sociais. É director do Museu da Indústria Têxtil da Bacia do Ave, situado em Famalicão, membro do Board do “TICCIH - The International Committee for the Conservation of the Industrial Heritage”, organismo consultor da UNESCO/ICOMOS para o património industrial, e presidente da APPI – Associação Portuguesa para o Património Industrial. É também director da revista Arqueologia Industrial. Tem inúmeros artigos e livros publicados nas áreas do património e arqueologia industrial, assim como da história económica e política contemporânea. É director do Mestrado em Património e Turismo Cultural da Universidade do Minho, e integra o CITCEM – Centro de Investigação Transdisciplinar «Cultura, Espaço e Memória».

Abstract Resumo

On 27 October 1887, with the presence of King Luís I and the Royal Prince D. Carlos, the Tua Railway was inaugurated between Mirandela and Foz-Tua. Engineer Dinis Moreira da Mota was the man driving the locomotive “Trás-os-Montes”, which proceeded to the inauguration. He was in charge of the construction of that narrow gauge railway almost since its inception in October 1884. Born in the Azores Islands, Dinis Moreira da Mota began his professional life on the Portuguese mainland, at the age of only twenty-four, shortly after the conclusion of his civil engineering degree at the Escola do Exército (Army School). By July 1892, when he had returned to the Azores, he had directed the construction of several railway lines in the Algarve, the Trás-os-Montes and the Beiras. Also in the Azores, where he developed his

• 279

Railroads in historical context: construction, costs and consequences

professional life until 1914, date of his death, he had not abandoned his interest in railways, having proposed a rail connection between Ponta Delgada and Ribeira Grande, which although approved, was never built.

In this paper, the author presents Moreira da Mota’s activity as a railway construction manager, with particular emphasis on the railway connection between Mirandela and Foz-Tua.

280 •

Em 27 de Outubro de 1887, com a presença do Rei D. Luís I e do Príncipe Real D. Carlos, era inaugurada a Linha do Tua, entre Mirandela e Foz-Tua. Tripulava a locomotiva “Trás-os-Montes”, que procedia à inauguração, o engenheiro Dinis Moreira da Mota, responsável pela direcção dos trabalhos de construção daquela via ferroviária, praticamente desde o seu início, em Outubro de 1884. Natural dos Açores, Dinis Moreira da Mota iniciou a sua vida profissional no Continente, com apenas vinte e quatro anos, logo após a conclusão do curso de engenharia civil na Escola do Exército. Até Julho de 1892, quando regressou aos Açores, esteve ligado à construção de várias linhas ferroviárias, no Algarve, em Trás-os-Montes e nas Beiras. Também nos Açores, onde desenvolveu a sua actividade profissional até 1914, data da sua morte, não abandonou o seu interesse pelo caminho de ferro, tendo proposto uma ligação ferroviária entre Ponta Delgada e a Ribeira Grande, a qual, apesar de aprovada, não veio a ser construída.

Nesta comunicação o autor apresenta a actividade de Dinis Moreira da Mota na construção de linhas férreas, com particular destaque na ligação entre Mirandela e Foz-Tua.

JosĂŠ Manuel Lopes Cordeiro

The man behind Tua railway: chief engineer Dinis Moreira da Mota JosĂŠ Manuel Lopes Cordeiro

A native of the Azores archipelago (Pico da Pedra, Ribeira Grande, S. Miguel island), where he was born March 2, 1860, engineer Dinis Moreira da Mota is nowadays a personality completely ignored, despite the recognized importance and prestige it has earned throughout his career. This is due, first, because there is still a considerable lack of knowledge of the professional activity that Portuguese engineers developed in late nineteenth and early twentieth century and secondly, due to the fact that in 1892 Dinis Moreira da Mota had departed permanently to the Azores, situation that contributed to the lack of knowledge of the intense professional activity he developed there, due to the relative isolation in which the archipelago was then.

After the frequency of the Faculty of Mathematics of the University of Coimbra between 1877 and 1881, he enrolled at the Military School, nowadays the Military Academy, in Lisbon, to study Civil Engineering, course which he completed in December 1883.

Soon after completing the course in civil engineering, Dinis Moreira da Mota began his professional activity in a Engineering Office in Lisbon, where he remained until April 1884 when, according to his son (MOTTA, 1950: 31), he was hired by the Companhia Nacional de Caminhos de Ferro, then still in formation, to make the technical studies for the railway from Foz Tua to Mirandela, that this company was preparing to build.

On 4 October 1884, Dinis Moreira da Mota was appointed civil engineer

• 281

Railroads in historical context: construction, costs and consequences

Figure 1 - Dinis Moreira da Mota, 26 years old.

282 •

of the Ministry of Public Works, Trade and Industry. The following month, he will have his first professional opportunity to lead the construction of a railway line, to be appointed Head of Section of the South and Southeast Railway (then called the Algarve Railway), in place of António José Arroio, having then fixed his residence at S. Martinho das Amoreiras, in Odemira. The contest for the construction of this railway line, which established the connection from Setúbal to the Algarve, had been opened by a law of March 29, 1883 and by a decree of September 17 of that year.

The conditions that the young engineer faced in Alentejo were extremely unfriendly, which led him to declare that it was a “… completely inhospitable region; … land poor, miserable; the house where I live is the best in the village but

José Manuel Lopes Cordeiro

has no glass in the windows and I see the roof tiles: … the paths are absolutely unusable, except by miracles of equilibrium” (MOTTA, 1950: 33). Isolation was almost total, and when he needed to send mail was necessary to go to Casével, Castro Verde, 38 Km away. As concerned the management of the railway works, he deplored that it was “… hard and meticulous work to supervise a construction in which the rule is practicing deception and fraud” MOTTA, 1950: 33). And the situation does not improved over time. In April 1885 he wreak: “I am degraded, in a complete exile”(MOTTA, 1950: 33). However, these moments of some despair were quickly overcome by the excitement resulting from the challenges faced by him: “… the snow, of a whiteness dazzling; is splendid ! … time is running well, the work will taking some development; the country is very healthy” (MOTTA, 1950: 33).

Dinis Moreira da Mota faced difficulties with the utmost determination. Were constant travels he made, either in Lisbon or Spain, in particular to study narrow gauge railways, as happened in January 1885, when bad weather made it impossible to work. He also accomplished, free of charge, work for the planned Tua Railway, in the expectation that the Companhia Nacional de Caminhos de Ferro be founded in order to begin the construction of that Railway.

On 1 October 1885 was finally officially established the Companhia Nacional de Caminhos de Ferro, which, through a Charter issued on the same date, obtained the concession of the Dão and Tua Railways. In the following month Dinis Moreira da Mota made a contract with that Company to act as head of the construction of the Tua Railway, but he was forced to wait for January 1886, when he obtained the necessary government license to start those functions.

TUA RAILWAY (1886-1887) The first difficulties that Dinis Moreira da Mota experienced in performing its new functions were not those resulting from the railway works. His limited experience did not inspire confidence in some of the investors of the Companhia Nacional de Caminhos de Ferro, particularly since they were known the great difficulties of the realization of the project. However, thanks to the trust of the director of that company, engineer António Xavier de Almeida Pinheiro, who had realized the competence of Dinis Moreira da Mota when he had made the early technical studies for the Tua Railway, still in Lisbon. In fact, the ground was “the most intractable that exists in Portugal, at the bottom of a very tight valley, flanked by abrupt mountains, completely stripped, that looks awful and

• 283

Railroads in historical context: construction, costs and consequences

terrifying; … with blocks and blocks of colossal dimensions, which seems they are about to crash down, towering rocks almost vertically cut … cliffs where the smallest disaster would be fatal. [Furthermore] is rare the day where there is no one death [due to a crime]” (MOTTA, 1950: 34-35).

Figure 2 - The highly scenic Tua Railway was opened in 1887.

284 •

This was consequently, the scenario which was presented to the young engineer, which however, not discouraged him. As concerned the plans to build of railway, he made numerous changes to ensure their viability, and about the other obstacles … he bought a revolver and was willing to face them. These obstacles ranged from complaints of many owners of expropriated lands for the construction of the railway, to the indiscipline of many workers and even the intimations of some criminals, which he had to threaten to shoot. With inexhaustible energy and an extraordinary determination, Dinis Moreira da Mota got out of bed “at daybreak, traveled miles, walked hanging by ropes on cliffs, he ate when it was possible and whether it was possible, returning late, tired” (MOTTA, 1950: 35).

The effort devoted to the management of the works not prevented him from performing other tasks, such as to travel abroad in 1886, in order to choose the future rolling stock. taking advantage to study very different realities of Portugal in

José Manuel Lopes Cordeiro

the railway field, spending two months in France, Belgium and Germany. In the latter, he visited the premises of Maschinenfabrik Esslingen in Stuttgart, where he contracted the purchase of several locomotives. some of them still exist, lying in different Museological Branches of Portuguese Railways (CP). According to the Maschinenfabrik Esslingen Web page, the following were the locomotives that company sold to the Companhia Nacional de Caminhos de Ferro, for service on Tua and Dão Railways: Table 1: Maschinenfabrik Esslingen locomotives sold to Companhia Nacional de Caminhos de Ferro Manufacturing year

Designation

1886

CN “E BRAGANÇA 82”

1887

CN “E 86 FOZ TUA”

1889

CN “E 52 VISEU”

1889

CN “E 53 SANTA COMBA”

1889

CN “E 54 TONDELLA”

1889

CN “E 55 DÃO”

Source: “Erhaltene Lokomotiven Triebwagen der Maschinenfabrik Esslingen”, www.werkbahn.de.

However, consulting the actual collection of the National Railway Museum, we found several locomotives made in Maschinenfabrik Esslingen, but not exactly those which are in the list of that company: Table 2: Maschinenfabrik Esslingen locomotives existing in the CP’s Museological Branches who operated in Tua Railway Location (Museological Branch)

Manufacturing year

Macinhata do Vouga

1886

CN 16

Bragança

1887

N 1 (E 81 ?)

Bragança

1889

E 55

Current designation

Source: “National Railway Museum”, www.cp.pt.

Finally, on 27 October 1887, with the presence of King Luís I and Prince

• 285

Railroads in historical context: construction, costs and consequences

286 •

Carlos, the Railway from Foz Tua to Mirandela was inaugurated. The work of the young engineer did not go unnoticed. The press at the epoch considered Dinis Moreira da Mota one “of the engineers which has greater future in our country, by outstanding skills of real builder, that everyone will recognize” (DIÁRIO DE NOTÍCIAS). For services rendered, was proposed to him a medal, he refused outright. The experience in the construction of Tua Railway, between Foz Tua and Mirandela – considered the most difficult engineering work until then done in Portugal – was in fact, an auspicious milestone in his career as an engineer, that will open doors to continue his connection to the Companhia Nacional de Caminhos de Ferro.

Also in this year of 1887 he was appointed head of management of the Tua Railway, a function that he accumulated with head of construction of the branch from Santa Comba Dão to Viseu in the Dão Railway, concessioned to the same Company. In 1889, he traveled again to Germany, on behalf of the Companhia Nacional, to acquire locomotives for the Dão Railway. He took the opportunity to visit the Universal Exhibition in Paris, which was held from March 6th to October 31st, in which the Eiffel Tower was inaugurated, built especially for this event, celebrating the centenary of the French Revolution of 1789.

In the contacts made, Dinis Moreira da Mota proved an exceptional technical expertise, leaving a positive image in the Maschinenfabrik Esslingen. which made him the most high praise in a letter sent to the Companhia Nacional, which asked “to that Company [that] dispensed to send its engineer and their assistants to Portugal, to assemble the locomotives and conduct the respective experiments, because the Portuguese engineer inspired so much confidence and would delegate to him the representation to make the experiments stipulated in the contract documents. The Companhia Nacional has agreed, the unassembled machines came to Santa Comba, and the assembling and experiments were exclusively directed by Moreira da Mota” (MOTTA, 1950: 38-39).

DÃO RAILWAY (1889-1891) Dinis Moreira da Mota across this period a phase of intense professional activity because, simultaneously to their tasks in Tua and Dão Railways, in 1889 he accepted the contract construction of all stations and halts buildings, and also the task of the construction of all the Beira Baixa railway, in the section between Abrantes and Covilhã – a total of 165.2 km –, which was completed on September 6, 1891. He settled down, then, in Alpedrinha (Fundão), south of the Gardun-

José Manuel Lopes Cordeiro

ha Mountain, to monitor the construction work. In addition to extension, the difficulties of construction of the track recorded other difficulties, particularly those deriving of having to accomplish over about 50 bridges –some of them quite extensive and high, such as Peral Bridge (130,7 m), São Pedro Bridge (160,7 m), Cerejal Bridge (209 m), Maçainhas Bridge (130,7 m) or the Gogos Bridge (130,7 m) –, and ten tunnels, some too lengthy, as Gardunha tunnel (646 m) or Sabugal tunnel (398 m).

After completing this section of the Beira Baixa Railway, Dinis Moreira da Mota returned to the service of the Ministry in Lisbon, where took up residence, having considered the possibility of competing for a position as professor of engineering at the Military School, an objective which he came to abandon.

RETURN TO THE AZORES With the works of the Beira Baixa Railway completed, he returned to work at the Ministry in Lisbon – which do not fit with his entrepreneurial spirit –, and without any other perspectives, Dinis Moreira da Mota decided to return to S. Miguel island after fifteen years of absence. The company that built the Beira Baixa Railway had failed – being due a considerable amount of money – and the Companhia Real found itself in a period of great financial difficulties, a result of mismanagement and of the crisis that the country was facing then. Personally, he also found himself facing a period of severe discomfort, whether by the political and economic situation of the country – British ultimatum, republican revolt in Oporto, financial crisis – whether by family troubles that occurred successively: in January 1892, a son died. in the following month, the eldest brother, and in April, his father. He had however, married Mary Margaret Botelho Riley, but due to the intense involvement in work, his family life was found to be very destabilized.

The solution was to return to his homeland, from where he departed in July 1892.

Returned to S. Miguel island, Dinis Moreira da Mota will divide, initially, his profession of engineer with agricultural activity, and also exercising political activity for a short period of time. As an engineer, he began to act as inspector of the 5th Industrial Circumscription, in Ponta Delgada, in which he remained between April 11, 1892 and January 10, 1894, then moving the availability for a short period of time. However, in the first years after his return to the Azores, it was agriculture that Dinis Moreira da Mota paid particular dedication to, not only because it was one of his passions, but also to compensate for the intense professional activity that he had developed on the Continent.

• 287

Railroads in historical context: construction, costs and consequences

288 •

As a farmer, he set up a small agricultural unit in Termo de Lagoa, where he devoted himself to the cultivation of land, animal husbandry, planting vineyards, among other activities. He sought also to provide expertise on agriculture, contacting with other more experienced farmers and constituting a considerable agricultural library, which included numerous books ordered from abroad. However, the initiative of major significance in this context developed by Dinis Moreira da Mota was the foundation in 1893, for his initiative and with the support of the district agronomist and large landowner António de Andrade Albuquerque Bettencourt, and other landowners on the island, the first agricultural union in the country – the Agricultural Union of Lagoa.

Probably, as a result of their readings, he had been aware of the outbreak of agricultural unions, who a few years earlier there had been in France, which also served as inspiration to the Portuguese legislators, which, in the following year, sought to adapt them to our country through the Decree of July 5, 1894. This, regulated precisely, the creation of “Agricultural Unions and Portuguese Chambers of Commerce” (decree of Carlos Lobo de Ávila, Minister of Public Works of the government of Hintze Ribeiro).

Indeed, Dinis Moreira da Mota, himself, who that year was deputy to the Parliament, presented on 6 and 7 July 1893, together with Antonio Barjona Alfredo de Freitas – founder of the first agricultural union of the Continent, in Montemoro-Velho, in 1894 – a draft law in order to establish in Portugal the advantages that the French law of 1884 provided to farmers who wanted to associate in agricultural unions, and that in that country, ten years after the enactment of that law, were already in the number of 900 unions (DIÁRIO DA CÂMARA DOS SENHORES DEPUTADOS, Nº 67, 1893: 6).

The draft law, which was supported by Bernardino Machado, then Minister of Public Works, Trade and Industry, and Mariano de Carvalho, president of the legislative committee that had to rule on the matter, would have been approved, but ended up not even be voted because of the dissolution of Parliament.

Contrary to its name, the agricultural unions that emerged at that time did not correspond to what today is meant by union organizations. They assumed the character of local associations, constituted by farmers land owners, which can also join individuals not being farmers but exercising professions related to agriculture. The agricultural unions were intended to study, promote and protect local agricultural interests, as well as his associates, to which did not distribute dividends. They were rather, associations of landowners.

José Manuel Lopes Cordeiro

DINIS MOREIRA DA MOTA AS DEPUTY A short passage of Dinis Moreira da Mota for political activity – largely as a form of solidarity with his brother, Aristides Moreira da Mota, paladin of the Azorean autonomy –, serving as Member of Parliament by the circle of Ponta Delgada, coincided with a period of great political effervescence in the Azores, due to the issue of autonomy.

Elected on 23 October 1892 by the Regenerador Party, it fell to him to present in Parliament, along with members Mariano Faria e Maya and Francisco de Almeida e Brito – both elected by the Progressista Party – the draft law on the autonomy of the Azores, prepared by an “Autonomic Commission of the District of Ponta Delgada”, where his brother Aristides played a leading role. However, the project was not approved because Parliament was dissolved on July 15, 1893.

Despite its non-approval – according to a deputy it had been “sentenced to the voracity of the moths’ – that project will give rise to the Decree of 2 March 1895, which inaugurated the first period of the Azorean autonomy. In Parliament, Dinis Moreira da Mota also integrated the committees of Public Works (1893) and Primary and Secondary Education (1893).

Apart from this project on autonomy, Dinis Moreira da Mota developed intense activity as deputy – considering the short duration of the legislature –, presenting projects on the issue of the alcohol industry and its tax regime, on the shipping routes between Lisbon and the archipelagos of Madeira and the Azores, and on claims of his voters, such as the registrars and notaries of the court of law of the district of Ponta Delgada, and also against the proposed law on industrial tax (DIÁRIO DA CÂMARA DOS SENHORES DEPUTADOS, Nº 62, 1893: 3).

However, a proposal that will take him a great pleasure, given the links to his professional education, was the establishment of the submarine telegraph cable to the Azores, that due to its geographical location, allowed the expansion of transatlantic submarine communications with North America and various parts of Europe. On the initiative of Europe and Azores Telegraph Company, Limited, a subsidiary of Eastern Telegraph Company Limited, the submarine telegraph cable was inaugurated on August 27, 1893, during the passage of Dinis Moreira Mota through Parliament..

BACK TO ENGINEER’S ACTIVITY After the parliamentary experience, Dinis Moreira da Mota returned to Ponta Delgada where, on March 16, 1894 assumed the functions of engineer attaché to

• 289

Railroads in historical context: construction, costs and consequences

the Directorate of Public Works of the District of Ponta Delgada – within which he developed numerous private and public works projects – having been appointed, in May 3, 1897 assistant engineer of the director of the works of the Ponta Delgada’ Artificial Harbour, and, in August 1, 1898, interim director of that body, then already in the dependence of the General Board created by the autonomist decree of March 2, 1895.

Figure 3 - Works of the artificial harbour of Ponta Delgada.

290 •

The first great work of Denis Moreira da Mota as an engineer after his return to the Azores, was the resolution of the problem of construction of the artificial harbour of Ponta Delgada – a work that had dragged on for decades – for the realization of which he was appointed on May 3, 1897, assistant engineer of the director.

The need for a harbour in Ponta Delgada was felt for a long time. São Miguel island had no natural harbours that could be used for the needs of maritime traffic, which caused a major constraint in the development of the necessary trade and, also, in the circulation of people.

In the first half of the nineteenth century, with the outbreak of the orange trade for exportation – mainly to Britain and the Baltic countries (DIAS, 1996) – have

José Manuel Lopes Cordeiro

worsened the needs of a harbour that could correspond to the economic development that this raised. Arise, then, several proposals for the construction of an artificial harbour – commonly referred to as dock – which, however, any results obtained, either by funding difficulties either by lack of knowledge and the technical means to achieve (GÓIS, 1838 and RIBEIRO, 1844).

To resolve this embarrassment, the wealthy american merchant Thomas Hickling, established since 1769 in Ponta Delgada and a major driver of the orange trade to Britain, hired in 1838 the famous engineer of harbours John Rennie for the preparation of construction project of the artificial harbour of Ponta Delgada.

Rennie sent his adjunct John Tucker Scott to gather evidence to enable him to write a memoir about the possibility of building a harbour.

John Rennie proposed the construction of a basin, sheltered by two breakwaters, both east of the Customs of Ponta Delgada (RENNIE, 1838 and RENNIE, 1839). However, the proposal did not have any development and it was necessary to wait fifteen years to Tucker be called to São Miguel to prepare a new draft.

The solution now proposed was the construction, towards the south, of two large breakwaters, one starting from the São Braz castle, and the other from the São Pedro Fortress , both changing direction at the same level: the first at a right angle to the east, and the second going to meet him to the West. Between the ends of the two breakwaters was an opening serving as an entry for a basin, in rectangular shape.

The draft presented by John Scott Tucker was approved, and a committee to implement it was appointed. However, a member of the committee for the construction of the artificial harbour of Ponta Delgada, chaired by José do Canto and Duarte Borges da Câmara e Medeiros (Viscount of Vila da Praia), considered it necessary to discuss the project with Rennie, having for this purpose, traveled to London. As, at that time Tucker was occupied at the Cape of Good Hope, the committee invited John Rennie, not only for directing the work of construction of the harbour, but also to promote the raising of a loan in London in order to cope with their cost.

Initially, all seemed well underway but, shortly after, Rennie said only accept technical direction, and the Portuguese State should assume the construction costs of the port of Ponta Delgada. As the Portuguese Government did not accept this charge, the efforts of the Viscount of Vila da Praia, and of the people os São Miguel represented by him, were frustrated. Only on August 9, 1860, the Parliament passed a law providing the funds needed to build the harbour, which would be recovered by levying customs duties on São Miguel and Santa Maria (BENSAÚDE, 1936).

• 291

Railroads in historical context: construction, costs and consequences

BEGINNING OF THE HARBOUR CONSTRUCTION The following year, on September 30, was formally launched the first stone, but only on 28 October 1862 the works really began with the implementation of a technical solution which provides for the establishment of the foundations of the piers through the accumulation of a number of large stones – extracted from the quarries of Santa Clara and conducted to the work zone through a small extrawide (2140 mm) gauge railroad where operated a steam locomotive and 39 wagons – the so-called “rockfills”. These would be “elevated to about 2 meters above the highest tides and [had] at lower tides, the width of some 40 meters. On these ‘rockfills’ there would be no fence or wall, or any wharf. The sheltered space would be about 16 hectares, considered sufficient for the 200 sail ships, with about 300 to 400 tonnes each one, which in that epoch visited us, especially for the export of orange” (MOTTA, 1950: 52).

Figure 4 - Santa Clara quarries where the stone was extracted for the construction of the breakwater.

292 •

The initial project was however, modified by John Rennie, which substantially reduced the breakwater of São Pedro (which was never built), keeping the other in front of the city of Ponta Delgada, but oriented towards East-North-East,

José Manuel Lopes Cordeiro

at the end. He also designed – to minimize the extremely rudimentary solution represented by the piers – the building of a “wall on the ‘rockfill’, with large stones and local clay mortar (or pozulana), about 4 meters wide by 5 high, and a wharf, 14 meters wide” ( MOTTA, 1950: 52).

The construction of the artificial harbour by the British engineers revealed that they were unaware of the exceptional violence of the waves in front of Ponta Delgada, as well as of the geological characteristics of the quarries of Santa Clara. This further increased the difficulties of the work, especially since they also lacked the necessary scientific and technical knowledge which are able to tackle the obstacles that stemmed from there.

In England, at that time, it was usual to use loose stone in this type of construction, obtained by blasting powder in quarries, which method was also applied in basalt quarries of Santa Clara. British engineers were unaware that the basaltic lavas were fragmented by grooves, resulting from the spontaneous contraction of the rock during its cooling (BENSAÚDE, 1936). They also ignored that in the basalt, the detonations produced radial cracks around the holes where the gunpowder was placed, reducing the rock into small fragments unsuitable for the construction of the piers in the Azores sea, whose violence “moves blocks of more than 2-3 tonnes to 20 meters below the surface of water” (MOTA, 1900). Therefore, it would be necessary to deposit the stones from the seaside, so as to create from the beginning, a stabilization rampart so that the stones resisted the erosive action of waves.

The construction of the breakwater was far from advancing as fast as John Rennie had anticipated. The deadline of 4-5 years for its conclusion proved to be unrealistic. The demolitions caused by bad weather during the winter forced that the repairs would be made next summer, which not only delayed the work and caused unpredictable increase in costs. When British engineers were aware of the failure of its processes and predictions, they tried to assign it, unsuccessfully, to the action of the Administrative Board. The only solution was to abandon the work which indeed came to be achieved at the September 2, 1866.

After the abandonment of the work by British engineers, the Administrative Board hired from 1866 to the spring of 1872, the Portuguese engineers Júlio Ricardo Ferraz (between 1866 and 1870) and Mariano A. Machado de Faria e Maia (from 1870 to 1872), which have continued that work. Although they had respected Rennie’s general plan, they introduced several modifications in order to reinforce the solidity of the foundations of the piers, in particular, in the inclinations of the ramparts or lateral ramps of the ‘rockfills’.

In 1872, the plan of works has been modified by engineers Manuel Afonso de Espregueira and Álvaro Köpke de Barbosa Ayala, “becoming the crowning wall 5 meters wide and 10 tall. Were directors, from 1872 to 1881, engineers

• 293

Railroads in historical context: construction, costs and consequences

Álvaro Köpke and David Xavier Cohen. The latter changed again the project, the achievement of which were commissioned in 1888, the French contractors Combemale & Michelon, who worked diligently noteworthy” ( MOTTA, 1950: 53) and concluded the contract in 1894

THE 1894’S CATASTROPHE AND THE ACTION OF MOREIRA DA MOTA

294 •

All the effort made for nearly four decades to build the artificial harbour of Ponta Delgada was suddenly almost completely annulled by successive storms that demolished the constructions almost completely. From 7 to 8 December 1894, a violent storm “destroyed 200 meters of the wall shelter, snatched everything with him and was moved the ‘rockfill’ up to 20 feet deep! ... It was a real catastrophe: more than 300 000 tons of stone thrown into the harbour, the 20 feet tall lighthouse was shattered, the huge crane ‘titan’ was thrown overboard, 25-ton blocks were dragged, and was sunk a floating dock !” ( MOTTA, 1950: 53-54).

To solve the serious situation of the construction of the artificial harbour of Ponta Delgada – already considered the most complex harbour work ever done in Portugal – the Administrative Board appointed in May 3, 1897, the engineer Dinis Moreira da Mota, as adjunct of the new director, engineer José Maria Cordeiro de Sousa. The new director organized a new project for further work, according to which “should protect the wall shelter from under the wall from the outside, with a row of blocks of cement concrete of at least 25 tons each one and 20 meters wide; the pier would also be clothed with this concrete; the wall shelter however, will continue to be built with mortar pozulana, to which add some cement, little” ( MOTTA, 1950: 54).

However, Dinis Moreira da Mota – that on September 30, 1898 had been appointed interim Director of Public Works of the district of Ponta Delgada – made a new design change, abandoning entirely the “mortar pozulana, which would be replaced by concrete slow-setting cement; he proposed to enlarge the wall shelter for about 10 to 13 meters (according to the locals) and left the pier (except in some points – curve and end) vaulted warehouses of about 5 to 6 meters deep by 6 of the mouth, interspersed 3.5 meters; and also proposed that the wall shelter finished with a circular head of 20 m radius based – up to 15 meters below sea level – in solid blocks of concrete, of 35 tons. Also fixed the width and shape of the wall shelter, in particular curve, as well as the inclination of the ramparts, the shape of the breakwater, etc” ( MOTTA, 1950: 54-55).

José Manuel Lopes Cordeiro

It is due, therefore, to Dinis Moreira da Mota, the technical solution for the construction the piers of the artificial harbour of Ponta Delgada – in fact, the first harbour – which until then had not been possible to built, despite the distinguished engineers that preceded him. His project was based on studies he done on the local conditions of sea and wind, on the observations of the effects of successive experiments performed by him, on the use of another type of material, and on the design of a different shape for the configuration the piers.

Only with the solutions pointed by Denis Moreira da Mota was it possible to stanch the destructive effects that until then the violence of the sea and the specific weather conditions of the Azores caused in the works that were being made.

Despite some reluctance that his solution was received, because withdrew the standards previously used – earned him the support engineer Adolfo Loureiro –, the project of Denis Moreira da Mota came to be adopted on 27 March in 1901.

The construction work continued thereafter, at an accelerated pace. As piers were taking configuration, the solution of Dinis Moreira da Mota appeared as the most suitable, having resisting storms even greater than in 1894 and its strengthening pursued continuously, having been thrown into the sea, until the date of death of Moreira Mota in 1914, about 5,000,000 tons of stone. However, will still have to wait long years for the full completion of the construction of the harbour, which only came about in 1942.

In 1910, the budgeted funds were exhausted. The amounts which since then have been available gave only to the maintenance of what had already been built. Given these difficulties, Dinis Moreira da Mota was charged with directing, cumulatively, the Public Works of the State and those who were in charge of the General Board, engaging in the construction and repair of roads. However in 1911, he had passed from adjunct director to director of Public Works of the Autonomous District of Ponta Delgada, a position he would maintain until his death.

THE SÃO MIGUEL RAILWAY The most important project that marked the final period of the life of Dinis Moreira da Mota was the construction of a railway on São Miguel , whose description and justification was presented by him to the General Board in November 1897 (MOTA, 1897a, and MOTA, 1897b). The project envisaged the establishment of a rail link between Ponta Delgada and four counties in the district: Ribeira Grande, Povoação, Vila Franca do Campo and Lagoa, as well as from the harbour

• 295

Railroads in historical context: construction, costs and consequences

to the dock shelter, in order to establish the interface with the long course navigation which apportioned regularly there.

To the achievement of the project was created a Promotion Commission, comprising various personalities of the island, including the engineer Dinis Moreira da Mota. The Commission has conducted several studies of economic feasibility, on the demographic structure of the island, on transporting of goods and passengers, as well as comparative studies with other similar realities – such as the Châteauneuf-sur-Charente to Barbezieux railway, in southwestern France, opened in 1872, or the Alençon Condé-sur-Huisne railway in Lower Normandy, also in France and inaugurated in 1873. The Commission concluded that the project had full viability, because expected results pointed to an annual revenue, at worst, close to 126 contos (REVISTA MICHAELENSE, 1919: 296). The Commission also presented an analysis of the construction cost of the railway lines, the acquisition of rolling and fixed stock, the outbuildings to the track, as well as land.

Figure 5 - Revista Michaelense, nº 3, 1919.

296 •

José Manuel Lopes Cordeiro

In fact, there were two railway lines that would be built, both going out from Ponta Delgada: one, called the south line, along the south coast to Povoação (via Lagoa, Vila Franca do Campo and Furnas), which presented no major construction difficulties as it follow the coastline, almost entirely, with the exception of the connection to Furnas; the other, towards Ribeira Grande, called the northern line, had had considerable difficulties in construction, due to having to cross the mountainous region of the island.

For the layout of the railroad, there were only two solutions, using two narrow passages between the mountains: one at the Pico das Murtas and the other in the Chã do Rego de Água. Dinis Moreira da Mota chose Pico das Murtas due to meet halfway between Ponta Delgada and Ribeira Grande. The proposed site for the Central Station of Ponta Delgada would be the Calhau do Laguim, given its proximity to the Customhouse. The track gauge would be of 1 meter, “because it is one that requires less maintenance care for transport heavy goods with more stability” (REVISTA MICHAELENSE, 1919: 301). The total length of lines to be built were 64.402 km, including the direct line to Povoação with 47.980 km, and the line to Ribeira Grande with 16.422 km.

On July 26, 1899, the Government issued a decree authorizing the General Board to undertake a tender for the construction of the railway – although, in the south line, it stops in Furnas, not reaching Povoação, an option that had been already abandoned by the Commission appointed by the General Board – and realize its tendering. The project was previously approved in the House of Peers, at its meeting on 17th July of that year, provoking demonstrations of joy by the population of São Miguel during three days, and determining the declaration of holiday on the 19th July by the secretariat of the General Board of the Civil Government. However, the difficulty to establish a company to undertake the construction of lines, mainly for lack of funds, meant that the project was never realized.

THE LEGACY OF DINIS MOREIRA DA MOTA While held the Public Works Department, Dinis Moreira da Mota developed a multifaceted activity as engineer, who has indelibly marked the physiognomy of the São Miguel island. Within these functions, he repaired all the small fishing ports in the district of Ponta Delgada and was also responsible for directing the work of enlarging and protecting the docks of Porto de Pipas, in the bay of Angra do Heroísmo. He also designed and built the premises of the Magnetic and Seis-

• 297

Railroads in historical context: construction, costs and consequences

mological Observatory of Fajã de Cima, São Miguel, in 1911, which included the first seismograph in the archipelago, as well as the new body of the former building of the Ponta Delgada Public Library, where currently is installed on the Regional Conservatory of Ponta Delgada.

Figure 6 - Dinis Moreira da Mota, 52 years old.

298 •

A remarkable aspect of the professional activity of Dinis Moreira da Mota was the fact that their projects also contemplated urban and landscape concerns, which at that time were virtually nonexistent. It was on their own initiative and as a means of recovery of the landscape adulterated by the extraction of stone for the harbour works, which were planted several species of trees in Pedreiras da Doca, Ponta Delgada, giving rise to a park that became known as the “Park Dinis da Mota”, which however, came to be destroyed by extending the runway of the airport of Ponta Delgada. He also designed the planting of pines in the sidebands of the road along the beach of S. Roque at the time of its construction, as well as

José Manuel Lopes Cordeiro

the construction of vacation homes in the zone of Pópulo and Milícias beaches, which still today constitute an urban complex of remarkable quality.

Another important facet of the life of Dinis Moreira da Mota was its commitment to social causes and community activities. On January 1, 1901 was one of the founders, in Ponta Delgada, of the Charity Association for the Promotion of Education “The Twentieth Century” – the name results from the date of its foundation – which housed and educated children of both sexes. He integrated, also the organizing committees of the District Exhibitions of arts and industries held in Ponta Delgada in 1895 and 1901 and, in the latter, was responsible for the installation of an ephemeral railroad in Ponta Delgada, to which he directed the construction of a locomotive to circulate during the Exhibition period. He was also, twice, in 1901 and from 1903 to 1912, provisory teacher of sciences at the Ponta Delgada Lyceum.

In 1914, Dinis Moreira da Mota was appointed Mayor of Ponta Delgada. On the morning of August 29th of this year, when, in the exercise of those functions, he was inspecting the construction of a bridge in Santo António da Bretanha, on the road between the Capelas and the Bretanhas, died suddenly struck by an angina pectoris.

Despite the important work that marked his work as an engineer in his native Azores, Dinis Moreira da Mota also deserves to be remembered by that who, without doubt, one of its greatest accomplishments: the construction of the Tua Railway.

• 299

Railroads in historical context: construction, costs and consequences

BIBLIOGRAPHY BENSAÚDE, Alfredo (1936), Vida de José Bensaúde. Porto: Litografia Nacional.

DIAS, Fátima Sequeira (1996), Uma estratégia de sucesso numa economia periférica: a casa Bensaúde e os Açores 1800-1870. Ponta Delgada: Jornal da Cultura.

GÓIS, Henrique José de Medeiros Cogumbreiro de (1838), Projecto de melhoramento para a Ilha de S. Miguel e plano para se fazer uma doca na enseada em Ponta Delgada, por um micaelense. Lisboa: Typographia de José Baptista Morando.

MOTTA, António Augusto Riley da (1950), Diniz Moreira da Mota (engenheiro). Ponta Delgada: [s.n.].

300 •

MOTTA, Diniz Moreira da (1897a), Caminho de Ferro em San Miguel. Relatório da Comissão Promotora da construção do caminho de ferro, memória descritiva e orçamento. Ponta Delgada: Typ. Elzeviriana.

MOTTA, Diniz Moreira da (1897b), Caminho de Ferro em San Miguel. Parecer da Comissão nomeada pela Junta Geral sobre o ante-projecto. Ponta Delgada: Typ. Elzeviriana.

RIBEIRO, Manoel José (1844), Projecto para a construção duma doca na ilha de S. Miguel.

Ponta Delgada: Typ. de F. J. P. de Macedo.

PERIODICALS Diário da Câmara dos Senhores Deputados, Lisboa, nº 62 and nº 67, 1893.

Diário de Notícias, Lisboa, Outubro de 1887.

Revista Michaelense, Ponta Delgada, nº 3, 1919.

Merrit Roe Smith

BECOMING ENGINEERS IN EARLY INDUSTRIAL AMERICA OS ENGENHEIROS NOS PRIMÓRDIOS DA AMÉRICA INDUSTRIAL M. Roe Smith (MIT, USA EUA) Merritt Roe Smith is Metcalfe Professor of engineering and Liberal Arts in the Program in Science, Technology, and Society at the Massachussets Institute of Technology.

Roe Smith é Metcalf Professor de Engenharia e Artes Liberais, MIT Massachusetts Institute of Technology, EUA. É autor de um dos livros mais conhecidos sobre os inícios da industrialização nos EUA.

Abstract Resumo

In 1859, a New Englander named Daniel Lester Harris traveled to Russia to examine and report on the condition of the bridges on the Moscow and St. Petersburg Railroad. Interestingly, he inspected a rail system that had been constructed largely under the supervision of two Americans, George Washington Whistler and Joseph Harrison, Neither man was in Russia at the time of Harris’ visit. But behind them remained a legacy of construction methods, structures, and machinery that revealed a distinctly American style of building. The careers of all three men go far to illustrate the diverse paths by which individuals became engineers during the early 19th century.

This paper explores that transformation by scrutinizing the lives of Whistler, Harrison, and Harris and by comparing their experiences with some of their contemporaries.

How did they come to call themselves engineers in a pre-professional era and what did this change signify? How did they differ from the traditional millwright and mechanic? What was their relationship to the business community and the emerging industrial system of America? To broach these questions is to suggest that the engineer was one of the key social inventions of the Industrial Revolution.

Em 1859, Daniel Harris, da Nova Inglaterra, viajou para a Rússia para estudar e reportar sobre as pontes de Moscovo e de São Petersburgo. Na altura inspecionou uma linha de caminho de ferro construída sob a supervisão de dois americanos, George Whistler e Joseph Harrison, que entretanto já não estavam na Rússia, mas tinham aí deixado um legado de métodos de construção, estruturas, e maquinaria

• 301

Railroads in historical context: construction, costs and consequences

com um estilo distintivamente americano. As carreiras destes três homens ilustram as diferentes vias pelas quais se chegava a engenheiro nos princípios do século XIX. Este trabalho explora essas transformações através das vidas de Whistler, Harrison e Harris, comparando as suas experiencias com a de outros contemporâneos.

Como é que chegaram a engenheiros e quem eram esses homens? Em que é que eram diferentes dos mecânicos tradicionais, e qual era a sua relação com a comunidade empresarial, e com o emergente sistema industrial americano? As respostas a estas questões acabam por sugerir que o engenheiro foi uma das invenções sociais mais importantes da revolução industrial.

302 •

Merrit Roe Smith

Becoming engineers in early industrial America Merritt Roe Smith

In 1859, a New Englander named Daniel Lester Harris traveled to Russia to examine and report on the con­dition of the bridges on the Moscow and St. Petersburg Railroad. Interestingly, he inspected a rail system that had been constructed largely under the supervision of two Americans, George Washington Whistler and Joseph Harrison.1 Neither man was in Russia at the time of Harris’ visit. Whistler had died in 1849, and Harrison had already returned to his native Philadel­phia. But behind them remained a legacy of construc­tion methods, structures, and machinery that revealed a distinctly American style of building.

What is more, the careers of all three men go far to illustrate the diverse paths by which individuals became engineers during the early 19th century. This paper explores that transformation by scrutinizing the lives of Whistler, Harrison, and Harris and by comparing their experi­ences with some of their contemporaries.

Like Creve-coeur, we might ask who were these “new men”? How did they come to call themselves engineers in a pre-professional era and what did this change signify? How did they differ from the traditional millwright and mechanic? What was their relationship to the business community and the emerging industrial sys­tem of America? Where did their essential loyalties and commitments lie? To broach these questions is to suggest that the engineer was one of the key social inventions of the Industrial Revolution.

1 For a fascinating account of Harris’ trip to Russia (particularly his impressions of European society at the time), see the D. L. Harris Diary, Harris Papers, Connecticut Valley Historical Museum, Springfield, MA [hereafter abbreviated CVHM]. Also see Harris to Messrs. Winans, Harrison & Winans, May 28, 1859, Ibid.

• 303

Railroads in historical context: construction, costs and consequences

GEORGE W. WHISTLER (1800-1849): WEST POINTER AS ENGINEER

304 •

Of the three, Whistler achieved the greatest renown during his lifetime. The son of an Irishman who had served under Burgoyne during the Revolutionary War and subsequently returned to America to join the U. S. army, Whistler spent his youth at various military posts in the Old Northwest Territory and upper Mis­sissippi Valley. He was a handsome lad of medium height and fine features who exhibited a particular aptitude for drawing and music. Indeed, friends often remarked that he could have just as well become an artist as an engineer. But that was not to be. Military life counted for much in the Whistler family. In addition to his father (an army major), two older brothers had made careers for themselves in the mili­tary and it was clear that George Whistler would do the same. On July 31,1814, at a mere fourteen years of age, he entered the United States Military Academy as a cadet.2 Whistler spent five years at West Point. Although his papers reveal little about the experience, it is clear that his school years proved critical in setting a course for his personal as well as professional life. At that time West Point was the only institution of formal engi­neering education in America. To be sure, it was first and foremost a military school. But under the rigorous discipline of its new superintendent, Major Sylvanus Thayer, the curriculum stressed mathematics and sci­ence as much as tactics and command. Like the French Ecole Polytechnique on which it was modeled, West Point turned out not just officers of the line but “soldier technologists” who could master the level as well as the sword. Everyone understood that knowl­edge of earthworks and fortifications had immediate applications in the construction of such “civil works” as roads and canals. Not surprisingly, many a cadet resigned his commission upon graduation (or soon after) to accept positions in the private business sector. Others like Whistler (who ranked 10th in his class and decided to stay in the Army), received assignments in the elite army bureaus of engineering and ordnance.

Only the poorest students became regular line officers in the antebellum period.3 2 KJeorge L. Vose, A Sketch of the Life and Works of George W. Whistler, Civil Engineer (Boston: Lee and Shepard Publishers, 1887), pp. 10-12. An excellent biography of Whistler during his later years is Albert Parry, Whistler’s Father (Indianapolis: The Bobbs-Merrill Co., 1939).

3 Vose, “Whistler,” pp. 12-15. For revealing treatments of the West Point educational system during the early nineteenth century, see Stephen E. Ambrose, Duty, Honor, Country: A History of West Point (Baltimore: Johns Hopkins Press, 1966), esp. chs. 4 and 5; Edward M. Coffman, The Old Army: A Portrait of the American Army in Peacetime, 1784-1898 (New York: Oxford University Press, 1986); Peter M. Molloy, “Technical Education and the Young Republic: “West Point as America’s Ecole Polytechnique, 1802-1833” (Ph.D. diss., Brown University, 1975); James L. Morrison, “The Best School in the World”: West Point, The Pre-Civil War Years, 1833-1866 (Kent, OH: Kent State University Press, 1986); and especially Keith Hoskins and Richard H. Macve, “The Genesis of Accountability: The West Point Connections,” Accounting. Organizations and Society 13 (1988): 37-73. An excellent dissertation on one of Whistler’s West Point contemporaries is Stanley Falk, “Soldier-Technologist: Major Alfred Mordecai and the Beginnings of Science in the United States Army” (Ph.D. diss., Georgetown University, 1959).

Merrit Roe Smith

For Whistler the West Point years were important in another respect. There he made life-long friendships and acquaintances that would prove critical to his professional advancement. Foremost among them was William H. Swift (18001879), a classmate whose older brother, Colonel Joseph Gardner Swift, became one of his closest friends and whose sister, Mary, he wed in 1821. Colonel Swift naturally took an interest in Whistler. Since he served as the Army’s chief engineer (1812-1818) and later became a highly respected civil engineer, he was in a position to provide important contacts for his younger brother-in-law. Ties of kinship and friendship proved very important in the 19th century business world. Clearly who one knew mattered as much as what one knew, and in Whistler’s case there were many such relationships. One of the closest was WOliam Gibbs McNeill (1801 -1853), a West Point classmate and early professional associate whose sister, Anna, would marry the wid­owed Whistler in 1831. Virtually every job he held in later years had connections that could be traced back to his West Point experience.4 Upon graduating from the academy in July 1819, Whistler was commissioned a second lieutenant in the Corps of Artillery and reported for duty at a garrison near the port of New York. It was a relatively uneventful assignment that included some topogra­phical surveying and a brief stint as an instructor of drawing and geometry at West Point. But a significant change came in the spring of 1822, when he received orders to report to a topographical unit charged with tracing the international boundary between Lake Superior and the Lake of the Woods in what is now northern Minnesota. It proved to be a difficult and trying experience that lasted nearly six years (1822-28). Yet in later life Whistler enjoyed regaling his family with tales of the Great Lakes region, its primitive frontier conditions, and how he and his partners traveled by snowshoe, slept in snow banks wrapped in buffalo robes, and ate various concoctions of tallow and hominy warmed over open fires. Having been raised on one frontier and served in another, he often looked back to these experiences with a certain wistfulness and as a source of inner strength during times of adversity. His intense pride in being an American seemed to be closely associated with his frontier experiences.5 Upon returning from the West in 1828, Whistler’s next assignment established a pattern he would follow for the rest of his life. The project involved a radically new technology - the steam-powered railroad.

Under the General Survey Act of 1824, Congress made available the engineering resources of the United States Army to private companies engaged in the construction of public works. Such enterprises included roads and canals, but the most exciting of all were railroads. By 1827 a major company, the Baltimore and Ohio Railroad, stood poised and ready to seek federal assistance.

4 Vose, Whistler, pp. 20-21; Parry, Whistler’s Father, pp. 23, 27, 50-52, 64,127-28.

5 Vose, Whistler, pp. 14-15; Parry, Whistler’s Father, p. 122.

• 305

Railroads in historical context: construction, costs and consequences

306 •

Encouraged by his board of direc­tors, the B & O’s president approached the Secretary of War about securing the services of a half dozen or so Army engineers.

Moreover, he specifically requested Whistler. With the secretary’s approval, the young of­ficer, accompanied by Lt. Colonel Stephen H. Long and his friend Captain William Gibbs McNeil], re­ported to the corporation’s Baltimore offices in Octo­ber 1828. At this time the railroad, without any apparent explanation, designated Whistler and his colleagues as “engineers.”6 Working on the B & O gave Whistler the opportunity to exercise and extend the engineering knowledge he had acquired at West Point as well as on the various topographical surveys. At the very outset of the project, he accompanied McNeill and Jonathan Knight (a civilian employed by the Army Corps of Engineers and detached to the B & O) on a six-month tour of Great Britain to examine the railroads of that country. As later events would reveal, this proved to be a pivotal experience that, in effect, introduced Whistler to railroading. Not only did he have the opportunity to meet such “distinguished persons” as Thomas Telford, the Stephensons, and “Jordon of Scotland,” he also examined first-hand the “noble” Liverpool and Man­chester Railroad “in all its stages of construction” and witnessed the “successful operation” of Hackforth’s steam locomotive on the Stockton & Darlington Rail­road. At every site he made copious notes and draw­ ings of what he saw: the manufacture of iron rails, the use of flanged locomotive wheels, the degree of ac­ceptable track curvature — even counting the number of strokes steam engines made as they moved from flat surfaces to inclinations and from straightaways to curves in their travels. While Whistler detected “no great display of the science of location” in the layout of British railroads and referred disparagingly to Telford and others as engineers “by instinct,” he nonetheless learned a great deal in Britain and soon put it to good use.7 Upon his return to the United States in May 1829, Whistler quickly became immersed in surveying and supervising the construction of the Baltimore & Ohio line. During this period he began to recognize that he possessed a special talent for administering large-scale enterprises. Such work not only encompassed surveys and construction but also day-to-day reports, account keeping, the maintenance and repair of rolling stock, the organization of traffic, and the coordination of work crews. What is more, colleagues as well as employees liked him. Unlike several of his talented associates, Whistler was an approachable person who dealt with people in a frank and open manner, and many acquaintances commented on his integrity and friendly demeanor. He also was a 6 Vose, Whistler, pp. 15-17; Forest G. Hi]], Roads, Rails, and Waterways (Norman: University of Oklahoma Press, 1957); Hill, “Formative Relations of American Enterprise, Government, and Science,” Political Science Quarterly 75 (1960): 400-419; Charles F. O’Connell, Jr., ‘The Corps of Engineers and the Rise of Modern Management, 1827-1856,” in Military Enterprise and Technological Change, ed. Merritt Roe Smith (Cambridge: The M.I.T. Press, 1985), pp. 87-106; O’Connell, “The United States Army and the Origins of Modern Management, 1818-1860” (Ph.D. diss., Ohio State University, 1982).

7 Whistler to Joseph G. Swift, Feb. 10, 1829, Swift Papers, New York Public Library [hereafter abbreviated NYPL].

Merrit Roe Smith

modest person who candidly recognized his foibles and limitations. Al­t hough he claimed several inventions during his life­t ime, he did not possess a particularly gifted mechani­cal mind. Rather his talent as an engineer lay as a technical synthesizer and organizer who could take command of far-flung operations and get results.8 Whistler’s stint on the B & O lasted nineteen months. From there he moved on to survey and initiate con­struction on three more eastern railroads before re­signing his Army commission on December 31,1833. The reasons for his decision to leave military service are not exactly clear. His biographer, George L. Vose, contends that he had grown tired of military life and found his work “more in the nature of an employment that a vocation”. To a degree this is true, but his letters also indicate a reluctance to leave railroad work and return to the more mundane port assignments of the Corps of Engineers, something which seemed imma­nent throughout 1832 and 1833. This, in addition to the fact that he had a growing family to support and found the monetary rewards and flexibility of the private business world too attractive to resist, sealed his decision. In effect, he was leaving the highly structured life the soldier to become a consulting engineer. In fact, he actually began to refer to himself as a “consulting engineer” at this time.9 Whistler was working on the Providence and Stoning-ton extension of the Boston & Providence Railroad when he decided to resign from the Army. Soon afterwards he received an inquiry from Patrick Tracy Jackson, one of the directors of the Proprietors of Locks and Canals on the Merrimack River, about moving to Lowell, Massachusetts to organize and superintend a new shop there devoted to the manufac­ture of railway locomotives. Jackson and his partners felt that Whistler was one of the few persons in America who could successfully initiate the venture, since he possessed considerable knowledge of the new technology through his British tour and his five and a half years experience on four railroads. In April 1834, the directors of Locks and Canals voted to make Whistler a company engineer at an annual salary of $3000 plus free housing. He immediately accepted.10 Whistler spent three years at Lowell and enjoyed considerable success in the locomotive business. Yet, even though he was comfortably situated and could support his family in a style befitting his social posi­tion, he felt uneasy and somewhat out of place in the manufacturing world of the Merrimack Valley.

Nowhere was this more apparent than in the products of the machine shop it8 Vose, Whistler, pp. 12,44-45; Parry, Whistler’s Father, pp. xiii, 25,46. For contrasting perspective, see Daniel H. Calhoun’s chapter on “Conflicts and Trend: The B & O Engineers, 1827-1838” in The American Civil Engineer (Cambridge: The Technology Press, 1960), pp. 123-140.

9 Vose, Whistler, pp. 18; Whistler to Joseph G. Swift, Dec. 2, 1831, February 12, 1834, Swift Papers, NYPL. 10 Vose, Whistler, pp. 18-19; George S. Gibb, The Saco-Lowell Shops (Cambridge: Harvard University Press, 1950), pp. 92-93.

• 307

Railroads in historical context: construction, costs and consequences

308 •

self. From the beginning the shop set out to replicate the 2-2-0 “Planet class” locomotives that Whistler had seen and admired in England. Under his supervision the Lowell Machine Shop built thirty locomotives and turned a handsome profit. Yet, as good as they were, the Lowell “Planets” continued to be copies of their British prototypes and little more. The fact that Whistler succeeded in organizing pro­duction at Lowell but made no significant design im­provements in the locomotives manufactured there suggests that he had a peripheral interest in mechani­cal things. His forte was “the science of location,” not machine production, and he recognized this early on. His stay at Lowell thus represented an interlude rather than a new departure in his career. Although his wife Anna and their children enjoyed Lowell’s bustling society, Whistler yearned to return to the work he knew best - railroad building. This, coupled with the onset of an economic depression and the consequent slowdown in the shop’s activities, prompted him to return to his old duties on the Providence & Stonington Railroad in 1837.11 Of all the places Whistler resided, Stonington, Con­necticut was his favorite, and he soon came to call it home. The nature of his work, however, virtually dictated that he could not remain in the same place very long. Railroad builders were like nomads, mov­ing from one place to another as opportunities arose.

Even before moving to Stonington Whistler had begun to consult with a number of other companies in the northeast section of the country. The largest of these was the Western Railroad of Massachusetts, a 160-mile trunk line that linked the cities of Worcester and Boston with Albany, New York and the burgeoning trade of the Erie Canal and America’s western regions.

Whistler joined the Western as a consulting engineer in the summer of 1836 and, in association with his old friend (and, since 1831, his brother-in-law), William McNeill, immediately became involved in the evalu­ation of surveys, the determination of possible routes, and the selection of locomotives and rolling stock for the road. Since the line had to cross some of the most forbidding terrain in the eastern United States — the Berkshire Mountains — these tasks presented Whis­tler with a formidable challenge. Aided by McNeill, William Swift, and another West Point classmate named John Childe, he settled on a route following the Westfield River west of Springfield that gradually as­cended the Berkshires until it reached the village of Chester. From Chester to the summit at Washington the grades became much steeper, often exceeding eighty feet per mile. Past Washington the grade leveled off and from that point the road stretched over undulating terrain toward Albany and the Hudson River Valley.12 During the first three years of the project (1836-39), Whistler resided in Ston11 Gibb, Saco-Lowell Shops, pp. 94-95; Vose,Whis-tler,pp. 18-19.23-24; Parry, Whistler’s Father, p. 50.

12 Charles E. Fisher, “Whistler’s Railroad: The Western Railroad of Massachusetts,” Railway and Locomotive Historical Society Bulletin 69 (May 1947): 7-9; Vose, Whistler, p. 23. On John Guide’s career, see Charles B. Stuart, Lives and Works of Civil and Military Engineers of America (New York: D. Van Nostrand, Publisher, 1871), pp. 177-94.

Merrit Roe Smith

ington and commuted to the railroad’s headquarters in Springfield. But once the line opened in 1839 between Worcester and Spring­field and construction began in earnest west of the Connecticut River, he began to spend more and more time in Massachusetts and eventually moved his family to Springfield. By 1840 he had become the Western’s Chief Engineer and, with the new full-time position, assumed primary responsibility for building the road to the Hudson Valley.

Although the work of excavating the roadbed and building bridges, stations, and terminals proceeded on schedule, not everything connected with the railroad went as smoothly. For one thing, Whistler experi­enced considerable difficulty finding locomotives that could haul profitable pay loads over the mountainous terrain. Neither the 2-2-0 “Planet class” locomotives purchased from the Locks and Canals Company nor Ross Winans’ Baltimore-built 0-8-0 (eight wheel) “Crabs” proved adequate to the task. Indeed, Winans’ ungainly engines, seven of which were delivered to the Western on Whistler’s recommendation, gave such poor service that they were either soon scrapped or sold. The Planets, on the other hand, performed relatively well for passenger service, but did not possess the tractive power to pull heavy freight loads over the rugged Berkshire range. Working in con­junction with the chief mechanic of the Locks and Canals Company, Whistler at length began to experi­ment with heavier 4-4-0 type locomotives in No­vember 1840, but he left the railroad without fully solving the problem. Eventually the Western adopted 4-4-0 locomotives for both freight and passenger service. Built initially at the Lowell shops and later by Holmes Hinkley of Boston, a variant of these ma­chines produced at the Western’s Springfield shops under the supervision of master mechanic Wilson Eddy acquired such a good reputation for reliability that they became know as “Clocks.” Compared with the early Lowell-built Planets which weighed twelve tons and had 12 x 18-inch cylinders with two 54-inch driving wheels, 4-4-0’s of the late 1840s weighed twenty tons and had 16 x 20-inch cylinders with four 54-inch drivers.

While it would be unfair to conclude that Whistler blundered in selecting locomotives for the Western Railroad, it would be equally amiss to characterize him as an innovator. Whenever he dealt with machinery, he sought the assistance of skilled mechanics. Even then, his plans and decisions, though thoroughly researched, lacked boldness and imagina­tion. Clearly locomotive design was not his strongest suit as an engineer.13 Whistler’s accomplishments on the Western Railroad brought him considerable prominence. By January 1842, the month service officially opened between Worcester and Albany, many dignitaries had already visited the New England line and praised the skill and dispatch with which it had been constructed. Acute observers also marvelled at the adroit measures Whis­t ler and his associates used 13 Fisher, “Whistler’s Railroad,” pp. 29-33,40-44,82-83; Vose, Whistler, p. 23. Also see Stephen Salsbury, The State, The Investor, and the Railroad: The Boston & Albany, 1825-1867 (Cambridge: Harvard University Press, 1967).

• 309

Railroads in historical context: construction, costs and consequences

310 •

to coordinate and control the efforts of hundreds of employees strung out over a 160-mile line. In large scale enterprises like the railroads, it was becoming increasingly evident that engineering and managing went hand-in-hand. Whistler certainly discovered this in his day-to-day activities as a consulting engineer.

In fact, he is probably as well remembered for developing administrative procedures for controlling traffic (and avoiding accidents) on the Western’s singletrack line as he is for building it. As Alfred Chandler has observed, Whistler’s contribu­t ions in this area helped inaugurate the advent of modern industrial management in America.14 Whistler’s dual competence as a manager and engineer set him apart from many of his contemporaries. A growing number of construction engineers could be found in the United States by the 1840s, but those who possessed a knack for successfully managing large-scale enterprises were still a rare commodity.

No­where was this more apparent than in the career of Whistler’s closest friend, “Mac” McNeill, a person who possessed abundant engineering talent but re­ peatedly foundered as a manager, bounced around from project to project, and eventually ended up an alcoholic. McNeill lacked the leadership qualities and temperament that people readily recognized in Whis­t ler. Consequently railroad and other business projec­tors often passed him by for positions of authority while they openly courted his brother-in-law.15 It did not take long before another invitation to build a major railroad arrived.

This time the call came from the court of Czar Nicholas I of Russia and the task involved linking the cities of St. Petersburg (now Leningrad) and Moscow, a distance of some 420 miles. The thought of building a railroad in Europe coupled with the promise of an annual salary of $12,000 attracted Whistler. Even though he would hold the title of “consulting engineer” and had to report to a Temporary Technical Commission consisting of twelve high ranking army officers headed

by Count Peter Kleinmichel (one of the Czar’s closest advisors), he nonetheless received assurances at the outset that the railroad was in fact his to build and that he would be its chief engineer. Given these pledges, Whistler decided to accept the appointment. On May 17,1842 he submitted a letter of resignation to the president and directors of the Western Railroad Corporation. Three weeks later the board of directors accepted his resig­nation, citing him for “genius and industry… in surveying, locating and constructing a railroad over a section of country and through mountain passes which seemed to bid defiance to the power of man.” While they were sad to see Whistler depart, they expressed pride in the fact that an American had been chosen to build the Russian raikoad.16 14 Alfred D. Chandler, Jr., The Visible Hand (Cambridge: Harvard University Press, 1977), pp. 97-98. Also see O’Connell, “The Army and Modern Management.”

15 Vose, Whistler, p. 37; Parry, Whistler’s Father, p. 23, 27, 50-51.

16 On Count Kleinmichel and the Russian railroad commission, see Vose, Whistler, pp. 28-29; Parry, Whistler’s Father, pp. 46-49, 55-57 and Richard M. Haywood, The Beginnings of Railway Development in Russia in the Reign of Nicholas I 1835-1842 (Durham, N.C.: Duke University Press, 1969), pp. 226-29.

Merrit Roe Smith

Whistler arrived in St. Petersburg in the summer 1842 and immediately plunged into the work of planning the railroad. The project differed from the Western Rail­road in several basic respects. Unlike the New Eng­land line, the projected Russian railroad had to trav­erse marshes and plains in nearly a straight line instead of mountains and winding river valleys. Along its entire length no grade exceeded twenty feet per mile. At the same time the Russian road was more that two and a half times longer than the Western. Although its construction presented Whistler with fewer technical challenges, its sheer scale involved him in an under­taking of monumental proportions. By the fall of 1846 over 60,000 serfs would be laboring on the railroad.

Whistler’s first year in Russia was intense but lonely. Having arrived in St. Petersburg without his family, he tried to alleviate his homesickness by immersing him­self in the affairs of the railroad. After meeting with Count Kleinmichel’s railroad commission and sitting through seemingly interminable conferences with its members, he began by reconnoitering the territory between St. Petersburg and Moscow. It was an ardu­ous task that had him slogging through bogs and riding through fields, meadows, and woodlands as many as seventeen hours a day. As he traced and re-traced the distance to be covered by the new railroad, he ques­tioned associates, made measurements, and took co­pious notes.

Upon returning to St. Petersburg and reporting to the Czar through Kleinmichel, he received royal permission to proceed. Whistler lost little time in establishing a survey line and dividing the work into two “directions” appropriately called the “Northern” and “Southern”. Colonel Paul Melnikoff headed the Northern distance while Colonel Nicholas Krofft took charge of the Southern. Both were trusted associates of Kleinmichel and both had visited the United States in 1840 to find an engineer for the St. Petersburg and Moscow Railroad. Their choice had been Whistler, but they agreed on little else. Indeed, Whistler’s biographer notes that their enmity toward each other was such by 1843 that not even “the thunderous Kleinmichel dared...(to)...bring them face to face.

They were, their compatriots jested, like the imperial eagle of Russia: the bird was one whole, yet its two heads gazed stubbornly opposite ways.

Their respective staffs aped the enmity, which led to much red tape and little cooperation Although Whistler liked both officers, their rivalry and hatred for each other did not ease the problems of building the railroad.17 While Melnikoff and Krofft headed the two main administrative divisions of the railroad, actual survey and construction work was further subdivided into On Whistler’s resignation, see Fisher, “Whistler’s Railroad,” pp. 96-97.

17 Parry, Whistler’s Father, pp. 63,67-68,71-72. The quotation is from Parry, p. 79.

• 311

Railroads in historical context: construction, costs and consequences

312 •

“districts” and “distances.” The basic working section of the railroad was the “distance,” each of which measured about 10 versts (5.8 miles) in length and was supervised by junior officers fresh from the Czar’s military engineering school. Altogether twenty-four engineer-lieutenants worked on the railroad, ten in the Northern direction and fourteen in the Southern. Supervising crews of a dozen or so members, they completed their surveys and had marked the line of construction by January 1843. However, since winter had already set it, the back-breaking work of excavat­ing and building the roadway had to wait until June when the six-month “working season” usually began in Russia.18 As he waited for the spring thaw to come, Whistler stayed in St. Petersburg patiently working on con­struction plans, conferring with associates, and preparing drawings for various structures along the line. Because long stretches of the railroad had to pass through poorly drained marshlands, his strategy called for the construction of a roadbed raised six to ten feet above ground level. Planned to accommodate double tracks, the trapezoidal earthworks, which measured thirty feet wide on top with six inches of gravel ballasting over a sand bed two feet deep, secured excellent drainage. Moreover, everything from the five foot track gauge to the line’s wooden Howe truss bridges conformed to a uniform pattern. Like many of his former Army comrades in the Corps of Engineers, Whistler was captivated by the concept of uniformity and regarded it as key concept in engineering and industrial practice. Indeed he wrote with reference to track gauges that “uniformity is of such importance that all deviations should be avoided until the advan­tages of a change are of such an obvious nature as to render a total change desirable.” The Czar and his military advisors warmly applauded Whistler’s ideas about system and uniformity and gave him complete latitude to implement his plan. After all, the railroad was being built primarily for military rather than commercial purposes, as the famous legend about the Czar taking a ruler and drawing a straight line between St. Petersburg and Moscow suggests. For Whistler it was like a dream come true. Other railroads had been built to uniform standards but none would be more thoroughly uniform than this one. The fortress-like earthworks, the rigid architectural uniformity of sta­t ions, engine houses, and repair shops, the military garb of employees, the manufacture of locomotives and rolling stock with standardized parts — all these things revealed the close connection that existed be­t ween military traditions and modern engineering prac­tice.19 Whistler’s faith in the power of technology to amelio­rate the condition of mankind reflected an idealistic attitude common among American engineers of the period. He firmly believed, for example, that the new mechanical technolo18 Ibid„ pp. 77-78.

19 On Whistler’s commitment to uniformity, see Parry, Whistler’s Father, pp. 91,158, 285; and Vose, Whistler, pp. 30-34,37-38. On the famous ruler incident, see Richard M. Haywood, “The ‘Ruler Legend’”: Tsar Nicholas I and the Route of the St. Petersburg-Moscow Railway, 1842-1843,” SJavicReyjew37(1978): 640-50.

Merrit Roe Smith

gies of the 19th century provided one of the vital pillars on which republican civilization and prosperity in America rested. Yet, as much as he believed in the connection between technological and social progress, he seemed puzzled, even paralyzed by social conditions in Russia.

Thousands of serf-laborers worked on the St Peters­burg and Moscow Railroad between 1843 and 1851. Most of them were bonded peasants who came from the western provinces of Russia and belonged either to private landlords or to the state. They had no political rights to speak of. The spade was their primary tool and their reputation as diggers was legendary. Through their back-breaking labor rose the earthworks of the railroad. As he travelled along the line on fortnightly inspections, Whistler witnessed first hand the appall­ing conditions under which these peasants lived and worked. To members of his family and in letters to friends he described their dank living quarters, their flea-infested sheepskin clothing, their perpetual drunk­enness, their obsequious behavior toward military of­ficers and noblemen, and their habitual tendency to pilfer things when the master’s back was turned. More often than not their diet consisted of hard bread, rotted potatoes, and maggot infested salt meat. Not surpris­ingly, they were perpetually ill with scurvy, diarrhea, and typhoid fever. Moreover, the discipline imposed by their military overseers was unfailingly vicious. If they complained or disobeyed, they were whipped over the buttocks with split birchcanes. Often times the birches were soaked in pickle brine to sharpen their effect. Those who tried to run away received even harsher floggings on their backs and over their sexual organs. A trip to the whipping post frequently meant death.

Not surprisingly, disease, exposure, and dis­cipline took their toll. According to one writer “the total cost of Whistler’s line was estimated at five thousand lives.”

For every mile of track laid, 12 workers died.20 Whistler witnessed this agony and suffering but felt powerless to do anything about it. Although he de­nounced the practices privately, he acquiesced to the country’s social traditions and hardened himself to their daily manifestations. First and foremost, he viewed his job from a technical and business perspec­tive. Even if, in his view, serfs were being mistreated and were dying by the thousands, his main task, as he saw it, was to complete the railroad. Clearly Whistler’s idealism had limits. Business took precedence over social concerns, just as it had with the treatment of Irish laborers on the Western Railroad in Massa­chusetts. Whistler recognized the moral dilemmas in his work, but he ultimately concluded that the ends justified the means. In the long run he believed that larger social benefits would accrue from the short term sacrifices and hardships that went into building the railroads. Such rationalization, however, gave little solace to those who suffered through the harsh disci­pline and hard living 20 Parry, Whistler’s Father, pp. 86-88, 156, 161, 204, 239-40, 258-59, 275. Cf. Reginald E. Zelnik, “An Early Case of Labor Protest in St. Petersburg: The Aleksandrovsk Machine Works in 1860,” Slavic Review 24 (1965).

• 313

Railroads in historical context: construction, costs and consequences

of the labor camps. To them, the price of progress came high.21 Whistler did not live to see the completion of the St. Petersburg and Moscow Railroad. In the fall of 1848 he contracted cholera while on one of his periodic tours of inspection and, after a prolonged illness, died the following spring. With his passing a fellow West Pointer, Major Thompson S. Brown, assumed the post of Consulting Engineer. Although railroad associates thought highly of Brown, for some unknown reason his Russian superiors made little use of his services. It was said that Count Kleinmichel simply refused to place the same confidence in Brown that he had in Whistler. To be sure, Whistler’s influence stood high among Russian officials and it doubtless lingered on after his death. When the first trains passed over the route in August 1851, members of the royal entourage celebrated the occasion with toasts to Whistler’s memory. They did so knowing full well that the conception, design, and execution of the system — the longest in the world — owed a great deal to the departed Yan­kee’s perseverance and diplomacy.22

314 •

II JOSEPH HARRISON, JR. (1810-1874): ARTISAN ENGINEER One of Whistler’s closest associates in Russia was Joseph Harrison, Jr. By background and training, Hanison came from a completely different world of experience. Yet, like Whistler, he would eventually end up referring to himself as a “civil” and, later, as a “mechanical” engineer.

Born in 1810, Harrison grew up in and around the Northern Liberties section of Philadelphia, Pennsyl­vania. Although his parents came from fairly well-to-do backgrounds, the family lost nearly everything when his father’s small retail business failed in the year of his birth. The experience of living under straightened circumstances coupled with the embar­rassment that accompanied his father’s bankruptcy made a deep impression on Harrison, one that he would carry throughout his life and try to rectify by regaining his family’s lost honor through personal achievement. Being poor as a child meant that, though “quick at learning,” he had limited prospects as he came of age. After attending grammar school “from time to time” until the age of fifteen, his father appren­ticed him to Frederick D. Sanno to learn “the art and mystery of a steam engineer.” Rather than pursue a mercantile career as his father and maternal grandfa­ther had before him, Harrison chose to become a machinist and make his fortune in the mechanical arts.

21 Parry, Whistler’s Father, pp. xvi, 121,156, 161.

22 Vose, Whistler, pp. 36-37; Parry, Whistler’s Father, pp. 325,331-35; Joseph Harrison to Mrs. Anna Whistler, Nov. 4,1851, to William L. Winans, August 21, 1851, Letterbooks, Joseph Harrison Papers, Historical Society of Pennsylvania, Philadelphia, PA [hereafter abbreviated HSP].

Merrit Roe Smith

Given his family’s straightened financial condition, he had few alternatives.23 Harrison spent only two years at Sanno’s Kensington machine shop. Owing to his mentor’s bankruptcy in 1827, he was forced to find another place to finish his apprenticeship. This time he turned to the larger, better know firm of Hyde & Flint, manufacturers of cotton machinery and stationary steam engines, and signed an indenture agreement with James Flint for a term of four years and three months “to learn the art and mystery of machine-making.” Harrison consid­ered the change advantageous because it allowed him in effect, to serve two apprenticeships in the time of one and thus secure “a double advantage.”

He clearly showed promise both as a machinist and as a “steady, reliable hand” because advancement came quickly. In less than two years, while still legally an apprentice to James Flint, he became foreman of the shop with thirty men and boys working under him.24 Harrison completed his apprenticeship in 1831 and remained with Hyde & Flint until the following year when he took a journeyman’s position with Philip Garrett, a Quaker gentleman well known among Phila­delphia mechanics for his “finer work” on such machinery as lathes and presses for bank-note engraving.

Although Harrison went to work for Garrett at a lower weekly wage ($8.50) than he received from Hyde & Flint ($9.00), he saw the new job as an opportunity to broaden his experience and profit himself in the long run. Indeed, he followed the same strategy in 1834 when he left Garrett and entered the neighboring American Steam Carriage Company of William Nor-ris and Colonel Stephen H. Long, one of the country’s earliest manufacturers of railway locomotives. De­ spite the fact that he considered the shop a “poor concern” with “bad workmen, bad tools, [and] bad foremanship - making locomotive engines of poor design,” he nonetheless gained an instructive intro­duction to locomotive building and established a good working relationship with Norris. From that point on, his primary occupation would be that of a railway engine builder and designer.25 Harrison ended his journeyman experience in 1835 when he returned to Philip Garrett’s shop as foreman of a newly established locomotive business called Gar­rett and Eastwick. In fact Harrison’s career underwent a subtle shift at this point, one that he probably did not fully recognize at the time. Not only had he achieved the status of a master mechanic, he also assumed a supervisory role in the firm and, with it, acquired an interest in its profitability. No longer a wageworker, he had become a manager and, as a consequence, began to separate himself from the everyday practices and pranks of co-workers. This new identity became manifestly clear in 1837 when Harrison became a junior partner in the firm of Garrett, Eastwick and Company. His new “interest,” to be sure, was that 23 Joseph Harrison, Jr., The Iron Worker and King Solomon. 2nd ed. rev. (Philadelphia: J. B. Lippincott & Co., 1869), pp. 31, 117. This rare limited edition volume is Harrison’s autobiography.

24 Ibid., pp. 117-18.

25 Ibjd., pp. 125-25; John H. White, Jr., A Short History of American Locomotive Builders in the Steam Era (Washington, D. C: by the author, 1982), p. 73

• 315

Railroads in historical context: construction, costs and consequences

316 •

of an employer rather than employee, a person of prop­erty and standing rather than a common shop hand.26 The years with Garrett and Eastwick proved to be among the most creative of Harrison’s career. Work­ing in concert with partner Andrew Eastwick, he de­ signed the locomotive “Samuel D. Ingham” in 1835 and, by 1837, was “building engines of a novel con­struction, with four connected driving-wheels, and a truck in front.” Incorporating flexible running gear, a single eccentric valve gear, and, most importantly, a three-point “equalizing lever” suspension system capable of adjusting the driving wheels to uneven tracks, these engines represented the prototypes of the standard 4-4-0 “American” locomotives that became so popular on railroads during the mid-nineteenth century. One engine in particular deserves special mention in this respect. Dubbed the “Gowan and Marx” and built in 1839-40 by Eastwick and Harrison for the Philadel­phia and Reading Railroad, this compact eleven ton 4-4-0 became famous for pulling a train of 101 loaded coal cars weighing 423 tons from Reading to Philadel­phia at a speed of 8-10 miles per hour. Significantly for Harrison and Eastwick, the feat attracted the atten­tion (and admiration) of the two Russian military engineers, Colonel Melnikoff and Colonel Krofft, who were visiting the United States on a tour of inspection of American railroads with orders from Czar Nicholas I to report on the best equipment then in use in Europe and America.27 By the time Melnikoff and Krofft witnessed the “Gowan and Marx” feat, Philip Garrett had retired from the firm and Harrison had become a full partner.

Times were not particularly auspicious, however. Owing to the depression that followed the Panic of 1837, Eastwick and Harrison limped along with “varied success,” producing no more than twenty locomotives over a four year period.

Consequently when Harrison re­ceived an invitation to visit St. Petersburg for the purpose of discussing a contract for building locomo­tives and “other machinery” for the St. Petersburg & Moscow Railroad, it seemed like a heavensent op­portunity. Not surprisingly, the letter transmitting the official communication came from his countryman, Major George W. Whistler.28 Harrison departed for Russia in March 1843, without any assurance that he would actually receive a con­tract. His purpose was to observe the situation, to discuss possibilities, and, if the situation looked at­t ractive, to make a bid. Once in St. Petersburg he discovered that a number of firms, including his old employer William Norris, had sent representatives to make presentations to Count Kleinmichel’s railroad commission. Yet, he also knew that he had strong sup26 Harrison, Iron Worker, pp. 125-26.

27 Ibid.; Joseph Harrison, Jr., “The Locomotive Engine, and Philadelphia’s Share in Its Early Improvements,” Journal of the Franklin Institute 63 (1872): 239-44; John H. White, Jr., American Locomotives: An Engineering History, 1830-1880 (Baltimore: Johns Hopkins University Press, 1968), pp. 48,152,287-95.

For good representations of these locomotives, see White, pp. 287-95.

28 Harrison, Iron Worker, pp. 126-27; Harrison, “The Locomotive Engine,” p. 244. Also see J. A. Cantrell, James Nasmyth and the Bridgewater Foundry (Manchester: Manchester University Press, 1985), pp. 158.

Merrit Roe Smith

porters in Colonel Melnikoff and, of course, Major Whistler. Indeed Whistler was well aware of Harrison’s work and, as early as 1840, even had begun to emulate it in acquiring 4-4-0 locomotives for the Western Railroad. As Consulting Engineer to Count Kleinmichel, he had considerable influence in deter­mining who would supply the locomotives and rolling stock to the Russian railroad. He also had long been an admirer of Ross Winans, a leading car and locomotive builder from Baltimore, and had already contracted with Winans to deliver a locomotive for testing on the new line. Consequently when Kleinmichel at length announced that a joint contract would be awarded to Ross Winans’ son, Thomas, and Eastwick and Hani-son, knowledgeable observers immediately recog­nized Whistler’s influence at work. Young Winans would bring a wealth of knowledge about car and engine building to Russia while Eastwick and Harri­son possessed the latest know-how in the production of the powerful 4-4-0 locomotives. Whistler’s quiet strategy was to bring them together as partners and offer them a lucrative three million dollar contract under the name of Harrison, Winans, & Eastwick.29 Signed in December 1843, the pact called for the delivery of 162 twentyfive ton locomotives with ten­ders and 2500 freight cars in five years. Harrison, Winans, & Eastwick also agreed to train 81 enginemen to operate the locomotives, something they ultimately failed to do. In return, the Czar agreed to provide the Americans with rent free use of the Alexandroffsky Head Mechanical Works, a large iron foundry and machine shop complex which employed 700 men and was located six miles from St. Petersburg on the eastern shore of the Gulf of Finland. Although Har­rison considered Alexandroffsky’s buildings “well de­signed,” he found its tools and machinery “of very old character and but poorly adapted to the work we con­templated doing.”30 This meant that the plant had to be re-tooled before manufacturing operations could begin. Since Eastwick and Harrison planned to close their Philadelphia shop in order to pursue the Russian contract, they shipped a portion of their equipment to the Alexandroffsky works. What else they needed, they purchased in England from such leading tool builders as Nasmyth & Gaskill, Joseph Whitworth, Francis Lewis, and Sharp and Co., all from Manch­ester or its environs. Of all Harrison’s business contacts in Great Britain, James Nasmyth proved to be the most constant and important. In addition to supplying the American builders and his partners with a varied assortment of machine tools, shop tools, stationary steam engines, and boilers, he introduced them to other British manufacturers and advised them on their purchases. Clearly not everything produced at the Al-exandroffsky works originated there. Harrison, Winans & Eastwick relied heavily on British suppliers such as William Crawshay for raw materials (spring steel, 29 Harrison, Iron Worker, p. 33; Harrison, “The Locomotive Engine,” p. 244.

30 Harrison, Iron Worker, pp. 52-54; Parry, Whistler’s Father, pp. 90-92, 126; Harrison to Gerard Ralston, Aug. 19/31, 1844, to William L Winans, Jan. 28, 1852, Harrison Letterbooks, HSP.

• 317

Railroads in historical context: construction, costs and consequences

318 •

cast iron, chisel steel, etc.) and iron products (car axles, car tires, rails, files,) as well as tools and machinery.31 In addition to needs for machinery and raw materials, Harrison, Winans, & Eastwick soon discovered that the government workers at Alexandroffsky knew little or nothing about locomotive- and car-making. As a consequence, the firm began to recruit skilled foreign workers to supervise major shops and departments at the Russian establishment. For their “top hand” and general manager of the works, Harrison, Winans, & Eastwick turned to a fellow American, Joseph Kirk. Although nothing is known about Kirk’s background, he probably had worked either for Eastwick & Hani-son or Winans before going to Russia in 1844. Of the other “principal” workmen at Alexandroffsky, six came from England, three from the United States, two from Germany, one from Norway, and six from Russia. All of them served as shop or department foremen. While the foreigners tended to oversee such specialty opera­t ions as the Locomotive Department, the Boiler Shop, the Truck Department, and the Foundry, Russians supervised more traditional work such as blacksmithing, spring making, model building, and woodworking. At peak production, the Alexandroffsky works employed over 1900 workers. Of that number the vast majority - some 1117 strong - were free Russian peasants while 495 were serfs either in possession of the Crown or of private landlords. The remainder of the workforce came mainly from the Sweden (164) and Germany (121), with only four from the British Isles.

Most em­ployees lived in houses “suitable to their various stations” at the Alexandroffsky head Mechanical Works.32 At an establishment as large as Alexandroffsky, it is not surprising that Harrison and his partners experi­enced disciplinary problems. Drunkenness proved to be a habitual problem, not just among Russian work­men but foreigners as well. Employees frequently came to work tipsy. Although he personally abstained from liquor and refused to serve it to house guests, Harrison seems to 31 Harrison, Iron Worker, pp. 79-82. The inventory of purchases from Nasmyth & Gaskill between 1844 and 1852 included one plate bending machine, several punching machines, one small shaping machine, one slide lathe, one planing machine, one slotting machine, five large engine lathes, five double flywheel steam engines, two boilers, and various shop tools (particularly files) of all sizes. On Harrison, Winans & Eastwick’s purchases of machinery, tools, and raw materials from British sources between 1844 and 1852, see Joseph Harrison’s Letterbooks, 1844-1852, HSP, esp. Harrison, Winans & Eastwick to Vickers & Co., April 15,1844, to Gerard Ralston, July 4, 7,11, 15/27, Aug. 2, 1844, to Nasmyth & Gaskill, July 9, 1850; Harrison to James Nasmyth, July 19, 23, 29, 1850, April 2, 7, 26, 1851, May 22, 29, 1851, to William Crawshay, July 22, Sept. 20, 1850, to William L Winans, July 23,1850, April 12, 26, May 24, June 21, July 19,1851, Jan. 16, 28,1852; to Thomas Winans, Aug. 16,1850, March 27, April 2,1851, Jan. 16,1852, to Brownlow, Pearson & Co., Sept. 22,1850, to A. Kamensky, April 25, May 1,1851, to C. Sanderson, May 13,1851, to Shanks & Co., Jan. 20,1852. On Harrison’s relations with JamesNasmyth, see Cantrell, Nasmyth, pp. 158,197-98,203,210-11. Also see R. Dickinson, “James Nasmyth and the Liverpool Iron Trade,” Transactions of the Historical Society of Lancashire and Cheshire 108 (1956): 99 ff.

32 For employment figures on Alexandroffsky, see Harrison, Iron Worker, table following p. 72; Parry, Whistler’s Father, pp. 155, 284; and Memorandum of an Agreement made by Joseph Harrison...” July 10, 1851, Letterbook 2, Harrison Papers, HSP.

Merrit Roe Smith

have accepted drinking among workers as a customary right But he frowned on drunkenness, especially when it resulted in lost time or ruined work, and did not restrain the police who guarded the works from dealing severely with native workers who showed up inebriated. As on the St. Petersburg & Moscow Railroad, the whip was used liberally at Alexandroffsky. As for foreign workers, they usually received warnings and fines for infrac­tions of the rules and, if the practice persisted, would be summarily dismissed without letters of reference.

On at least one occasion, however, two enginemen -a German and a Swede were publicly whipped with birch rods and sentenced to three years hard labor for wrecking two locomotives on the railroad. At the same time, Harrison and his partners received a rep­rimand from Count Kleinmichel’s railroad commis­ sion “for putting these two unskillful men...in charge of the locomotives.” Harrison later wrote approvingly in his autobiography that Russian railroads ran “on the principle of military discipline, - prompt and severe punishment, and no excuses.” As a result, he noted, “fewer and less serious accidents have occurred on the Russian railways than anywhere else. Severe punish­ment had at least this effect.”33 Despite his ambivalent feeling about alcohol, Harri­son held strong opinions about wages and working hours. When his father-in-law wrote to him in 1851 about the movement in Philadelphia for an eight-hour work day, he bristled at the notion.

It seems always that trouble begins with the workpeople whenever there is a tolerable long run of plenty of work. They are not so much in the wrong either in trying to get good wages in brisk times, as they are screwed tight enough when little work is to be had. Still sometimes I have known this class of men to be very ungrateful. My sympathies will always be with the workingmen except in regard to the hours they wish to work I always thought that ten hours per day was rather too little for a man to work and that short hours induced idleness and dissipation. You have said in some of your late letters that workmen have turnedoutfor 8 hours per day. With the majority of workmen it is all non­sense to talk of short hours on account of giving more time to intellectual improve­ment. Those who have a desire to cultivate their minds will do it if they work 15 hours a day andnothingwillstop them.

Withmostthe short hour scheme is only an excuse for getting more time to be idle. Give a man as high wages as you like but do not shorten their hours of labor too much. I was never more happy than when I was working (and hard too) eleven or twelve hours per day. I never would have anything to do with any trade unions or turn outs for wages or hours. I think every 33 Harrison, Iron Worker, pp. 77, 87-88.

• 319

Railroads in historical context: construction, costs and consequences

individual ought to be free to make any bargain he likes with his employers on these points. 34

320 •

These words reveal just how far Harrison had moved away from his craftoriented, working class back­ground by 1851. While he begins by acknowledging the right of workers to get “good wages,” he ends by denying them the means usually associated with ob­taining such demands. Individuals, not unions or strikers, should bargain with employers. Shorter hours, on the other hand, are the work of the devil and should be avoided at all costs. The ideal worker, Harrison thought, “wiD most assuredly do more justice to him­self and give more satisfaction to his employers by pursuing a steady, straightforward course in all things entrusted to him without interfering or complaining of what others have done or are doing.” He impressed on employees that “we did not want anyone to turn upside down and revolutionize what already existed but to quietly go to work and make the best of what he had entrusted to him.” Quiet compliance and steadiness thus became the order of the day at Alexandroffsky. As an employer, Harrison seemed increasingly to imbibe the ethos of his military employers.

In doing so, he ironically denied the very individuality that had once allowed him to express himself creatively as a journeyman mechanic and eventually to achieve status as a man of affairs.35 Once underway, manufacturing operations moved swiftly at Alexandroffsky. “Our work is in full activity now,” Harrison wrote excitedly to a business associ­ate in August 1844. As the principal partner in the firm, Harrison headed the Locomotive Department while Thomas Winans supervised the newly erected Car Department. For reasons that were never made completely clear, the third partner, Andrew Eastwick, played a minor role in the venture. When Eastwick abruptly left the firm and returned to America in 1849, Harrison explained to his father-in-law that “he leaves the concern, because his conduct during the whole time he has been here has been so distasteful to Mr. Winans and myself, and had become so intolerable that we were glad to let him go out of the concern.”

“Mr. Eastwick goes out with a large sum of money,” Harrison wrote his sister in October 1850, “a great deal more than he ever deserved.”

He is now a rich man but owes nothing that he possesses to any good that he has ever done by himself. He has been our Millstone, our incubus, my contempt and disgust. Oh! how my blood boils when I think of that man’s conduct to me.... It was no doubt wisely ordered that I did bear it as I did.

Had I turned upon him at times when I had so much provocation, some violence might have been done by myself which might have embittered my34 Harrison to Stephen Poulterer, June 20,1851, Letterbook 2, Harrison Papers, HSP. 35 Harrison to William Winans, May 24, 1851, to Gerard Ralston, May 30, 1851, to Thomas Winans, April 7, 1853, Letterbooks, Harrison Papers, HSP.

Merrit Roe Smith

wholelife. Lethimgo, he will find his own punishment. He finds it now in the utter dislike which all look upon him with.

In Harrison’s opinion, Eastwick was a “humbug,” “the nastiest meanest man that ever lived.”36 Personal animosities aside, the American enterprise at Alexandroffsky flourished. By the fall of 1845, some 1600 mechanics employed in the complex’s nine major shops were turning out ten cars a day and one locomotive a week. A year later the workforce had expanded to 1920 workers and output had jumped to ten freight cars a day, four passenger cars a month, and six loco­motives and tenders a month. On July 26, 1847, Harrison wrote jubilantly to his father-in-law that “We shall have all the Locomotives done in three Months from the present time at the rate we are now making them - eight a month or two in six working days.” “This beats our old operations at home,” he continued, “and it is about two engines a month ahead of any establishment in any country that we have any knowledge of.” The prediction proved correct. By mid-September Harrison again reported that “All our business matters are going smoothly as we could wish.”

Our original contract for Locomotives and cars will be entirely finished in six weeks from the present time, with the exception of putting the Machinery in operation on the road We shall have built all the Engines in fourteen months shorter time than our con­tract calls for, and could make 100 more Lo­comotives within the time fixed by our con­tract, if they were needed. 37 In the end, Harrison, Winans and Eastwick delivered 162 twenty-five ton locomotives and tenders, 2000 eight-wheel box cars, 500 eight-wheel platform cars, 70 eight-wheel passenger cars on “the American plan,” 6 eight-wheel post cars, and 1 sixteen-wheel “Imperial car” to the Russian government. On several occasions Count Kleinmichel expressed satisfaction with the work and complimented the American build­ers for their “superbly good and even graceful prod­ ucts.” But the most gratifying moment had come in the spring of 1847, when Czar Nicholas, accompanied by the Grand Duke Alexander, Grand Duke Constantin, Major Whistler, and an entourage of a dozen or so dignitaries visited the Alexandroffsky shops and pro­ceeded to review every aspect of the manufacturing operation. After inspecting the foundry with its mammoth Nasmyth steam hammer, the locomotive erecting shops, and the car shops, the group ended up at the train sheds where, according to one writer, “a bright array of one hundred and eighteen locomotives freshly painted and cleaned, all of uniform shape 36 Harrison to Gerard Ralston, Aug. 19/31, 1844, to Stephen Poulterer, April 4,1849, to Sarah Harrison, Oct. 4, 7,1850, Letterbooks, Harrison Papers, HSP. 37 Harrison to Poulterer, March 14/26, 1846, July 26, Sept. 9/21, 1847, to “brother Charles,” March 21, 1846, to C. C. Harrison, Nov. 8/20,1847, Letterbooks, Harrison Papers, HSP; Parry, Whistler’s Father, pp. 238, 283.

• 321

Railroads in historical context: construction, costs and consequences

322 •

and color... pleased the Emperor’s soldiery eye.”38 To be sure, uniform standards counted for much among the military officials of the St. Petersburg & Moscow Railroad. Whistler emphasized uniformity as did Count Kleinmichel and his staff. Indeed, Whistler’s biographer claims that “the engines of each class [freight and passenger] were made throughout from the same patterns, so that any part of one engine would fit the same position on any other.” The same author also contends that “the use of standard patterns, uni­formity of design, and duplication of parts, was ap­plied ... to the rolling stock” as well. However, the record is unclear about how much emphasis Harrison and his partners actually placed on building locomo­tives and rolling stock with standardized parts. Other than mentioning the need to have “iron of uniform goodness” for car axles, Harrison makes no reference to the subject in his personal correspondence. Lacking both physical evidence and first-hand testimony, one is forced to conclude that Harrison, Winans & East­wick worked from standardized patterns but did not necessarily produce locomotives and cars with inter­changeable parts. Their’s was nonetheless an early manifestation of the “American system” of manufac­ turing in the railroad business.39 Czar Nicholas expressed satisfaction with what the Americans had accomplished at Alexandroffsky by sending each of the partners a beautifully embellished diamond ring worth, in Harrison’s estimation, “not less than three thousand dollars in gold.” At the same time he bestowed the Cross of the Order of St. Ann of the second degree on Major Whistler and promoted Colo­nels Krofft and Melnikoff to the rank of general. With these recognitions came more work for the Ameri­cans, including a two million dollar contract to provide the materials for a cast iron bridge (“the largest cast iron Bridge in the world,” according to Harrison) over the Neva River at St. Petersburg. Harrison claimed to have “suffered large losses” in the venture and, per­haps as partial recompense as well as recognition for the achievement, the Emperor decorated him with the Order of St. Ann upon its completion in the fall of 1850. By this time Harrison was enjoying the profits from a number of ventures, including, interestingly enough, the importation of sugar to Russia. But the greatest opportunity came earlier in 1850 when Har­rison and Winans, in partnership with Winans’ younger brother William, signed a twelve-year contract for maintaining and repairing the engines, tenders, and cars of the St. Petersburg & Moscow Railroad. Al­though the exact financial terms of the agreement remain vague, Harrison indicated that payment was based on the number of versts run by the machinery and that the contract was “of much greater magnitude than all the former contracts put together,” that is, in excess of five million dollars. Flushed with enthusi­asm over the new 38 Harrison, Iron Worker, p. 52; Parry, Whistler’s Father, pp. 238, 283-85.

39 Vose, Whistler, pp. 33-34, 38. On the so-called “American system,” see David A. Hounshell, From the American System to Mass Production (Baltimore: Johns Hopkins University Press, 1984): Merritt Roe Smith, “Army Ordnance and the ‘American System’ of Manufacturing, 1815-1861,” in Smith, Military Enterprise and Technological Change (Cambridge: MIT Press, 1985), pp. 39-86.

Merrit Roe Smith

arrangement, he informed Thomas Winans (who was visiting Baltimore at the time) that “William and I are going into things about the works with a good deal of spirit, rearranging old things and projecting new ones, and we think when all our plans are carried out that we shall have about the finest establishment in the world. Everything goes very pleasantly and we find our present operations amusing rather than the contrary.” Even the cautious Harrison had to admit that the profit margin on the second railroad contract was “very large.” When the pay­ments for their services began coming in, he was as­tonished by the amounts.

“Our revenue this month with 2 trains in each direction [from St. Petersburg and Moscow] will be 60,000,” he informed Thomas Winans in January 1852, “next month with 3 it will be 85 or 90,000 and the prospect is that in ‘52 we will have one and a half or two million. The Count [Kleinmichel] begins to think we will take too much of the revenue of the road and I see there will be a tendency to break us down.” But, he quickly added, “they can’t faze us.”40 Although he continued to draw a handsome income from the second contract for more than a decade, Harrison’s sojourn in Russia ended in 1851 when he moved his family to London and conducted business there as a special agent for Winans, Harrison, & Winans. The following year he returned to Philadel­phia, leaving William Winans in charge of the com­pany’s affairs at Alexandroffsky.

Ostensibly he did so because of his wife’s ill health, but it also is clear that other considerations -principally family ones- beck­oned him home. He liked Russia a great deal and, to a degree, even “dreadfed] living in America again.”

But, he wrote Thomas Winans, “I have been away from America within a month of nine years - and I must return... or I may never see those who have the greatest claims upon my duty.”41 Harrison retired from Russia with a large personal fortune. The extent of his wealth is not precisely known, but judging from his partners’ affluence, he was worth at least two million dollars. While still in Europe, he began to invest in real estate in and around the city of Philadelphia and, according to his own account, “erected numerous and costly buildings.” In addition to his activities as a landlord, he bought a large block of stock ($50,000) in the Pennsylvania Railroad and strongly supported J. Edgar Thomson for the road’s presidency. He also continued to devote time to mechanical pursuits and, in 1859, produced a sectional “safety boiler bearing his name which re­ceived considerable acclaim and won prestigious medals at the London International Exhibition in 1862 and from the American Academy of Arts & Sciences (Rumford Medal) in 1872. Per40 Parry, Whistler’s Father, p. 286; Harrison, Iron Worker, pp. 34-35, 55, 67, 78, 94-99; Harrison to Stephen Poulterer, Sept. 29, Oct. 24, 1850, to William L. Winans, Dec. 23, 1851, to Thomas Winans, Oct 18, 1850, Jan. 16, 1852, Jan. 11, April 7, 1853, Letterbooks, Harrison Papers. By March 1851, the Russian government owed Winans, Harrison & Winans 300,000 silver roubles. See Harrison to Thomas Winans, March 27,1851, ibid.

41 Harrison to his parents, Oct. 20, 1850, to Thomas Winans, July 11, 1851, Jan. 16, 1852, to William Winans, April 26,1851, to Capt. J. F. Kruger, May 1, 1851, Letterbooks, Harrison Papers; Harrison, Iron Worker, pp. 34, 36,127-28,137,151.

• 323

Railroads in historical context: construction, costs and consequences

324 •

haps the most intrigu­ing turn his life took at the time, however, was his interest in art. Harrison had always been an avid reader and, judging from book invoices in his personal papers, read widely. Whether his interest in the fine arts first took root in Russia with his exposure to the great collections at St. Petersburg and his close friend­ship with the Whistler family or whether it developed earlier is not known. Whatever the case, he began to purchase original oil paintings by distinguished artists as early as 1851, and continued to collect for the rest of his life. Among the works represented in his collec­t ion were Benjamin West’s “Penn’s Treaty with the Indians” (purchased from George Catlin for $500), West’s “Christ Rejected” (c. 1815), Charles Willson Peak’s “The Artist and His Museum” (1822), Rem­brandt Peale’s “George Washington, Patriae Pater” (c.

1824), and, a personal favorite, Christian Schuessel’s “The Iron Worker,” specially commissioned in 1865. Clearly Harrison prided himself on his patronage of the arts and claimed credit for establishing “the most extensive, and probably the first gallery of Art in Philadelphia.” Today a portion of his rather extensive collection may be found at the Pennsylvania Academy of Fine Arts.42 What do these varied activities reveal about Harrison’s career? Can some deeper meanings be attached to them? The answer, I believe, is yes, and an indication is found in how Harrison perceived himself both as a person and as a practitioner of the “mechanics arts.” Central to this identity process is how he came to call himself an engineer.

Harrison’s autobiography, published in 1869, reveals a number of things about his personality: his sensitiv­ity, his integrity, as well as his deep commitment to the protestant work ethic and classical liberal political views. But perhaps most deeply felt of all was his determination to atone for his father’s embarrassing business misfortunes - “this dark hour of our history” - and to redeem the family’s good name and lost fortune. Because his family lacked the resources to establish him, as his father and grandfather before him, in the mercantile business, he became a craft apprentice. Yet even as a journeyman machinist, he prided himself on the fact that employers like William Norris considered him a “gentleman.” Indeed, Harri­son notes in his reminiscences, “1 was a gentleman, compared with the ill-mannered and ignorant crew around me.”43 Being recognized as a gentleman carried great weight with Harrison because it reinforced his feelings about self-integrity and gave him a status beyond that of an ordinary mechanic. Not surprisingly, once he began to experience success as an inventor and eventually became a partner of Garrett & Eastwick, he began to dissociate himself from the shop culture of the ma­chinist and identify himself as “a man of business”. Nonetheless, he never completely abandoned his iden­tity 42 Harrison to Stephen Poulterer, April 24, May 29, July 15,1851, Jan. 6, 22,1852, to J. Hubbard & Co., June 16,1851, to George Catlin, Aug. 1,1851, Letter-books, Harrison Papers; Harrison, Iron Worker, pp. 36-37.

On Harrison’s sectional boiler, see Louis C. Hunter, A History of Industrial Power in the United States, 1780-1830. Vol. 2, Steam Power (Charlottesville: University Press of Virginia, 1985), pp. 332-35.

43 Harrison, Iron Worker, pp. 101-105,124.

Merrit Roe Smith

as a craftsman and, as late as 1852, referred to himself as “an engineer and machinist.” The interest­ing change here, of course, is that he had begun to call himself “an engineer.” The record is not clear about how this transition occurred.

It doubtlessly took place while Harrison was in Russia during the 1840s and probably owed something to the influence of his close personal friend, George W. Whistler, whose official title was that of “Consulting Engineer” to the Russian railroad commission. In any case, we find Harrison asking his father-in-law in 1850 to address him as “Civil Engineer” because some other J. Harrison had received his mail in St. Petersburg.44 Harrison continued to refer to himself as a civil engi­neer after returning to the United States in 1852. But sometime during the 1860s, probably after being awarded a medal for his safety boiler at the London Exhibition of 1862, he began to designate himself as a “Mechani­cal Engineer.” No explanation is given for this change in appellation. Perhaps Harrison began to realize that, contrasted with his friend Whistler, the core of his work lay in the mechanical arts. He probably also felt that it was no longer necessary to refer to himself as a “civil engineer in order to distinguish himself from the military authorities he dealt with on a daily basis in Russia. Whatever the case, the shift is significant because it indicates that Harrison, like many contem­poraries, had no clear or formal idea of what engineer-ingmeant. Harrison died in 1874, at the very time such disciplinary definitions and distinctions were begin­ning to be formalized through the establishment of professional engineering societies and schools. His identity as an engineer denoted a conscious differen­t iation in status that, as an employer and manager, set him apart from the shop worn mechanic yet main­ tained his identity as a member of “the useful classes.” It had virtually nothing to do with formal engineering disciplines as we know them today. In Harrison’s mind, becoming an engineer meant being a business­man and a gentleman. To the end his skill as an inventor continued to be identified with his early background and training as a Philadelphia crafts­man.45 44 Ibid., Harrison to Stephen Poulterer, Oct. 25, 1846, July 12, 1850, to Hon. Stephen A. Benson, Jan. 13, 1852, Letterbooks, Harrison Papers. On Harrison’s friendship with Whistler, see Harrison to Poulterer, March 28, 1849 (“the best friend I ever had in my life”), July 25,1849 (“truly the best friend I had ever found out of my own family”). For similar observations about the connection between engineering, science, and gentlemanly standing, see R. A. Buchanan, “Gentleman Engineers: The Making of a Profession,” Victorian Studies 26 (1983): 407-29; Arnold Thack-ray, “Natural Knowledge in Cultural Context: The Manchester Model,” American Historical Review 79 (1974): 672-709; Calhoun, American Civil Engineer, pp. 193-98.

45 Harrison, Iron Worker, pp. 35,152; Harrison, “The Locomotive Engine,” pp. 161,233. On the emergence of engineering professionalization, see Monte A. Calvert, The Mechanical Engineer in America. 1830-1910 (Baltimore: Johns Hopkins University Press, 1967); Raymond H. Merritt, Engineering in American Society, 1850-75 (Lexington: University Press of Kentucky, 1969); Bruce Sinclair, A Centennial History of the American Society of Mechanical Engineers. 1880-1980 (Toronto: University of Toronto Press, 1986); A. Michael McMahon, The Making of a Profession: A Century of Electrical Engineering in America (New York: IEEE Press, 1985); David F. Noble, America by Design (New York: Alfred A. Knopf, 1977), pp. 3349; Robert Perucci and Joel Gerstl, eds., Engineers and the Social System (New York, 1969).

• 325

Railroads in historical context: construction, costs and consequences

DANIEL LESTER HARRIS (1818-1879): ENGINEER AS BUSINESSMAN

326 •

Compared with Joseph Harrison, who came to call himself an engineer rather late in his career, Daniel Lester Harris’ odyssey to engineering seems more forthright. Raised in and around the cotton textile manufacturing village of Plainfield, Connecticut, Harris’s ancestry went back to the days of Roger Williams and the early settlement of Rhode Island. His father, a modestly successful mill owner and merchant, de­cided early on that his son should receive “a thorough English education” and, to that end, sent Daniel first to the Plainfield Academy and thence to Wesleyan Uni­versity in 1835. His expressed wish was for Daniel “to acquire such a knowledge of mathematics, and the various branches of natural philosophy, as will qualify him for an efficient, practical man.”46 Being practical counted for much in the Harris house­hold. From the outset the father “impressed upon his children the importance of being industrious and use­ful.” (150). To give an air of reality to his teachings he sent Daniel to work in the mill “at a very early age, during his school vacation.” The experience had its intended effect. By age fourteen the youth was work­ing sixteen-hour days and keeping an account book in which he recorded “every pecuniary transaction he made.” Indeed, Harris prided himself on being a hard worker. Industry, self-reliance, diligence, prudence, honesty, candor — these “old fashioned virtues” pro­vided the values around which he shaped his life. “He was a Puritan by descent, and a Puritan in his training,” a friend later observed. Such training made him a rather plucky, matter of fact person whose tastes “ran in the line of the plain, the substantial, and the useful.” In Harris’ life the puritan ethic, the pursuit of engi­neering, and the spirit of capitalism became closely intertwined.47 Harris’ turn toward engineering came early. Capti­vated by the internal improvements boom that swept the nation during the 1830s, he expressed a desire to become a “civil engineer” upon entering Wesleyan. As it turned out, his “thorough English education” quickly deviated away from the classics (subjects which he “never regarded...as of sufficient importance to merit my attention”) toward the study of algebra, geometry, trigonometry, calculus, and “everything else pertaining to mathematics”.48 Harris spent three years in the “scientific department” at Wesleyan. During that time he rounded out his program with subjects that he considered of “greater utility” than “the dead languages.” Indeed, he wrote his brother that “The sciences are every day becoming of more and more consequence, and as they increase it is found necessary to make room for them in courses of study, at the expense of less important branches.” For Harris “the less important branches” 46 Henry M. Burt, ed., Memorial Tributes to Daniel L. Harris, With Biography and Extracts from His Journal and Letters (Springfield, MA: Privately Printed -Henry M. Burt, 1880), pp. 143-55.

47 Ibid., pp. 18, 22, 36, 40-41, 44-46, 114, 150, 179, 236.

48 Ibid., pp. 153-59, 250-51, 319.

Merrit Roe Smith

included Greek, Latin, French, and literary studies; the important branches embraced mathematics, astronomy, chemistry, and geology. He derived particular “pleasure...from the propositions of geometry, the demonstrations of trigo­ nometry, and the problems of mensuration.” In them he discovered “hidden beauties” and, as he wrote his brother, felt “sublime emotions...in contemplating the order and regularity with which the heavenly bodies pursue their courses through space; the astounding velocity with which they move; and the unerring precision which attends man’s calculations respecting them.” Harris’ association of order, regularity, speed, and precision with an almost poetic sublimity speaks volumes about the values to which he ascribed and ultimately how his social beliefs linked up to his worldly endeavors.49 As much as the rigor and symmetry of science influ­enced Harris, a collegiate religious experience served to reinforce his inclination toward ascetic behavior.

As a youngster he had gone through the motions of religious belief but had never felt called to a Christian life. With the sudden death of a college friend in the fall of 1836, however, that changed. “Will you believe it?” he wrote his father, “I am firmly resolved, by the Grace of God, henceforth to lead a Christian life.”

I have, for a long time, occasionally felt that I had need of an interest in the atoning merits of the Redeemer.... Suffice it to say, I firmly trust I have found pardon at the throne of grace, and am firmly resolved henceforth to walk according to the precepts of the Bible, so far as the frailty and inconsistency of my nature will permit.50 Walking according to the precepts of the Bible pro­vided Harris with a sort of moral analog to the disci­pline and regularity he had found in science. A friend later observed that “He firmly believed that God governs the world, including all the affairs of men, by definite and established laws — laws that men can know and ought to obey.” For Harris appropriate conduct in this world meant being temperate, not only in clothing and other forms of outward display but also in the avoidance of alcohol and tobacco. It also meant embracing a work ethic that emphasized diligence and industry and castigated its opposites, idleness and dissipation. For Harris moral affairs and practical affairs were inseparable.

The certainty with which he pursued business and engineering reflected the cer­ tainty of his religious convictions.51 When Harris left Wesley an in 1838, he was well educated in the sciences but lacked practical experi­ence. Intent upon learning the “first principles” of railroading, he found a job with James Laurie, a well-known Scottish engineer who was surveying the line for the Norwich and Worcester Railroad in Connecti­cut.

Getting at first principles meant, in effect, work­ing from the bottom up just as craft apprentices learned by doing. Indeed, in later years Harris frequently re49 Ibid., pp. 245, 247.

50 Ibid., pp. 156-57.

51 lbid., p. 97.

• 327

Railroads in historical context: construction, costs and consequences

ferred to Laurie as his old master. To be sure, much of the structure of early civil engineering in America was patterned after the craft tradition.52 Railroad builders, like millwrights, were constantly on the move. Harris came to realize this fact early on. In the space of five years he worked for six railroads: the Norwich & Worcester (from April to September 1838; returning again from June 1839 to January 1840); the New York & Erie (from September 1838 to June 1839); the Albany & West Stockbridge (from January to August 1840); the Troy & Schenectady (from August 1840 to April 1843); and the Springfield & Hartford (April to September 1843). “My busi­ness,” he wrote his brother, “is of that roving, unsettled kind which does not allow me to count with any certainty on the events of the future. Of civil engineers it is perhaps more literally true than of any other class of the community, that they “know not what a day or an hour may bring forth.’

328 •

They are children of fortune, and the old dame never wearies of playing off her odd freaks upon them. They are here, there and everywhere - perfect cosmopolites, wandering up and down the earth without a fixed residence upon its surface. It being their business to prepare ways for other people to keep in motion, they are constantly moving about themselves, by way of example.53 Harris started out by carrying the survey chain for Laurie, and from that rather menial position, moved in succession from rod-man to leveler, to surveyor (compass-man) to group leader. “The leader,” he informed his father, “has nothing to do with the instru­ments, but he spends the most of his time in exploring the country and directing the surveyor as to the loca­tion of the lines.” Two years of hard labor, however brought few real monetary advances and by the spring of 1840 Harris had grown restive about his future in railroading. “Well, here I am,” he wrote sarcastically to his father in June 1840, “engaged in the very poetry of engineering, - wallowing through rye fields, climb­ing fences, and wading swamps and rivers.”

I might have staid some two weeks longer on the Norwich Road: but, as I might, after that, be entirely out of employment, I thought it preferable to accept the offer here [the Al­bany & West Stockbridge Railroad]. My compensation is not at present what it should be, but I have a reasonable hope of prefer­ment. In accepting a low situation here, I plead the excuse of the apothecary who sold poison to Romeo: ‘My poverty, not my will consents’.... Surely, these are hard times with our profession.54 52 For documentation at this point, see Calhoun, American Civil Engineer and Calvert, Mechanical Engineer.

53 Burt, Memorial, p. 268.

54 Ibid., pp. 168, 282-83.

Merrit Roe Smith

Owing primarily to the depression that scourged the United States following the Panic of 1837, Harris never received the preferment he had hoped would be forthcoming from either the Erie or the Albany & West Stockbridge railroads.

Instead his big break came in the summer of 1840 when his old mentor, James Laurie, asked him to join the survey of the Troy & Schenectady Railroad in upstate New York as assis­tant engineer. With the completion of the survey, Laurie placed Harris in charge of the First Division, eight miles of the heaviest construction on the line.

This was the young man’s first real opportunity to demonstrate his skills as a high level supervisor. He evidently carried out his duties to the utmost satisfac­ tion of his employers because, with the line’s comple­t ion in 1843, he was offered the position of general superintendent of the railroad.55 As it turned out, Harris declined the position because he felt that the salary ($800 per year) was too low. By then, however, several things had happened that would alter the course of his career.56 While working on the Norwich & Worcester Railroad, Harris had become friendly with a construction con­tractor named Azariah Boody (1815-1885). After leaving the Norwich & Worcester, Harris kept in touch with the Boodys and on several occasions visited them at their new home in Pittsfield, Massachusetts.

There he met Mrs. Boody’s sister, Harriet O. Corson, and, after year’s courtship, married her in May 1843. The couple found time to visit Harriet’s family in upstate New York and take a brief honeymoon trip to Niagara Falls, but no sooner had they returned to Pittsfield than Harris left his wife at the Boody’s and headed for Hartford, Connecticut to undertake a survey for a proposed railroad between that city and Springfield, Massachusetts. Since the distance between Hartford and Springfield was relatively short, Harris completed the survey by mid-July and finished the task of pre­paring estimates of the cost of construction by the end of August. The Hartford-Springfield survey consti­tuted a sort of milestone in his career because it was the first job he took on as an independent survey contrac­tor with his own hand-picked crew. Interestingly at this very time one also begins to see Harris shift from calling himself an engineer to a new appellation, that of businessman.57 Hams enjoyed working in the bustling picturesque area of the Connecticut Valley and decided to move there permanently in the fall of 1843. Working out of a small office in Springfield, he spent the next year doing land surveys for local clients like John Ames, a local paper manufacturer, and Charles Stearns, an­other local businessman whose water power rights on the Mill River were in dispute with the Springfield Armory. Harris was just getting comfortably settled 55 Ibid., pp. 174,176, 285, 291, 296.

56 Harris to the President and Director of the Schenectady & Troy Railroad, Sept. 30, 1842, D. L. Harris Papers, Baker Library, Harvard University Graduate School of Business Administration [hereafter abbreviated HBS].

57 Burt, Memorial, pp. 177-80.

• 329

Railroads in historical context: construction, costs and consequences

330 •

in Springfield, however, when his brother-in-law urged him “to give up his civil engineering” and join him in partnership with another New Englander, Amasa Stone (1818-1883), in the railroad construction business. Harris agreed to the proposal and the firm of Boody, Stone & Harris was formed in 1845 to exploit the upswing in railroad building that began to appear with the end of America’s first great depression.58 Boody, Stone & Harris proved successful because it brought together three individuals whose talents nicely complemented each other. Boody was a wellknown construction contractor; Harris had a good reputation as a surveyor/engineer and cost analyst; and Stone was thoroughly acquainted with bridge building.

Friendship and kinship helped bring Harris and Boody together as business associates. Kinship also played an important role in Stone’s relationship to the firm. Through him the partners made contact with his brother-in-law, William Howe, and acquired the exclusive rights to Howe’s patented wooden truss bridge for the New England region and all of Canada. With the Howe patent under their control, Boody, Stone & Harris moved expeditiously to contract for all sorts of railroad and bridge projects during the late 1840s and early 1850s.59 Boody, Stone & Harris’ largest contract was for build­ing the New Haven, Hartford & Springfield Railroad. In fact, it appears that the three men initially estab­lished their partnership in order to bid on the New Haven, Hartford & Springfield contract. When the project ended in 1846, Boody left the firm and moved west to take on the construction of other railroads such as the Rochester, Lockport & Niagara Falls Railroad and the Toledo, Wabash & Western Railroad. Be­cause of his family ties with Harris, he continued to keep in close touch with his friends in Springfield but he never returned to the area. Stone & Harris contin­ued the business, specializing in bridges, passenger terminals, and roundhouses built on the Howe truss principal.60 The pattern of shifting business partnerships and temporary business associations continued for the rest of Harris’ career and indeed seems quite common to the business history of the period. In 1850, for ex­ample, Amasa Stone decided to leave the firm and moved to Cleveland, Ohio, where he subsequently amassed a fortune in the railroad construction and banking businesses. Although he had numerous op­portunities to move west as well, Harris decided to remain in Springfield because he had strong New England attachments and felt that it was a good place to raise his family. With Stone’s departure, Harris as­sumed control of the Howe truss rights for New England and made them the focal point of his business. Between 1850 and 1876, Harris entered into at least four different 58 Ibid. p. 181.

59 The History of Local Bridges and of the Bridge Industry,” June 11, 1899, in John C. Newell, “Scrap Book, 1878-1896” (Springfield, MA: Genealogy & Local History Division, Springfield Public Library), p.

171; n. a., Amasa Stone (Cleveland, OH: The De Vinne Press, c. 1886), pp. 4-5.

60 Burt, Memorial, p. 183; Azariah Boody newspaper obituary notices, D. L. Harris Papers, Springfield (MA) Public Library [hereafter abbreviated SPL].

Merrit Roe Smith

partnerships with four different people. He doubtlessly had other partners since a number of his business associations extended to special contracts for specific projects. His lengthiest partnership, four­teen years, was with William Biraie, a Scottish-born stone mason and friend of both Azariah Boody and Amasa Stone, who joined the firm in 1848 and exe­cuted all the masonry work on which Harris’ bridge-work rested. During Birnie’s tenure the firm began to expand its market by taking on bridge contracts in Pennsylvania, New York, Maryland, and Virginia, including nearly all the bridges on the Richmond & Danville Railroad, the Providence & Stonington Rail­road, and the Harlem Railroad. In all these undertak­ings, Harris & Birnie followed the common practice of hiring supervisors and work crews for specific proj­ects and putting out the work for various bridge com­ponents to local foundries and machine shops. Once a project ended, the crews either moved on to another job or were temporarily laid off until a new one began. Such practices were common in the building trades throughout the 19th century and contributed to a lot of moving about on the part of construction workers. They also relieved Harris & Birnie of having to pay for keeping a permanent labor force.61 The bridge building business flourished during the late 1840s and early 1850s. In 1851, for example, Harris and his associates had forty four bridges under con­struction with equally good prospects for 1852. Partly to supply his own needs and partly to service a rapidly growing market for bridge components, Harris had expanded his business by building a foundry and machine shop in 1849. Conveniently located on a siding of the Western Railroad in Springfield, the “Agawam Foundry” produced a wide variety of cast­ings for bridges, railroads, and machinery. Harris’s neighboring Springfield Machine Shop proved equally versatile, offering a general line of machine tools (particularly bolt cutters for bridge work and large engine lathes, drills, and planers for railroad shops), steam pumps, and turn tables as well as “bolts, & bridge & machine work generally.” Initially Harris equipped his shops with tools and machinery obtained from such firms as the Ames Manufacturing Company in nearby Chicopee and the more distant shop of Samuel Flagg & Company in Worcester. The business grew from 19 employees in 1849 to 32 in 1852. By the latter date the shops were supplying bridge compo­nents and machinery to customers (primarily rail­roads) throughout the northeast and, thanks to Amasa Stone’s continued influence, the midwest. As the shops acquired a reputation for good work, they also became a stopping point for railroaders and other visitors who frequently toured the New England re­gion in search of “improvements.” In 1854 the local 61 Amasa Stone, pp. 6-8; Burt, Memorial, p. 183; Business correspondence, 1852-68, D. L. Harris Papers, SPL; Stone & Harris Accounts, 1851-54, Stone, Harris & Birnie Payrolls, 1850-52, and Bills for Labor, 1848-52, Harris Papers, HBS; Biographical Review: The Leading Citizens of Hampden County Massachusetts (Boston: Biographical Review Publishing Co., 1895), p. 23. On Birnie, see Biographical Review, pp. 22-24; Springfield Republican. Dec. 2,1889; Stone & Harris Bills, 1848-51, Harris Papers, HBS.

• 331

Railroads in historical context: construction, costs and consequences

correspondent for R. G. Dun reported that Harris and his partners had “grown rich in the railroad business.” Harris, he noted, “is worth $100,000 or more & con­stantly making money, & living prudently.”62 Harris’ nineteenth century biographer, Henry M. Burt, observed that “Bridge building gave Mr. Harris a good start in business, but it was his long association with the Connecticut River Railroad, as its President (1855-79) that gave him his greatest reputation.” “This was the crowning achievement of his life,” Burt continued, and few men so completely mastered the subject of railroad construction and man­agement from beginning to end. He not only understood the theory of construction, op­eration and management, but he had passed through so many different stages of practical experience, that he became a complete mas­ter of his business.63

332 •

Since Harris equated engineering with “the practical, every-day affairs of business,” it seemed obvious that he eventually would become a railroad executive. His friends and former partners Amasa Stone and Azariah Boody had done so. Harris seemed to be waiting only for the right opportunity to move up the ladder of preferment and power. That opportunity arrived early in 1855 when Chester W. Chapin, a Springfield banker and president of the Western Railroad, asked Harris to succeed him as president of the Connecticut River line.64 The Connecticut River Railroad was ten years old in 1855, and in trouble.

Originally built to connect the up-river towns of Hadley, Northampton, and Green­field with the Western Railroad and routes south at Springfield, the line’s stock had been paying small dividends and its value had fallen from $100 to a little above $40 a share. Harris’ charge was to put the company in the black, and he did so by initially curtailing service, laying off employees, standardiz­ing equipment, and moving the line’s repair shops from Northampton to Springfield.

Since a number of the railroad’s largest stockholders came from Northamp­ton, the new policies aroused bitter resentment and even resulted in a court suit aimed at stopping them. Harris nonetheless stood his ground and the business slowly improved. By 1860 the company began to return eight percent profits to the stockholders as well as expand service northward into central Vermont and western New Hampshire. Such an impressive per­formance served to quiet his critics and allowed him to consolidate his power. It also helped to insure his rep62 Payrolls, 1848-54 (vols. 7-8), Accounts, 1851-54 (vol. 4), Bills, 1852-54 (vol. 5), Springfield Too] Co. folder, 1864 (vol. 9), Harris Papers, HBS. Harris to E. B. Philips, March 6, April 23,1860, Harris & Briggs Letter Book(vol. l,pp. 525,604), Harris Papers.HBS; Dun and Bradstreet Records, vol. 41, pp. 90,156,173, 179,191, 218, 221, HBS. 63 Burt, Memorial, p. 185.

64 On Harris’ reputation as “an old fashioned businessman” see Ibid., pp. 37,42, 49,178, 232, 242.

Merrit Roe Smith

utation as one of the most talented early railroad executives in New England.65 Owing to his reputation as a successful businessman, Harris received numerous requests for his services. In 1859, as alluded to earlier, he traveled to Russia to inspect and report on the condition of the Howe truss bridges on the Moscow & St. Petersburg Railroad.66 On another occasion, he declined the presidency of the Southern Railroad, yet served on the boards of direc­tors of the Vermont Valley, the Ashuelot, and the Union Pacific railroads. His most important position was as secretary and manager of the Eastern Railroad Association, an organization established in 1867 to pool patents among members and resist threatening lawsuits by outsiders. Beyond the railroad commu­nity, Harris also served on the boards of the Spring­field Fire & Marine Insurance Company, the Chapin Banking & Trust Company of Springfield, the Hol-yoke Water Power Company, and the Crocker (Paper) Manufacturing Company of Holyoke. During his earlier years as assistant engineer on the Troy & Sch­enectady Railroad, he had spoken disparagingly of “capitalists” and their narrow economic interests. Indeed, he wrote his father in 1840, “Their aim is to make a road in the shortest time, and for the least money, with scarcely a thought for safety, utility or durability.”

Those opinions had long since passed. The table had turned. Now he was a capitalist and doubtless his interests and priorities had shifted accordingly.67 By the early 1860s, Harris had become one of Spring­field’s richest and most influential citizens. Estimates of his worth at the time ranged from $150,000 to $500,000, most of which came from three sources: his salary as president of the Connecticut River Railroad, profits from his bridge building and affiliated busi­ nesses, and income from investments in railroad stocks often acquired inpayment for materials his companies sold to various railroad clients. With wealth came social standing. Not surprisingly, Harris’ circle of friends included the leading families of Springfield and the surrounding region. Among his closest friends numbered his old partners William Birnie and A. D. Briggs, the local railroad magnate Chester Chapin, Chief Justice Chapman of the Massachusetts Supreme Court, Springfield Republican publisher Samuel Bowles, and the Reverend Samuel G. Buckingham, pastor of the South Congregational Church of which Harris was a member. Together these men viewed themselves as the rep65 Ibid., pp. 185-90; “The Connecticut River Railroad,” September 10, 1916, a newspaper clipping in the Harris Papers, SPL; Moses King, ed., King’s Handbook of Springfield Massachusetts (Springfield: James D. Gill, Publisher, 1884), pp. 89-91.

66 During his five-month visit to Europe, five weeks of which was spent in Russia, Harris kept a diary in which he recorded his impressions of the “flawless” precision that characterized the St. Petersburg & Moscow Railroad. At the same time he spoke disparagingly of the “slick and greasy looking priests” he saw everywhere and leveled bitter criticism against the religious superstitions and filthy living conditions that pervaded societies along his route. See Harris diary (footnote 1 above); also Burt, Memorial, pp. 197227; Harris to Father, June 10, 1859, Harris Papers, CVHM. For Harris’ report on the bridges of the St. Petersburg & Moscow Railroad, see Harris to Messrs. Winans, Harrison & Winans, May 28, 1859, Harris Papers, CVHM. 67 Burt, Memorial, pp. 24, 59-62,192, 278, 342.

• 333

Railroads in historical context: construction, costs and consequences

resentatives of the established order in Spring­field.68 In politics Harris was a staunch Republican. He had been raised in a Whig household and on more than one occasion he had spoken out against the “rich, lazy, smoking, drinking, bragging Loco Focos” within the Democratic Party.

As a young man he freely con­fessed that he harbored no “abolition sentiments.”

Indeed, he wrote his brother in 1841, The more I ponder the subject the more I am bewildered. I abhor slavery, but I cannot bear the radicalism of those who preach im -mediate emancipation.

334 •

By the early 1850s, for no apparent reason, his posi­tion on abolitionism changed. In addition to support­ing temperance, Sabbatarianism, and “all the great moral reforms of the day,” he became deeply involved in the affairs of the Kansas Emigrant Aid Association and made his Springfield office available as a depot for arms shipments to John Brown and his followers in Kansas.69 Harris held a number of elective offices during his career. His baptism in local politics came in 1854 when he was elected to the Springfield City Council, a post he returned to three times during the 1870s. From the city council he went on to serve two terms as a member of the Board of Alderman (1859,1863), one term as mayor of Springfield (1860), and five terms as a member of the Massachusetts state legislature. Even when he was out of office, people considered him one of the powers behind the throne in Springfield politics. Much to his unease even friends playfully referred to him as “Boss Reformer.”70 As the years wore on, “retrenchment, reform and reduced taxes” became Harris’ political trademark. As a fiscal conservative, he frequently opposed munici­pal and state projects because they portended “too lavish expenditure of public money.” Indeed, his vehement stand against state support for building the Hoosac Tunnel in northwestern Massachusetts be­came a cause celebre during the 1860s and 1870s. In this case, however, Harris had more than public fiscal concerns in mind. As president of the Connecticut River Railroad, he well knew that the construction of a tunnel would promote the interests of a competing railroad to the north (the Troy & Greenfield) and siphon off business from his line.71 Harris lost that battle, but he won others. Nonetheless, friends felt that he “was not a success as a politician or as a legislator.” “He could never practice the 68 Dun and Bradstreet Records, vol. 41, pp. 156, 218, 221; King’s Handbook of Springfield, pp. 188-90; Letter Book, vol. I (Stone & Harris Papers), pp. 323, 347,374, Harris Papers, HBS; Harris to Cammann & Co., Oct. 5,1859, Harris & Briggs Letterbook, vol. 1, pp. 323-24, HBS. 69 Burt, Memorial, pp. 27-28,35,284,293; Biographical Review: The Leading Citizens of Hampden County, p.

10; “Daniel L. Harris.” Springfield Republican. July 12,1879.

70 Burt, Memorial, p. 194; “Daniel L. Harris,” Springfield Republican. July 12,1879; Unidentified newspaper clipping dated July 12,1879, Harris Collection, SPL. 71 Burt, Memorial, pp. 34, 37; Dun and Bradstreet Records, vol. 41, p. 218, HBS. For correspondence and materials on the Hoosac Tunnel and the Troy & Greenfield Railroad, see D. L. Harris Papers, vol. 20, HBS.

Merrit Roe Smith

arts of the politician,” one supporter observed, because he tended to antagonize people with his blunt manner and out­spoken views. More than one contemporary realized that “he was of too positive and unyielding character and scorned too heartily the arts of a politician, to secure influence at Boston.” Indeed, a writer for a local newspaper wittily remarked, “it was often said that if any one wished to carry a measure in the House, he had only to get Mr. Harris to oppose it.”72 Harris devoted himself so completely to business and political affairs that he had little time for other activi­ties. Two exceptions, however, are worth noting: his membership in the South Congregational Church and his affiliation with the Springfield City Library Asso­ciation. Of the first little needs to be said other than that Harris maintained an active association with the Congregational sect ever since his conversion at Wesleyan and, through it, became an ardent advocate of temperance and sabbath keeping. The second occupied a place close to his heart and represented his only involvement in what could be called philan­thropy.

Harris’ interest in establishing a city library grew out of a paternalistic belief that it would provide young people unable to afford “higher branches” of study an opportunity to better themselves. Such an institution, he believed, would contribute significantly toward making them “good and useful citizens.” I do not want a City Library for myself or my family,” he remarked. / have plenty of books at home; but I want to make a goodplace for young men, where they can spend their evenings in improving their minds -just such a place as I longed for when I was a boy.

Harris contributed $10,500 to the library fund, chaired the committee that oversaw the building’s construc­tion, and personally raised an additional $25,000 to pay off the institution’s early debts. Such a large persona] commitment certainly attests to Harris’ al­t ruism, but it also indicates the extent to which he was willing to go to safeguard his vision of a well-ordered society by erecting new institutions of social im­provement and control.73 When Harris died in 1879, those who eulogized him tended to portray him in a heroic mode and thus foster a legend equal to any that existed for 19th century America’s more famous captains of industry. In his funeral oration Harris’ pastor and friend, Rev. Samuel Buckingham, epitomized this adulatory spirit when he characterized Harris as “a man of the strictest integ­rity,” a faithful steward of the Lord who achieved “success of the noblest kind...that can come of legiti­ mate business -wealth, usefulness and reputation-” not tarnished “in the least by any dishonesty.” Buck­ingham’s flowery sermon included all the encomiums appropriate to a person of high standing in the com­munity: his Puritan descent, his settled principles, his industriousness, his sense of duty, his courage as a 72 Burt, Memorial, pp. 34,42; unidentified newspaper clipping dated July 12,1879, Harris Collection, SPL. 73 Burt, Memorial, pp. 31, 59-60, 195-97, 224-25. The quote is from p. 195.

• 335

Railroads in historical context: construction, costs and consequences

336 •

Christian reformer, his moral purity, and so on.74 Yet one must ask to what extent these and other remarks by the rich and well-born reflected the feelings of the entire community? Clearly there is evidence that not everyone felt warmly toward D. L. Harris. His nu­merous political enemies attest to that. Even his friends acknowledged that he was a blunt, no-non­sense person who had an irascible temperament and often used extreme methods to accomplish his ends. One thus cannot help but wonder whether those who worked for Harris harbored different sentiments about him than his friends.

For the most part, the record is silent about Harris’ relations with his employees, but what evidence exists suggests tension if not open conflict. According to an associate from the Connecticut River Railroad, Harris developed “a strange prejudice against amusements and holidays for the people, as if they interfered odiously with the proper attendance upon business duties.” Such an attitude did not endear him with employees. He could not tolerate wastefulness and frequently chastised workers about their “careless, wasteful spirit.” His insistence upon system and regularity in all things also resulted in sharp reprimands to those who failed to meet his standards of workmanship. Whether these confrontations ever resulted in strikes, walkouts, or other group expres­sions of discontent is not known. On at least one occasion, however, Harris faced public wrath when a mob, led by local firemen upset by his budget-cutting practices as a city councilman, conducted a “riotous demonstration” in front of his house on the eve of July 4, 1876. What started out as a “midnight horning party,” according to one newspaper account, ended up destroying part of Harris’ property. Yet the local police stood by and, much to Harris’ chagrin, refrained from arresting the perpetrators. Harris was shaken by the experience but refused to bow to the crowd’s demands. He received further threats in 1877 and 1879, but nothing came of them. He was already suffering from a “dyspeptic disease” (probably stom­ach cancer) and his biographer suggests that the “apprehension of a second attack upon his property” coupled with his extreme agitation about the great railway strike of 1877 hastened his demise.75 Apprehensive that “the general unsettled condition of our country” threatened to undo it, Harris went to his grave a strong proponent of social order. He excori­ ated union activity as a menace to that order and did everything in his power to discourage workers from forming labor organizations on the Connecticut River Railroad and in his bridge and machine works. What­ever one may think of his personality, politics, or behavior, Harris’ career well illustrates one of the main paths early 19th century Americans followed in be­coming engineers.

74 Ibid., pp. 11-20.

75 Ibid.. pp. 38, 43, 104, 115, 238-39; Springfield Telegram. July 9, 1876. Thanks are due to Guy McLain of the Springfield Public Library for pointing out the last citation. For informative discussions of mobbing as a form of controlled protest in 19th-centruy America, see Sean Wilentz, Chants Democratic: New York City and the Rise of the American Working Class. 1788-1850 (New YOrk: Oxford University Press, 1984), pp. 64-67, 71; and Philip Scranton, Proprietary Capitalism: The Textile Manufacture at Philadelphia. 1800-1885 (New York: Cambridge University Press, 1983), pp. 384-90.

Merrit Roe Smith

EARLY ENGINEERING CULTURE What do the preceding biographical sketches reveal about the emergence of early engineering culture in America? Did George Whistler, Joseph Harrison, and Daniel Harris share a common viewpoint about their work as engineers?

Drawing on their experiences, it is possible to discern the emergence of an engineering ethos in the United States prior to professionalization? While evidence gleaned primarily from the lives of three men cannot provide definitive answers to these questions, it nonetheless is suggestive of important trends underway long before the Civil War.76 Whistler, Harrison, and Harris lived in the midst of what is often referred to as the “Great Transformation” to modernity. Although they probably never heard of such concepts as the “Industrial Revolution,” industri­alization, or the “rise of industrial capitalism,” their papers clearly reveal an awareness of living in an age of extraordinary change. Like many of their contem­poraries, they ascribed to a progressive view of history and readily attributed much of the “progress of the age” to the advent of radically new mechanical technolo­gies. Indeed, their enthusiasm for progress through technology is so strong that it must be considered a primary tenet of an emerging engineering ideology. Contrasted with the older Enlightenment view of prog­ress, which considered technology as a means of improving all the conditions of life (material, social, moral), the new view tended increasingly to construe technology as an autonomous force and end in 76 In preparing this section I have drawn on the following sources for purposes of corroboration and comparison: The James H. Burton Papers, Yale University Library; LamontDu Pont Papers, Hagley Library; J. Morton Poole Papers, Historical Society of Delaware; Calhoun, American Civil Engineer; Calvert, Mechanical Engineer in America; Merritt, Engineering in American Society; Sinclair, Centennial History of ASME; McMahon, Making a Profession; Anthony F. C. Wallace, Rockdale (New York: Alfred A. Knopf, 1978); Eugene F. Ferguson, ed., Early Engineering Reminiscences (1815-40) of George Escol Sellers. U. S. National Museum Bulletin 238 (Washington: Smithsonian Institution, 1965); Neal Fitz Simons, ed., The Reminiscences of John B. Jervis. Engineer of Old Croton (Syracuse, NY: Syracuse University Press, 1971); George A. Morison, George S. Morison. 1842-1903: A Memoir (Peterborough, NH: Peterborough Historical Society, 1940); Bruce Sinclair, Philadelphia’s Philosopher Mechanics (Baltimore: Johns Hopkins University Press, 1974); Elt-ing E. Morison, From Know-How to Nowhere (New York: Basic Books, 1974); Noble, America by Design; Edwin T. Layton, Revolt of the Engineers (Cleveland, OH: Case Western Reserve University Press, 1971); Layton, “Millrights and Engineers, Science, Social Roles, and the Evolution of the Turbine in America,” in Krohn, E. Layton, and P. Weingart, eds., The Dynamics of Science and Technology (Dordrecht, Holland: Reidel, 1978), pp. 61-87; William Sloan, Benjamin Franklin Isherwood. Naval Engineer (Annapolis: Naval Institute Press, 1965); Louis C. Hunter, A History of Industrial Power in the United States. 1780-1830. vol. I, Waterpower in the Century of the Steam Engine (Charlottesville: University Press of Virginia, 1979); also vol. 2 on Steam Power (see note 42); Duncan E. Hay, “Building ‘The New City on the Merrimack’: The Essex Company and Its Role in the Creation of Lawrence, Massachusetts” (Ph.D. diss., University of Delaware, 1986). Terry Shinn provides an excellent overview of early French engineering in “From ‘Corps’ to ‘Profession’: The Emergence and Definition of Industrial Engineering in Modern France,” in R. Fox and G. Weisz, eds.. The Organization of Science and Technology in France. 1808-1914 (Maison des Sciences de l’Homme and Cambridge University Press, 1980), pp. 1833-208. For a perceptive treatment of the British scene, see R. A. Buchanan, “Gentlemen Engineers.” Also see Edwin T. Layton, Jr., “European Origins of the American Engineering Style of the Nineteenth Century,” in N. Reingold and M. Rothenberg, eds., Scientific Colonialism (Washington: Smithsonian Press, 1987), pp. 151-166.

• 337

Railroads in historical context: construction, costs and consequences

338 •

itself. To be sure, engineers like Whistler, Harrison and Harris were not the only ones to adopt this perspective. Joining them, for instance, were politicians like Daniel Webster of Massachusetts and merchant-industrial­ists like Patrick Tracy Jackson of the Boston Associ­ates. Nevertheless, they became among the strongest proponents of a technocratic conception of progress because it not only enhanced their personal economic interests but also served their larger social vision of bringing order and prosperity to a wild and unruly natural world. For them the clock and the steam engine became master symbols of the new order.

Such exquisitely precise engines of change served as mod­els of what could be achieved by a machine civiliza­tion. Above all, they felt the new technologies would replace the unpredictable and irregular human ele­ments that comprised the handicraft tradition.77 It should come as no surprise that early engineers became closely associated with the “business interest” of the country. Leading engineers frequently referred to the “business of engineering” and at times even defined their work as “the science of making money for capital.” Their endeavors centered on mechanical power technologies that, in sheer cost and size, stood beyond the reach of small shopowners. In effect, this meant that most early engineers either received their training in large corporate organizations like the Erie Canal or soon became associated with such ventures. In the cases of Whistler, Harrison, and Harris, it is virtually impossible to distinguish between their busi­ness commitments and their undertakings as engi­neers. Indeed, the technologies they worked with represented a new form of capital, distinctive to in­dustrial civilization, that set them apart from the common mechanic and bound them ever closer to large business enterprise. In antebellum America becoming an engineer meant being an expectant capi­talist. It also involved a growing antagonism toward labor.78 The so-called “labor problem,” as it came to be known during the 19th-century, became a sticking point with early engineers like Whistler, Harrison, and Harris and, not surprisingly, gave rise to another facet in their emerging mode of thought and action. On more than one occasion they expressed disapproval of the “ir­regular habits” that seemed to pervade their labor forces. Such criticisms, to be sure, were not confined to engineers. Complaints about carelessness, inso­ lence, pilfery, drunkenness, absenteeism and the like could be heard from a wide variety of 19th century employers. Yet engineers became the most persistent critics of traditional labor practices — particularly those associated with the craft tradition — and ulti­mately acted in concert to do something about them.

They did not like having to rely on skilled workers who followed their own dic77 Leo Marx, “Does Improved Technology Mean Progress?” Technology Review (January 1987): 34-41,71; Merritt Roe Smith, “Technology, Industrialization, and the Idea of Progress in America,” in Kevin Byrne, ed., Responsible Science: The Impact of Technology on Society (San Francisco: Harper & Row, 1986), pp. 1-30.

78 On the business aspects of engineering, see Calvert, Mechanical Engineer in America, pp. 13,154,255 ff.; Noble, America by Design, pp. 33-49; Edwin Lay ton, “Science, Business, and the American Engineer,” in Robert Perrucci and Joel E. Gerstl, eds., The Engineer and the Social System (New York, 1969), pp. 51-72.

Merrit Roe Smith

tates about doing things and, at any moment, could halt production by making de­mands on employers. Control over the work process thus became a key issue in the emerging engineering ethos. As agents of industrial capitalism, engineers like Whistler, Harrison and Harris sought ultimately to overcome the labor problem by devising strategies aimed at diminishing worker decision making, de­mystifying craft knowledge, and eliminating work traditions that carried over from the handicraft mode of production. Implicit in their assault on the craft tradition was the picture of progressive engineers fighting backward and selfish tendencies in the American work tradition. In their struggle to wrest control of the workplace they adopted procedures that inculcated utilitarian values and emphasized rationality in all things.79 Mechanization, modern management practices, and formal codified knowledge became earmarks of the engineering approach. Enough has already been said about mechanization and management in the previous pages to indicate that engineers like Whistler, Harri­son, and Harris emphasized mechanization and man­agement because they viewed them as ways of stan­dardizing practice and achieving control over produc­tion. Since technical uniformity held the key to achieving system and order in the workplace, knowl­edge of every aspect of production loomed large in the engineering enterprise.

What really distinguished engineers from their con­temporaries in the first industrial world was their approach to knowledge. Contrasted with the craft tradition, which emphasized highly individualized cut-and-try methods, the engineering approach focused increasingly on calculation and accountability.

Cal­culation referred to the use of mathematics in analyz­ing and solving technical problems in a deductive manner while accountability referred primarily to numerical methods of monitoring, evaluating, and controlling matters once projects moved from design and planning to actual production. Until recently historians understood relatively little about the larger implications of these mathematically-based approaches to knowledge. Thanks to several scholars, we know that the engineering approach to knowledge had mili­tary roots and denoted a significant shift in power relations in early industrial communities.

Higher mathematics (algebra, geometry and eventually cal­culus) provided engineers with an effective tool in re-conceptualizing and appropriating craft knowledge to their uses. The result of this restructuring of knowl­edge meant empowerment for those who possessed the new techniques and disempowerment for those who lost their monopoly over the mysteries of craft knowledge.

79 On the persistence of traditional labor practices, see W. J. Rorabaugh, The Craft Apprentice From Franklin to the Machine Age (New York: Oxford University Press, 1986); Sean Wilentz, Chants Democratic: New York City and the Rise of the American Working Class, 1788-1850 (New York: Oxford University Press, 1984); Merritt Roe Smith Harpers Ferry Armory and the New Technology (Ithaca. NY: Cornell University Press, 1977); Michael H. Frisch and Daniel J. Walkowitz, eds., Working-Class America (Urbana: University of Illinois Press, 1983); Hugh G. J. Aitken, Taylorism at Watertown Arsenal (Cambridge: Harvard University Press, 1960); Walter Licht, Working for the Railroad (Princeton: Princeton University Press, 1983).

• 339

Railroads in historical context: construction, costs and consequences

340 •

Equally significant, the new engineering approach showed in no uncertain terms that technical knowledge could be reduced to compact but sophisticated sets of mathematical equations. As the outpour­ing of printed engineering textbooks and mechanical handbooks in the 19th century well attests, engineer­ing leaders demonstrated that technical knowledge could be controlled, sub-divided, and sold like a commodity. Codification and standardization thus stood at the center of engineering thought.80 Acquiring and displaying new forms of knowledge and, in the process, calling oneself an engineer had other important social implications. For one thing, it served to differentiate engineers from skilled artisans and provided a psychological edge as well as an added degree of authority over workers that had not existed before. Interestingly enough, the term “mechanic” began to take on a pejorative meaning in society at the very time individuals like Whistler, Harrison, and Harris began to call themselves engineers. Clearly this phenomenon is more than mere happenstance. Calling oneself an engineer while referring conde­scendingly to factory “hands,” “common mechanics,” and “engineers by instinct” aptly illustrates the new class consciousness that began to emerge with indus­trial civilization. Indeed, class consciousness seems to have been far more evident among engineers and other expectant capitalists than among workers. A career in engineering provided a means of asserting merit and superiority in an increasingly hierarchical society.

Addressing oneself as an engineer had another sig­nificant implication. It brought a degree of respect from merchants, bankers, and other more highly placed members of the business world that obtained for few craftsmen. With respect came status, something which engineers like Whistler, Harrison, and Harris were acutely aware of .at least by the 1840s. Reference has already been made to Harrison’s intense concern with his family’s honor and social standing. To a lesser extent, though at times no less forcefully, Whistler and Harris expressed similar concerns about their social status. Being an engineer meant being a gentleman, and gentility counted for much in class recognitions of 19th century America. Not surprisingly, when the earliest professional engineering societies began to take shape after the Civil War, they emphasized gentlemanly standing as a primary criterion for member­ship. Indeed, Monte Calvert points out with reference to the American Society of Mechanical Engineers that for 80 On the origins and implications of the engineering approach to knowledge, see Hoskin and Macve, “The Genesis of Accountability”; David D. Bien, “Military Education in 18th Century France: Technical NonTechnical Determinants,” in Monte D. Wright and Lawrence J. Paszek, eds., Science, Technology, and Warfare. The Proceedings of the Third Military History Symposium United States Air Force Academy, 8-9 May 1969 (Washington, D.C.: GPO, 1970), pp. 51-59. Also see Sally Hacker, “Mathematization of Engineering” in J. Rothschild, ed., Machina Ex Pea: Feminist Perspectives on Technology (New York: Pergamon Press, 1983), pp. 38-58. I am indebted to Harry Marks for pointing out the significance of the codification of engineering knowledge. A typical source is John C. Trautwine, The Civil Engineers Pocket-Book (Philadelphia: Claxton, Remsen & Haffelfinger, 1872). For more citations, see Eugene S. Ferguson, Bibliography of the History of Technology (Cambridge: The M. I. T. Press, 1968); Brooke Hindle, Technology in Early America (Chapel Hill: University of North Carolina Press, 1966).

Merrit Roe Smith

the first twenty or thirty years of its existence “the society did not function as a professional association, but primarily as a clearinghouse for practical and scientific information and an elite social club.” The elitist orientation of engineering in antebellum Amer­ica is extremely important to understanding and ex­plaining the social gradations, tensions, and conflicts that accompanied early industrialization.81 Lest one leaves this chapter with an image of Whistler, Harrison, and Harris and their “engineer” contempo­raries as enlightened individuals stepping off smartly down the road toward late 19th and early 20th-century formulas of scientific management and engineering professionalism, several caveats are in order. For one thing, early engineers like Whistler, Harrison, and Harris lived in a world populated and still largely dominated by farmers, shopowners, craftsmen, and other small producers. Large-scale enterprise of the types associated with the rise of factories and rail­roads, though growing, was still in its infancy.

Con­sequently it is important to bear in mind that much of what came to be part of the engineering ethos derived from traditional culture. Early engineers, despite their concern for social differentiation, held a good deal in common with craftsmen. Whistler and Harris fre­quently had recourse and worked closely with ma­chinists, masons, and millwrights. Harrison was so closely bound to the craft tradition that he referred to himself as a “machinist-engineer.” Indeed, John B. Jervis, the highly esteemed railroad builder and chief engineer of the Old Croton Aqueduct, initially consid­ered engineering methods (of surveying, etc.) as much a “mystery” as learning a craft. Clearly engineering and craftsmanship overlapped at many points during the antebellum period. In short, we need to temper our tendency to adopt a progress-oriented, hence linear, historical perspective by acknowledging that technol­ogy is shaped as much by social traditions as it is by visionary thinking.82 When Whistler, Harrison, and Harris identified them­selves as engineers they thought and acted as if doing so were a matter of personal choice, an option that an individual could -and had to- exercise without the kind of positive “organizational” innovations, sup­ports, or sanctions that later seemed necessary to postCivil war founders of engineering societies, profes­sional journals, and the like.

That is, persons like Whistler, Harrison and Harris were concerned about trying to find a niche in society and to define vocations for themselves in the good old Calvinist sense of “callings” and in light of problems specific to their own lives and situations. They were decidedly not looking to organize a new profession or to add a new line item to the Census Bureau’s list of recognized oc­cupations.

Whistler, Harrison and Harris viewed engineering in a highly personalized fashion because they came out of a world where people organized -and expected 81 Calvert, Mechanical Engineer in America, p. 126. Cf. Shinn, “From ‘Corps’ to ‘Profession’”; Buchanan, “Gentlemen Engineers.”

82 Fitz Simons, John B. Jervis, pp. 27-28,37. Also see Calhoun, American Civil Engineer, p. 193.

• 341

Railroads in historical context: construction, costs and consequences

342 •

to organize- their lives primarily around ties of friend­ship, kinship and other equally “local” constraints rather than in relation to functionally defined careers thought to have the same contours everywhere. Even in Whistler’s case, where the influence of military thought and bureaucracy loomed large, personal rela­ tionships played a critical role in defining his career. What is known about these individuals -to wit, the overriding importance of family connections, inter­ marriages, shop friendships, old school ties, eta-provides telling information about the type of society they inhabited. The degree of intimacy and coopera­ tion that existed within various communities and re­gions is indeed remarkable.

Put crudely, antebellum America consisted of more or less independent com­ munities where what one knew often mattered less than who one knew. What really counted in business and engineering as well as daily life was one’s friends, relatives, and neighbors.83 This last point is important historiographically be­cause it calls into question the “first new nation” vision of the United States as a society with no feudal or other pre-capitalist past.84 To depict Whistler, Harrison, Harris, and their engineer contemporaries as men ascending the road to professionalization is a grave oversimplification. What we have here are not “pre-professional” engineers looking to the future and not quite making it, but rather people very much caught up in “the world we have lost.” The past tugged at individuals like Whistler, Harrison, and Harris at the very time they initiated technological changes and struggled somewhat unwittingly to define the future. In this respect, we need perhaps to emphasize the anomaly rather than the prescience of individuals who began to call themselves engineers prior to the Civil War.

83 This point is well illustrated, for example, by Wallace, Rockdale; and Ferguson. George E. Sellers. Also see Smith, Harpers Ferry Armory; Hay, “Essex Company,” pp. 48-56,68-87.

84 See, for example, Louis Hartz, The Liberal Tradition in America (New York: Harcourt, Brace, 1955); and Seymour M. Lipset, The First New Nation (New York: Basic Books, 1963).

Ian Kerr

THE BUILDING OF THE BHOR GHAT RAILWAY INCLINE IN WESTERN INDIA IN THE MID-19TH CENTURY A CONSTRUÇÃO DO GRANDE DECLIVE DO CAMINHO DE FERRO EM BHOR GHAT, NA ÍNDIA OCIDENTAL, EM MEADOS DO SÉCULO XIX Ian Kerr, (U. Manitoba, Canada Canadá) Ian Kerr is senior scholar in the University of Manitoba (Canada) and Professorial Research Associate in the Department of History, School of Oriental and African Studies, University of London (U.K.). He published several books of reference about the construction and impact of railways in India.

Ian Kerr é Senior Scholar no Departamento de História, Universidade de Manitoba (Canada) e professor associado e investigador no Departamento de História, Escola de Estudos Orientais e Africanos, Universidade de Londres. Publicou vários livros sobre a construção e impacto dos caminhos de ferro na Índia.

Abstract Resumo

The successful construction, 1856-1863, of the Bhor Ghat railway incline to carry the southeastern line of the Great Indian Peninsula Railway (GIPR) through the precipitous Western Ghats was one of the most difficult feats of mid-19th century engineering The end result was a pivotal development in the history of Britain?s Indian Empire. The formidable 15 mile incline with a 1 in 37 gradient in its steepest area, and a reversing section to handle the most difficult stretch was preceded by arduous surveys and succeeded by considerable repair work because of landslips and collapsed structures. Indian workers in tens of thousands up to an average daily maximum of 40,000 men, women and children in 1860 and 1861 were required. The death toll (some 25,000) from diseases and accidents was enormous, and the abilities (organizational and technical) and endurance of the supervising engineers were stretched to their utmost.

O sucesso da construção do caminho de ferro de Bhor Ghat (1856-1863), com uma grande inclinação, para passagem da linha sudeste da Great Indian Peninsula Railway (GIPR) através dos precipícios de Western Ghats, foi um dos feitos mais

• 343

Railroads in historical context: construction, costs and consequences

difíceis da engenharia do século XIX. O resultado final foi um desenvolvimento fundamental na história do Império Britânico na Índia. A formidável inclinação, com 15 milhas, e um gradiente de 1 em 37 na sua área mais inclinada, e com uma secção para inversão na passagem mais difícil, foi precedida por levantamentos árduos e exigiu consideráveis trabalhos de reparação, devido aos constantes deslizamentos de terras e ao colapso de estruturas. Foram necessários dezenas de milhares de trabalhadores indianos, com uma média diária máxima de 40 mil homens, mulheres e crianças, entre 1860 e 1861. O número de mortos, quer por acidentes como por doença, foi enorme (cerca de 25 mil), e as competências (técnicas e organizacionais) e resistência dos supervisores foram levadas aos limites.

344 •

Ian Kerr

The building of the Bhor Ghat Railway Incline in western India in the mid-19th century Ian Kerr

This paper focusses on the the construction, 1856-1863, of the Bhor Ghat railway incline to provide the permanent way of the southeastern line of the Great Indian Peninsula Railway (hereafter GIPR) through the precipitous Western Ghats: a construction project that ranks among the most demanding and important accomplishments of mid-19th century civil engineering. The building of the formidable 15 mile incline with a 1 in 37 gradient in its steepest area, and a reversing section (see figure #4) to handle the most difficult stretch, was preceded by extended deliberation and arduous surveys, and succeeded by considerable repair work because of landslips and collapsed structures—indeed there is a continuing history of major engineering works on the Bhor Ghat incline through the 20th century including the elimination of the reversing station in 1928 by building three additional tunnels. Indian workers in tens of thousands were required. The death toll from diseases and accidents was enormous, and the abilities and endurance of the supervising engineers stretched to their utmost.1 Figure # 1 provides a sectional map and diagram of the railway incline.

Completion of the Bhor Ghat railway incline required demanding inputs ranging from huge requirements for labour to organizational and engineering skills of the highest order. The management of labour, and civil and mechanical engineering (technology in the broadest sense encompassing knowledge and 1 Some pages about the Bhor construction can be found in Ian J. Kerr, Building the Railways of the Raj: 1850-1900 (Delhi: Oxford University Press, 1995, paperback ed., 1997; Kerr, “Labour Control and Labour Legislation in Colonial India: A Tale of Two, Mid-Nineteenth Century Acts”, South Asia. Journal of South Asian Studies XXVII, 1 (April 2004): 7-25; Kerr, Engines of Change: The Railroads That Made India (Westport: Praeger, 2007; 2nd ed.; Hyderabad: Orient BlackSwan, 2012) but no full-scale study exists as yet.

• 345

Railroads in historical context: construction, costs and consequences

practice, tools and machines) anchored in British experience had to be transferred and adapted to a vastly different physical and cultural setting thousands of miles away. Much had to be learned from hard-gained experience on the Bhor Ghat itself. The engineers (the “transfer agents” in Headrick’s terminology) had nothing to guide them in the Indian context, and virtually nothing from elsewhere in the world.2 The consulting engineer for railways to the Government of the Bombay Presidency wrote in 1855 “there is nothing as far as I am aware of in any English Railway which can be looked upon as a parallel undertaking”.3 A late 19th century author wrote: There are probably but few travellers now daily passing up and down the magnificent Thul and Bhore ghat inclines, quietly seated in comfortable railway carriages, who can at all adequately realise the extraordinary nature of the obstacles which have been so succesfully overcome, and the great skill and daring of all those engaged— especially during the first years—in shaping and carving out of the rocky mountain sides those wide and luxurious roads on which they now so easily and securely travel.4 346 •

The conception and construction of the incline was the compelling story of death, struggle, suffering, perseverance, heroism, brutality, venality, ingenuity and eventual triumph played out on a larger-than-life scale for almost two decades. The Indians—men, women and children—must be at the centre of the story because they did most of the work (and most of the dying—perhaps 25,000 over the course of the project) but some fascinating Britons were also present. The latter included the Chief Engineer, James John Berkley (1819-62), formerly a trusted associate of Robert Stephenson, who died an early death in considerable measure because of his exertions on the Ghats; William Frederick Faviell (1822-1902), of whom it was said he treated his workpeople like dogs, who took and then abandoned the contract to build the incline; Solomon Tredwell who took over the contract after Faviell and then died from a disease acquired on his first visit to the line of works; and the remarkable Alice Tredwell, Solomon’s wife, who continued the contract and saw it through to a successful conclusion thanks to the labours of her two resident agents, the former GIPR engineers S. Adamson and G.L. Clowser.

2 Daniel R. Headrick, The Tentacles of Progress. Technology Transfer in the Age of Imperialism, 1850 – 1940 (New York: Oxford University Press, 1988), pp. 9-13.

3 India Office Library and Records, British Library, L/PWD/d/3/251, Bombay Railway Letters, #11, 19 Mar. 1855, enclosures, item 464.

4 G.W. Macgeorge, Ways and Works in India. Being an account of public works in that country from the earliest times up to the present days (Westminster: Archibald Constable and Company, 1894), p. 358. This author provides some twenty pages of descriptive material about the Bhor and Thal inclines but the Indian workers and the British engineers appear primarily as part of aggregate statements.

Ian Kerr

The Bhor project provides an intrinsically dramatic story set in a colonial situation and a harsh, socio-spatial environment. A contextualized microhistory adopting Frederick Cooper’s dictum that historians should have “a willingness to change focus back and forth from the intimacy and complexity of relationships in specific places and their connections to distant places and long-term processes of change . . .” enables one to detail the complex interactions between the human actors, their institutions, their technologies, their work processes, and the physical environment in which they worked and sought to subdue.5 In short, one can tell a compelling story and explore certain analytical issues while doing so.

Figure 1 - The Bhor Ghat Incline, sectional map and diagram

• 347

The Bhore incline was an exercise in imperial engineering, what one writer in the Engineering Magazine in 1899 referred to as “a more certain and enduring form of attack than military power, and that the railway, the canal and harbour are the real weapons in the conquest of a colony”.6 Roads, railways, 5 Frederick Cooper, “African Labor History”, in Global Labour History. A State of the Art, ed. Jan Lucassen (Bern: Peter Lang, 2006, 2nd ed. 2008), p. 94.

6 Quoted in Casper Andersen, “’The Civilizers’. Consulting Engineers, Imperialism, and Africa, 1880-1914”, Ph.D. thesis, Department of the History of Ideas, University of Aarhus, 2009, p. 39.

Railroads in historical context: construction, costs and consequences

348 •

telegraphs, bigger and better harbours, fortifications, government buildings, and hydraulic works provided the material infrastructure upon which the political and economic control of a colony was secured and exploitative possibilities enhanced. To use Daniel Headrick’s term, infrastructural projects were among the “Tools of Empire”.7 Each project, each major public work, had to be constructed via workprocesses that typically were labour-intensive in character. The demands for labour were substantial: substantial in their cumulative aggregations, impressively large in their localized demand for workers within short time periods at particular sites. Railway construction in India in the last half of the 19th century may have employed a cumulative ten million people.8 Stromquist may overstate when he writes “railroad construction workers became the preeminent migratory workers” but there is no doubt railway building figured large among the infrastructural projects of empire.9 Each project, moreover, involved technological transfers from the metropolitan power to the colony, from core to periphery. That which was transferred included concepts of what was to be built—the very idea of a dam or railway—to the practice, techniques and tools needed to do the actual building.

Most fundamentally, knowledge was transferred: knowledge that existed in a multiplicity of forms from the abstract to the specific, from the idea, the template, of what a harbour, dam or railway should be to the tools and machines that represented objectified knowledge, and onwards to the knowledge needed to use the tools and the machines; from what practice might be, to what was practiced.10 It was through these complex transfers and/or adaptations of existing African or Asian technologies utilized within particular labour processes that an imperial infrastructure project came into being.

The end result was an example of technological transfer but the labour processes that created that result also involved technological transfers—be it in the form of exogeneous means of production (e.g., a steam dredge or an electric dynamo), in the form of a new standard for the quality of an earthen embankment, or a new grouting technique to seal a dam. Equally if not more important was the transferral of managerial technologies—“bodies of techniques, structures and This fascinating thesis is now a book: British Engineers and Africa, 1875–1914 (London: Pickering & Chatto, 2011).

7 Daniel R. Headrick, The Tools of Empire. Technology and European Imperialism in the Nineteenth Century (New York: Oxford University Press, 1981); also Headrick Power over Peoples: Technology, Environments, and Western Imperialism, 1400 to the Present (Princeton: Princeton University Press, 2009) and his The Tentacles of Progress.

8 Kerr, Building the Railways of the Raj, esp. pp. 196-226.

9 Shelton Stromquist, “Railroad Labor and the Global Economy: Historical Patterns”, in Global Labour History. A State of the Art, ed. J. Lucassen (Bern: Peter Lang, 2006, 2nd ed. 2008), p. 633.

10 I draw inspiration from the ideas presented in Raymond Williams, “Communications Technologies and Social Institutions”, in Raymond Williams, ed., Contact, Human Commuication and Its History (London: Thames and Hudson, 1981), p. 227.

Ian Kerr

principles”— literally embodied in the European supervisory transfer experts who used them to provide project management.11 Marx would have labelled the physical results “the power of knowledge, objectified” although he also would have recognized the socio-institutional framework that made the physical systems work.12 Technology is social and material. The relationship is continually recursive. But that, surely, is exactly that with which studies of technology transfers must deal, namely the transferral of reliable knowledge, validated techniques, skills, and devices to societies which had existing means and social relations of production; what Ian Inkster has explored as “useful and reliable knowledge” whose transference and adoption, he argues, “are better explained at the more mundane level of how”.13 Because of imperialism the infrastructural projects were imposed on areas of Africa and Asia but the labour processes through which the projects were built required adaptations between exogeneous and indigeneous forms of production, between localization and globalization because, as Peter Robb puts it, “In some senses, labour is a universal condition, but in many other ways it is plainly various; its situation in India depended on indigenous characteristics as well as colonial force majeure”.14 The Empire in the case of the Bhor Ghat incline was the British Empire, within which India was the crown jewel, but most other European powers or their settler offshoots, e.g., the United States, plus an imperially-minded Japan, had colonies such that on the eve of the First World War most of the world was controlled, directly or indirectly, by a small number of more developed states, the core states in Wallerstein’s much used vocabulary.15 The construction of the infrastructures of the British Empire, and particularly the massive consequences of that for the mobilization and utilization of Afro-Asian labour as suggested above, provides a wider context within which a detailed examination of the Bhor project can be framed.

In the rest of this paper I briefly gloss some of the highlights of the struggle to build a railway line through the Western Ghats where the railway builders (British engineers and overseers, and Indian workers) directed by colonial authorities, encountered and overcame what Hughes conceptualized as a 11 Richard Roberts, “French colonialism, imported technology, and the handicraft textile industry in the Western Sudan, 1898-1918”, Journal of Economic History, XLVII, 2, (June, 1987), pp. 461 and 471.

12 Karl Marx, Grundrisse, translated by Martin Nicolaus (London: Penguin Books), p. 706.

13 Ian Inkster, “Potentially Global: ‘Useful and Reliable Knowledge’ and Material Progress in Europe, 1474-1914’, The International History Review, XXVIII, 2 (June 2006), p. 241 et passim 14 Wim Ravesteijn, “Between Globalization and Localization. The Case of Dutch Civil Engineering in Indonesia, 1800-1950”, Comparative Technology Transfer and Society, 5:1 (April 2007), pp. 3265; Peter Robb, “Labour in India 1860-1920: Typologies, Change and Regulation”, Journal of the Royal Asiatic Society, 3rd series, 4:1 (April 1994), pp. 52-53.

15 “Core, periphery and semi-periphery”. See Immanuel Wallerstein, The Capitalist World-Economy.

Essays by Immanuel Wallerstein (Cambridge: Cambridge University Press, 1979), passim.

• 349

Railroads in historical context: construction, costs and consequences

350 •

technological frontier: “Wherever and whenever nature in her nonanimal manifestations frustrates man in the pursuit of his objectives, there exists a technological frontier. To penetrate the frontier man must develop techniques or a technology allowing him to adapt to, modify, or obliterate nature”.16 Duration, difficulty, employment totals and significance place the Bhor Ghat project among the greatest feats of 19th century engineering. Additionally, construction of the Bhor incline overlapped with the construction of the nearby, almost as difficult, Thal Ghat incline—nine miles 26 chains with a maximum gradient of 1 in 37—that carried the north-eastern branch of the GIPR through the Ghats, and for which large numbers of workers also were required. As for significance, the two railway inclines were important to the development of western India, and to the overall history of Britain’s Indian Empire.17 For example, the rapid growth of Bombay (Mumbai) in the last half of the 19th century—a development that led Bombay to earn the sobriquet “Gateway of India”—was enabled by the railways’ conquest of the Western Ghats. The railway made possible bulk shipment of commodities to and from the expanding port, and from Bombay’s similarly enabled cotton textile industry.18 Railway construction in India began in 1850. Of the two initial lines, the GIPR (the other was the East Indian Railway in Bengal) from Bombay faced by far the most difficult obstacle. Some 30 miles east of Bombay City the precipitous escarpment of the Western Ghats presented what some had considered to be an insurmountable obstacle to any attempt to establish a rail link directly northeast or south-east from that city to other parts of India. Although the crest of the Ghats was only some 2000 feet above the narrow littoral of India’s west coast the ascent was abrupt and devoid of gentle passes that could facilitate the construction of a railway. Figures #2 and #3 provide a cartographic depiction of the routes through the Ghats and the overall development of railways in India during, 1861-1871.

After extended debate and the arduous surveys of various routes, railway inclines at the Bhore and Thal Ghats (other routes were considered) were eventually approved. Over 3000 maps, drawings and cross-sections—many based on surveys personally carried out by Chief Engineer Berkeley—were needed to guide the construction of the Bhor line, construction that began in January of 1856.

Almost eight years later (1863) the Bhor incline was completed and officially 16 Thomas Park Hughes, “A Technological Frontier: The Railway”, in The Railroad and the Space Program. An Exploration in Historical Analogy, edited by Bruce Mazlish (Cambridge: The M.I.T. Press, 1965), p. 53.

17 On the importance for western India read “Railways in Western India”, Bombay Quarterly Review, I (Jan.—Apr. 1855), pp. 281-322.

18 Frank Broeze, “The External Dynamics of Port City Morphology: Bombay 1815-1914”, in Ports and Their Hinterlands in India (1700-1950), edited by Indu Banga. (Delhi: Manohar, 1992); Meera Kosambi, Bombay in Transition: The Growth and Social Ecology of a Colonial City, 1880-1980 (Stockholm: Almqvist & Wiksell International, 1986).

Ian Kerr

Figure 2 - Railway Lines Through the Western Ghat

Figure 3 - India, Railways, 1861 and 1871

opened at a grand ceremony held at Khandalla (near the crest of the incline) on 21 April 1863. The Governor of Bombay Presidency, Sir Bartle Frere, gave a florid address in which he felt “assured, that in future ages the works of our English engineers on these Ghauts will take the place of those works of their demigods, the great Cave Temples of western India, which have so long, to the simple inhabitants of these lands, been the type of superhuman strength, and of more than mortal constructive skill”.19 Perhaps Frere’s words were not all that far of the mark because the completed incline included 25 tunnels, eight arched masonry viaducts, the cutting of 54 million cubic feet of hard rock, and the 19 Frere’s speech is reprinted in Ian J. Kerr, ed., Railways in Modern India (“Oxford in India Readings. Themes in Indian History”). (New Delhi: Oxford University Press, 2001; paperback ed., 2005), pp. 69-76.

• 351

Railroads in historical context: construction, costs and consequences

352 •

embanking of 67.5 million cubic feet of material at a cost of some £1,100,000 (over £70,000 per mile). Some embankments were over sixty feet high with slopes exceeding three hundred feet. The work force numbered some 10,000 in 1856, surpassed 20,000 in early 1857, and peaked at 42,000 in January 1861.20 Average employment over the construction period was roughly 25,000, and the total number of deaths likely exceeded 25,000, although detailed records of mortality among the Indian workers was not kept. These figures are larger than those needed to build the iconic Suez Canal (built 1859-1869).21 The substantial demand for construction workers set labour in motion across an extensive part of western India, and beyond when people with particular skills were needed. The mobilization of so many people (100,000 at any one time keeping in mind the requirements for labour on the Thal Ghat construction, and for railway lines being built above and below the Ghats) required, in Berkeley’s words, “considerable trouble.” He went on to write that “the wants of the work, have, however, been supplied by unusual exertions in sending messengers in all directions, and by making advances to muccadums, or gangers, upon a promise to join the work with bodies of men at the proper season”.22 Work in the precipitous and isolated conditions of the Ghats was extremely dangerous. No footholds existed on some cliff faces, so workers had to be suspended by ropes in order to drill and blast the right-of-way. Some workers lost their hold and smashed into pieces in the ravine below that “had the effect of deterring his fellows, altogether, from working for days”.23 Another hazard was extensive blasting necessitated by the hard rock of the Ghats. Powder accidents 20 Archival record pertaining to the Bhor incline construction exist in quantity in the India Office Library and Records (British Library, notably the L/PWD/3 series of “Bombay Railway Letters and Enclosures”), the National Archives of India (New Delhi) and in the Maharashtrian State Archives (Mumbai). Bombay newspapers also provided accounts of what was happening on the Bhor Ghat. However, a good, initial understanding of the Bhor (and Thal) construction can be found by reading the following: James John Berkeley, “On Indian Railways: With a Description of the Great Indian Peninsula Railway”, Minutes of Proceedings of the Institution of Civil Engineers, vol. 19 (1859--1860), pp. 604-605; Berkley, Paper on the Thul Ghaut Railway Incline: Read at the Bombay Mechanics’ Institution in the Town Hall, on Monday, December 10, 1860 (Bombay: Education Society’s Press, 1861). Berkeley, Paper on the Bhor Ghaut Railway Incline of the Great Indian Peninsula Railway: Read at the Bombay Mechanics’ Institution in the Town Hall, on 21 December 1857, with an Appendix by A.A. West (Bombay: Education Society’s Press, 1863); R.W. Graham, “Description of the Bhore and Thul Ghat Inclines, on the Great Indian Peninsula Railway”, 1866. Institution of Civil Engineers, Archives (London), “List of Original Manuscripts and Drawings in Existence”, Mss. No. 1161.

21 The Panama Canal, built 1904-1914, required a maximum of 45,000 constructions workers. We have a splendid account of the Panama Canal construction in Julie Green, The Canal Builders: Making America’s Empire at the Panama Canal (Penguin Press, Penguin History of American Life, 2009) but the more general point is that many infrastructure projects employed large numbers, e.g., Nile Dams such as the first Aswan Dam and the Sennar Dam required ten thousand plus workers at any one time. Railway building in northern Nigeria in 1911 employed 13, 000 in 1911.

22 Berkeley, “On Indian Railways”, pp. 604-605.

23 Maharashtrian State Archives [hereafter MSA], PWD (Railways) 1859, vol.25, compilation 206, ‘Bhor Ghat Incline. Surrender by Mr. Faviell Of His Contract’, E. Swan to Faviell, 3 Nov.1858.

Ian Kerr

were frequent, and resulted in a heavy loss of life. Falling rocks, slips and cave-ins added to the toll. Many cuttings proved more difficult, dangerous and expensive than expected when “masses of detached rock embedded lying upon the sloping substratum of traprock. . .” “slipped when the stratum in which they were embedded was intersected by the cutting”.24 Regardless of the hazardous nature of the work itself, it was epidemic diseases that most ravaged the workforce. The Indian workers lived in rudimentary, difficult and often deadly conditions. Despite the heavy precipitation (150–200 inches) during the rainy season, many parts of the Ghat were waterless in the dry (main working) season. Drinking water had to be carried to the worksites.

Strong winds buffeted the workers and their flimsy, lean-to dwellings. Poor sanitation led to frequent outbreaks of cholera that mounted in intensity as the size of the workforce grew. Ten workers per day died during an outbreak that started in late 1859 and continued into January of 1860; another that flared up in April–May 1860 killed 25 percent of the European overseers and “of the natives so numerous as to be beyond accurate calculation”.25 Each major epidemic caused the workers to flee, thus slowing or even stopping construction until they returned. “Jungle fever” (likely malaria) also took its toll—when it did not kill people, it disabled them. A contractor’s agent reported in 1858 that during the eight month working season the Europeans on site were routinely disabled by sickness more than a quarter of the time.26 The conditions on the Ghat bred violence and lawlessness. Many Indian workers (but certainly not all) came from the deprived margins of Indian society—tribals and members of low or untouchable castes—overseen by a rough lot of Europeans among whom racist notions of white superiority were present. Overseer-level Europeans sometimes were non-commissioned military men who had taken their discharge in India when their service with an India-stationed regiment expired. Relationships between the few permanent inhabitants of the Ghat and the seasonal construction workers also were strained.

Workers were implicated in robberies and various expedients were suggested for their better control including appointing a mucaddum (headperson) for each fifty workers to ensure that they were in their hutments by eight in the evening.27 Unrest among the workers intensified when the initial contractor, Faviell, began to convert his paid agents into risk-accepting sub-contractors after Berkeley rejected Faviell’s request at the end of the 1856-57 working season (i.e., the first season) for better rates (i.e., more generous schedules in the contract). Faviell’s case was based on his view that labour had become more difficult and costly 24 Graham, “Description”, p. 20.

25 MSA, PWD (Railways) 1862, vol.6, compilation 317.

26 MSA, PWD (Railways) 1859, vol.25, compilation 206, Appleby to Fowler, 31 Oct. 1858. The rains halted most work except for some tunnelling.

27 MSA, PWD (Railways) 1862, vol.36, compilation 227, ‘Robberies in the Vicinity Of the Railway Works on the Ghats’.

• 353

Railroads in historical context: construction, costs and consequences

354 •

to obtain and retain thanks to railway-created demands for labour throughout western India. Berkeley’s view was that Faviell and his agents had difficulty obtaining labour because they did not pay well or promptly and because they did not provide much in the way of housing. Berkeley could point to the Thal Ghat construction where labour was more readily obtained, and where the contractors (Wythes and Jackson; the latter died in India before the contract was completed) used more capital intensive techniques. Eventually (March 1859) Faviell was permitted to terminate his contract on favourable terms with a new contract (which was more favourable) subsequently assumed by Solomon Tredwell. His unfortunate story was recounted above, as was the completion of the business end of the contract by his widow, Alice Tredwell supervised on-the-ground by two former GIPR engineers, Adamson and Clowser.

Regardless of the merits of the positions adopted by Faviell and Berkeley, the agents turned sub-contractors had a personal stake in cutting costs and squeezing the workers. Discontent among the workers intensified culminating in an affray at Khandalla in January 1859 when workers, some owed months of wages, became violent after the subcontractors tried to pay out at half the promised rate.

A Briton was killed, the colonial authorities became more alarmed particularly in a context where railway building had taken on urgent importance in the light of the mutinies and civil disturbances that had seen the British almost lose control of north India in 1857-58, and legislation was introduced quickly to provide the authorities with greater statutory authority to “empower Magistrates to decide disputes between contractors and workmenn engaged in railway and other public works”.28 Life and work on the Bhor Ghat continued to be hard, and some unrest persisted. However, there was improvement in the conditions of work, better payment of wages, the size of the workforce did reach its maximum in 1860 and 1861, and the work was completed in 1863 such that Governor Frere and other dignitaries could have their big opening ceremony on the 21st of April 1863.

Even this brief description of some of the highlights of the construction of the Bhore Ghat railway incline suggests its fascination for anyone who wants to explore the ways in which histories of imperialism/colonialism, of railways and technology transfer, and of labour can be made to intersect within a contextualized microhistory of one important infrastructural project—among many examples which can also be investigated and subsequently compared and contrasted— undertaken to advance the interests of the British Empire writ large, and the British Indian Empire specifically. Additionally, the building of this railway incline provides the basis for a compelling narrative full of accomplishment and 28 Kerr, “Labour Control and Labour Legislation” explores the text and context of this legislation, Act IX of 1860.

Ian Kerr

suffering. This historian, at least, remains committed to the idea that accurately telling a good story is one of the important reasons we write history. Something more of that story is captured in the image below in figure #4.

Figure 4 - The Reversing Station and the “Triumphant” (?) Worker

• 355

Anne McCants

FINANCING PUBLIC GOODS AND SOCIAL OVERHEAD CAPITAL: SOME HISTORICAL LESSONS O FINANCIAMENTO DE BENS PÚBLICOS E DO CAPITAL SOCIAL: ALGUMAS LIÇÕES HISTÓRICAS Anne McCants (MIT, USA EUA) Anne McCants is professor and Head in the Department of History, MIT Massachusetts Institute of Technology, Boston (U.S.A.). She has published extensively in the areas of economics of Gothic church building, wealth and income inequality, global trade networks and European consumerism, women’s work and access to credit, history of nutrition and social welfare, migration and labor market participation, historical demography.

Anne McCants é professora de História e diretora do departamento de História do MIT Massachusetts Institute of Technology, Boston (E.U.A.). Tem publicados trabalhos nas áreas da economia da construção das catedrais e igrejas góticas, desigualdades de riqueza e de rendimentos, redes comerciais globais e consumismo europeu, trabalho feminino e acesso ao crédito, história da nutrição e bem estar social, migrações e participação no Mercado de trabalho, demografia histórica.

Abstract Resumo

Public investment: what is it good for anyway? In the summer of 2011 most of the industrialized world finds itself once again in heated debate about this question, with expected answers ranging from an extreme perspective of nothing to a great deal indeed. This paper raises a number of questions about the economics of public good provision and social overhead capital investment, in particular how such investments have been financed in the past, and the historical performance of such investments for generating (or stifling) human welfare and economic growth. A number of historical examples will be reviewed (medieval church and bridge construction, 18th c. canal and road transport improvements, and 19th c. investment in railroad construction) to assess the macro-economic and social welfare implications of public investment, and the strategies employed to secure the financing for such investments in the past. Given the current world retrenchment of resources for public investment, we might wonder if there is a tendency to systematically under invest in public goods and social overhead capital? History has

• 357

Railroads in historical context: construction, costs and consequences

much to teach us here of direct relevance to the contemporary debate about public spending.

358 •

Investimento público: que bem e para quem? Esta questão é presentemente objeto de uma discussão acesa em muitos dos países industrializados do mundo, com (esperadas) respostas extremas que variam desde nada até muitíssimo. Este trabalho levanta uma série de questões acerca da economia dos bens públicos e dos investimentos em capital social adicional, em particular como é que esses investimentos foram feitos no passado, e a sua performance histórica para gerar (ou sufocar) o bem estar social e o crescimento económico.

Faz-se uma revisão de vários exemplos históricos (igrejas medievais, construção de pontes, canais do século XVIII e melhorias no transporte rodoviário, investimentos na construção de linhas de caminho de ferro no século XIX) para avaliar as implicações macroeconómicas e no bem estar social dos investimentos públicos, e as estratégias empregues para assegurar no passado o financiamento de tais investimentos. Dada a atual retração de recursos para investimentos públicos, pode-se perguntar se existe uma tendência para subinvestir sistematicamente em bens públicos e em capital social adicional. A história tem muito para nos ensinar sobre a relevância direta do debate contemporâneo sobre o investimento público.

Anne McCants

Financing Public Goods and Social Overhead Capital: Some Historical Lessons Anne McCants

Public investment: what is it good for? In the summer of 2011, most of the industrialized world found itself in heated debate about this very question. Of course, this was hardly the first occasion for such a debate; the merits, or demerits, of collective, and often coercive, public expenditure lie at the very foundations of the discipline of political economy as it emerged in the literary conversation of an elite group of intellectuals located primarily in northwestern Europe in the seventeenth and eighteenth centuries. Indeed, the political struggles that lie behind the formal working out of the rules of political economy are as old as the emergence of polities themselves. Under what circumstances, and for what purposes, is it justified, or even desirable, that collective bodies tax themselves to fund large-scale, and often long-term, projects for common use and in the name of the public good?

The acute revival of the public investment debate in the wake of the financial crisis of 2008 and the widespread resulting crises of the public fisc across nations both rich and poor brings renewed salience to the debate. So what is public investment good for anyway? The expected answers range broadly from nothing (or even worse if we consult those who assert axiomatically that public investment of necessity crowds out what would be superior private investment) to a great deal indeed. The occasion of the first Foz Tua Conference to investigate the economic, social and technological spillover effects of the building of the Tua railroad (begun 1887) seemed an especially propitious moment to revisit

• 359

Railroads in historical context: construction, costs and consequences

360 •

the broader question about the impact of public investments in infrastructure.

To this end, my contribution raises a number of questions about the economics of public goods provision and social overhead capital investment. In particular I consider how such investments have been financed in the past, and the historical performance of such investments for generating (or stifling) human welfare and economic growth. I look primarily to one important but easily misunderstood historical example (the building boom of the High Middle Ages of which the most spectacular remaining evidence is the presence of numerous Gothic Cathedrals dating from the twelfth and thirteenth centuries) in order to assess the macro-economic and social welfare implications of one form of public investment. In particular, I review the strategies employed by contemporaries to secure the financing for building investments in an age that was characterized by neither the presence of strong centralized states nor extensive markets for private building activity. Nevertheless, substantial investments in stone building projects were made by North-western Europeans in this period, the legacy of which is still evident on the landscape today nearly eight hundred years later.

Given the current world retrenchment of resources for public investment, it is not an unreasonable moment to ask, how did they do it? And why? And with what benefit? History has much to teach us here of direct relevance to the contemporary debate about public investment and the means to finance it.

As with some of the other contributions to this volume my direct focus then will not be on the Tua Railroad per se, but rather on an episode that shares (perhaps surprisingly given the distance of time and focus that separates them) some important features in common with the building of an elaborate railroad network (in Europe and increasingly around the world) beginning in the middle of the nineteenth century. The extension of that network even into the difficult and remote terrain of rural mountainous areas such as the Tua Valley is, like the towering heights of the Gothic cathedrals before them, evidence of the deep commitment of their societies to making profound investments in social overhead capital. I begin then by reviewing some of those key points of commonality.

Looking first to the issue of scale, the movement to build, or rebuild as was often the case, all of the major cathedrals located in the economically advanced regions of the High Middle Ages (beginning in the Paris Basin in the twelfth century and radiating west to Anglo-Norman England, north to the Low Countries and even Scandinavia, and eventually east and south to the central German lands and the Iberian Peninsula) was the culmination of an even broader medieval building boom. Significantly, the material of choice for the projects begun in this period was stone, both more durable and more expensive than the timber

Anne McCants

that had largely preceded it as the predominant building material. In the year 1003, Raoul Glaber, a monk from the reform Abbey of Cluny had this to say in his Historia: As the third year that followed the year one thousand drew near, there was to be seen over almost all the earth, but especially in Italy and in Gaul, great renewal of church buildings; each Christian community was driven by a spirit of rivalry to have a more glorious church than the others. It was as if the world had shaken itself, and, casting off its old garments, had dressed itself again in every part in a white robe of churches.

Not surprisingly Brother Raoul expresses his amazement in terms of the great many churches under construction. But this was not the only locus of building effort, or indeed, perhaps not even the most significant. (Unfortunately it is very difficult to assess this comprehensively as the survival rate of lay and religious buildings was so very different over time, a point that architectural historians forget to their peril.)

Broadly speaking there were four types of building in stone that merit our attention. The first is easy to overlook entirely, and that was the true rural building effort, comprised of walls, hedges, terraced vineyards, minor fortifications, and roads built or lined with stone. For these there is simply no systematic survey that would allow us to document either the absolute amount of such stone-work, nor the increase in the same at the advent of the new millennium.

The second type, by contrast, is extremely well documented. Thanks to the painstaking work of countless local historians, and the collating efforts of Eltjo Buringh and Jan Luiten van Zanden, we now have a comprehensive accounting of the establishment of new monastic houses for all of Europe from the inception of the monastic movement in the sixth century until the end of the Middle Ages.

What is readily apparent from their work is that there was an explosive growth in the number of monasteries at the start of the period under consideration here, more or less doubling in number over the eleventh century and then growing by another 63% across the twelfth century – that is from 6,343 monasteries at the end of the tenth century to 21,948 at the start of the thirteenth. While new growth largely ceased at this point, the absolute number of monasteries remained at an astonishingly high level with only modest decreases in the late Middle Ages. Each of these monastic establishments, except for the very poorest among them, would have expected to build at a minimum an abbey church and cloister in stone, and for most of them there would also have been a great many additional buildings including dormitories, kitchen and refectory, chapter

• 361

Railroads in historical context: construction, costs and consequences

362 •

house, and in the grander communities separate, sometimes even palatial, accommodations for the Abbot and his household.

The third category of building was occasioned by the rapid expansion in urban life that has been so well documented for this period. Not only did urban governments commission public buildings in which they could meet to conduct business, regulate the economic and political affairs of the community, and put up defense against enemies (so this list must include not just burger halls, but also weigh houses, mints, granaries, gate houses, walls and other fortifications), but private men of wealth and standing did so as well. Cities both large and small participated in this process, but none more so than Paris and London/ Westminster where the apparatus of an increasingly focused royal power was manifested in a plethora of buildings ranging from the spectacular to the bureaucratically mundane. Moreover, in keeping with a status befitting an urban presence, many town dwellers were eager to have their local parish church built of stone and not the more typical wood still found in the small churches that dotted the countryside. In some cases even, the urban parish church came to supersede in either size or grandeur the bishop’s church (i.e. the cathedral) that may have resided nearby in the same city.

Despite these rare occasions of parish preeminence, however, it was diocesan church building, our forth category of interest, that was typically the most spectacular example of them all. Over the course of the High Middle Ages, bishops gradually became more important than their monastic equivalents, the abbots. Increasingly the secular clergy assumed the headship of the most important clerical offices and were tapped for the most prestigious royal appointments. In England, where the anomalous situation of Cathedral Sees affiliated with monastic houses persisted from the Anglo-Saxon period, bishops worked increasingly hard at separating their seats of power from monastic entanglements, successfully for example in the case of the transfers of the Bishop of Coventry to Litchfield and from Bath/Glastonbury to Wells (since 1245 the home of the so-called Bishop of Bath and Wells). This effort proved to be spectacularly unsuccessful at Canterbury, where more than one bishop tried in vain to establish a secular chapter of canons just outside the urban limits in Hackington.1 (This failure was almost certainly a reflection of the peculiar dynamic set in place by the murder of the Archbishop Thomas Becket in 1170 in the cathedral itself, and the subsequent establishment of the most important English shrine on its premises.) Not surprisingly given the broader trend however, the most prominent examples of the new Gothic style, except for the very earliest, the Abbey Church at St. Denis, are all to be found in buildings of the urban secular clergy.

1 Some scholars speculate that Archbishop Thomas Becket had this as one of his goals, for which he would have incurred the enmity of the monks of Christ Church. See Draper, 2006, p. 14.

Anne McCants

This competition between the monastic and the secular clergies contributed to what was overall a tremendous presence of churches of all types on the medieval landscape. Jean Gimpel estimates that by the High Middle Ages in a then almost fully converted Christian Europe there was a church or chapel for every 200 inhabitants. For example, Norwich, Lincoln and York, all cities of less than 10,000 inhabitants, boasted 50, 49 and 41 churches and chapels respectively. The medieval Diocese of Chartres had 911 parish churches, not counting those in the city itself, in addition to the Bishop’s seat in the cathedral.

Consider that between 1050 and 1350 “several million tons of stone were quarried in France for the building of 80 cathedrals, five hundred large churches and some tens of thousands of parish churches. More stone was excavated in France during these three centuries than at any time in ancient Egypt, although the volume of the Great Pyramid alone is 2.5 million cubic meters.” (Gimpel, 1980, p.7). Moreover, France was not merely exceptional in this regard as is often assumed. Eric Fernie argues that in the first hundred years after the Norman Conquest, triggering as it did a massive rebuilding of Anglo-Saxon churches and a new (for the English anyway) architectural expression in the form of stone castles and keeps, the level of quarrying in France “was probably approached if not exceeded” (Fernie, 2000, p. 19).

As these figures suggest then, the church building of the High Middle Ages was not only carried out with a high density of construction, but also with individual projects attainting to a grandiose scale. Anyone who has ever entered into a still-standing Gothic cathedral dating from this period will need no convincing of this fact, despite having the jaded perception of an age well inured to the charms (or terrors) of out-sized scale. In short, church building in the Middle Ages might be thought of as the counterpart of the large-scale public works projects of the nineteenth and early twentieth centuries. More than one commentator has noted the remarkable similarity of feeling evoked by Gothic cathedrals and the grand train stations of the early railway era.

From an economic point of view, the building boom represents a major effort of investment and therefore raises the question of how the funds were mobilized, particularly for those projects that were not intended for private use.

This was, after all, a relatively poor and deeply fragmented society, one in which neither markets nor states were well developed. The strongest institution was no doubt the Church, but its resources, besides being spread over many member components, had to cover a multitude of needs: sustaining a large clergy, succoring the poor and sick, preserving learning and the arts, and providing pastoral care and oversight to the entire population. The vast extent of piously motivated

• 363

Railroads in historical context: construction, costs and consequences

364 •

building in this time and the scale and magnificence of churches, most especially in northern France and in England, is thus all the more striking. There is no doubt that the cathedrals in particular were meant to impress and awe the faithful, but the question of how it was actually managed remains unresolved.

My work aims to answer this question by reviewing the material aspects of the building boom freed from the simple piety of the folklore long associated with medieval cathedrals and other religious structures; and further, to use the Gothic moment as an especially rich example of the capacity of the much-neglected (by economists certainly) social sector to provide economic goods in a context where markets were comparatively thin and the public sector limited in its capacity to mobilize resources.

I begin with a basic fact well established by architectural historians: once it was demonstrated that churches could be made light and airy, and moreover large, such churches were desired everywhere, giving birth to one of the more spectacular episodes of fervent emulation in history. Furthermore, the technical advances in building with stone associated with the Gothic style also gave rise to significant improvements in the stability and durability of other kinds of built structures. The most important of these were obviously castles and other fortifications; but also equally important but less noted by historians, were stone bridges, a form that benefited greatly from new vaulting techniques, and often shared a similar source of impetus in the Church. All of this building had economic consequences, but also economic requirements, most importantly the refinement of a material process, the mobilization of new resources, the development of capacity for long-term financing and capital formation, and the establishment of new forms of managerial organization. Once again, there are striking parallels with railroad building in the nineteenth century.

The actual construction of cathedrals was overwhelmingly carried out by people paid to do so, notably by itinerant skilled craftsmen working in groups under the direction of a master mason. A few of these masons are known by name, and many were likely associated with particular stone quarries. The elite component of this workforce was definitely comprised of the men who actually worked the stone. Carpenters and blacksmiths, not to mention haulers, carters, and numerous low skilled or even unskilled laborers could be drawn from the local area. A number of critical points follow from these observations. First, successive building sites could learn from the experience (and most especially the difficulties and failures) of other sites, since what we might call the engineering decisions were made by men who traveled from one project to the next. Competition between projects along the lines of scale and scope was easily facilitated by this mobility as well. Third, any technological spillovers from large-scale

Anne McCants

masonry construction into other knowledge domains could be widely diffused with relative ease, again because of the geographic reach of the knowledge embodied in the masons themselves as human capital. Finally, agricultural labor rhythms did not have to be overly disrupted by even major building campaigns with high demand for labor because local men could easily shift back and forth between a building site and their fields. For example, at Westminster Abbey in 1253, a year for which we have unusually detailed records for the fabric, the payroll was halved during the month of September as compared to the summer months that preceded it. Labor had to be released for the taking in of the harvest.

Just as labor could move between occupations and projects, building a cathedral, even in those periods when there was real impetus behind the work, was never the sole or even the primary concern of the bishops or other clerical promoters. Cathedral construction competed with building other parts of the ‘holy city’ next to it (the Bishop’s palace, cloister, chapter house, etc.), with building churches and monasteries elsewhere in the diocese, with other municipal projects (bridges, towers and walls), and with the worldly struggles of bishops, chapters, and monasteries among themselves, or against kings, nobles, and communes, for power and advantage. The capacity to build was also impacted by major outside distractions, such as wars, crusades, and outbreaks of epidemic disease.

All of these enterprises required resources and funding was always problematic, often to the point of bringing work to a halt for either shorter or longer periods. In the worst cases, such cessations of building activity might extend even to centuries. Many resources that might have gone toward the building were instead devoted to ornamentation and to ritual, such as elaborate coronations, funerals and processions, or to the ongoing commemoration of the dead in chapels within the structure. Work at Reims was slowed on more than one occasion by the great expense of crowning a new king, and the cathedral at York was severely compromised by the payments required by Rome on the accession of each new Archbishop, a burden magnified by the conflict with the Archdiocese of Canterbury over primacy in England. On the other hand, the idea has been put forward that lavish church ornamentation and the acquisition of particularly efficacious, and therefore rare, saint’s relics tended to attract more donations.

The secular power and property of churchmen, including bishops, the cathedral chapter, and individual canons had a mixed impact on the process. On the one hand, lordship provided revenue in the form of rents, feudal dues, tithes, etc. that could fund construction; on the other, it often pitted the clerics against the local population and involved the various actors within the church in potentially costly disputes and conflicts not least among themselves. Similarly, the rise of

• 365

Railroads in historical context: construction, costs and consequences

366 •

towns greatly increased the available resources for building but led to endless conflicts over lordship (with its attendant fiscal returns), involving both who would rule and whether towns and townspeople might free themselves of the constraints imposed by feudal institutions. On this showing, one reason for the chronic financial difficulties is that, though an urban institution, cathedrals were funded within an agrarian system in which both wealth and social rank depended on control over land and those who worked it. By the thirteenth century, this system, which did not grant full legitimacy to commerce, mobile capital, or rank based on wealth (rather than the other way around), was already moving toward obsolescence, however protracted.

With these points in mind, it is possible to return to my starting point and ask, what was the role of the cathedral and does it qualify as social overhead capital as bridges and railways undoubtedly do? There is no question that cathedrals proclaimed the power and glory of a particular city and diocese, of its bishop, and of the Church and its God and saints in general. Given the role of the Church, this can be seen as proclaiming the civilization of Christian Europe, or as defending the primacy of the Church against competing forces such as kings and nations, or even vis-a-vis urban commerce and manufactures. On a more mundane level, the cathedral enabled and symbolized the dual role of the clergy, to be in the world and yet apart. It served as a gathering place, especially on feast days and other great occasions. The largest cathedrals could hold most of the population of a moderate size town, up to 10 000, (Gimpel, 1979, p. 7). As Ball reminds us, ‘when a service was not in progress, [the people] would meet their friends here, bring in their dogs and their hawks, arrange trysts, eat snacks, etc. The poor might even bed down for the night in the gloomy recesses. Stalls clung like limpets to the walls of the building. At Strasbourg, the mayor held office in his pew in the cathedral, meeting burghers there to conduct business.

Wine merchants, perhaps employed by the chapter itself, even sold their wares from the nave of Chartres – selling inside the church, they were exempt from the taxes imposed by the count of Blois and Chartres’ (Ball, 2008, p. 56). Cathedrals also drew visitors, in some cases large numbers of pilgrims (Santiago de Compostela, Chartres, and Canterbury to name only the most famous). They might also be seats of learning, but no more so than monasteries. This is perhaps not a lot of immediate return for the enormous effort of building them. However, if one uses a zero rate of discount, appropriate to unworldly matters, and reckons as a return their drawing power for tourists in the last two centuries, the costbenefit calculation changes dramatically.2 If the cathedrals only partly qualify as social overhead capital from an economic point of view, what of the possibility that their construction was a drain 2 A recent article in the New Yorker profiles an Italian producer of luxury goods who donates to, among other causes, restoring a monastery in a small Umbrian town. The abbot of the monastery is quoted as saying that ‘…monks take a long view of things, so if it takes fifty years to do it we’ll manage,’ Mead, 2010, p. 79.

Anne McCants

on more productive uses of capital and labor? Robert Lopez raised the point in a short and superficially modest article many years ago (Lopez, 1952). He pointed to towns that appear to have exhausted their substance in a disproportionately magnificent cathedral, while other communities grew and prospered with less extravagant churches. Among the former one may cite Beauvais, Salisbury, Bourges, and many more, while the latter include most Italian city-states (but not Pisa) and Bruges. Of course, from Roman times to ours, other forms of conspicuous investment (baths then and stadiums then and now) have proved be to white elephants. Lopez is careful to point out that the major cities, from Paris to Venice, could absorb the cost of a major cathedral with no loss of momentum. In the case of both these cities, moreover, cathedral building benefited from plunder. Venice profited from the crusades and Paris from the annexation of Normandy by the French Crown. We would add to these another possibility, namely that the long duration of construction may simply have outlasted a phase of prosperity of the city. Merchant towns tended typically to experience temporary booms, perhaps again a working of Cardwell’s Law, or in keeping with the Network System of competitive and fluid trade flows and shifting economic leadership among cities.

Indeed, Herman van der Wee has suggested a very different way of thinking about the resource demands of Gothic construction itself. Rather than viewing it as does Lopez as investment forgone for more productive uses, he concentrates instead on the contribution made by Gothic to a general increase in technical knowledge, understood as either practical or abstract or a combination of both.

He argues that ‘cathedral building concentrated resources in a sector where the few potentialities of technological progress could be actualized in the most promising way’ (van der Wee, 1984, p. 268). He seems almost to suggest that even if the buildings were used for no good social purpose, the very act of building them contributed in ways not available elsewhere in the economy of the High Middle Ages to the development of human capital. A similar argument has been made by the historian of science, David Turnbull, who argues that Gothic cathedrals were akin to laboratories with geometry as the common language employed to transfer knowledge from one place to another. Thus, they contributed to both ‘science and technology,’ and it is this combination that ‘allows ad hoc progress to constitute real progress in collective practice’ (Turnbull, 1993, pp. 321 and 332). Or in Maarten Prak’s formulation, cathedral building allowed for something akin to ‘full-scale’ experimentation (Prak, 2011, p. 395). All of this suggests, of course, that there are more than just intriguing parallels between the medieval stone building enterprise, and the funding practices and mechanisms that undergirded it, and the public investments that culminated in the expansion of nineteenth century railroad networks into technically difficult terrain and across a

• 367

Railroads in historical context: construction, costs and consequences

wide swath of geographical space. Indeed, these historical cases even have much to teach us as we struggle to fund contemporary large-scale investments in social overhead capital.

368 •

Anne McCants

BIBLIOGRAPHY Ball, Philip (2008), Universe of Stone: a Biography of Chartres Cathedral, New York: HarperCollins.

Buringh, E. and J. L. van Zanden (2009), ‘Charting the ‘Rise of the West’. Manuscript and Printed Books in Europe,’ in: Journal of Economic History, 69, 410-46.

Draper, Peter (2006), Gimpel, Jean (1983), The Cathedral Builders, Salisbury: Michael Russell.

Krauss, Henry (1979), Gold was the Mortar: The Economics of Cathedral Building, New York: Barnes and Noble.

Lopez, Robert S. (1952), ‘Economie et architecture médiévales: cela aurait-il tué ceci?’ in: ANNALES (E.S.C.), 7, no. 4, 433-438.

Prak, Maarten (2011), ‘Mega-structures of the Middle Ages: the construction of religious buildings in Europe and Asia, c.1000–1500,’ in: Journal of Global History, 6, 381–406.

Scott, Robert A. (2003), The Gothic Enterprise: A Guide to Understanding the Medieval Cathedral, Berkeley, CA: University of California Press.

Turnbull, David (1993), ‘The Ad Hoc Collective Work of Building Gothic Cathedrals with Templates, String, and Geometry,’ in: Science, Technology and Human Values, 18, no. 3, 315-40. van der Wee, Herman (1984) ‘Review of De Financiering van de Kathedraalbouw in de Middeleeuwen’ in: Journal of the Society of Architectural Historians, 43, no. 3, 267-68.

• 369

PART 4 The future of historical railroads O futuro das linhas hist贸ricas

Dominic Fontana

RAILWAYS: INDUSTRIAL AND MARITIME ARCHAEOLOGY, GEOGRAPHIC INFORMATION SYSTEMS, HISTORY AND CULTURE CAMINHOS DE FERRO: ARQUEOLOGIA MARÍTIMA E INDUSTRIAL, SISTEMAS DE INFORMAÇÃO GEOGRÁFICA, HISTÓRIA E CULTURA Dominic Fontana (U. Portsmouth, United Kingdom Reino Unido) Dominic Fontana is Senior Lecturer in Geography at the University of Portsmouth and is a Fellow of the Royal Geographical Society. As a mature student he read for a BA in Geography at Portsmouth Polytechnic and then continued his studies, taking his PhD at University of Portsmouth, during which time he also became a lecturer in Geography.

Earlier in his career Dominic worked for British Railways for five years following which, he spent five years working as archaeological photographer on the Mary Rose project. This major maritime archaeological project excavated and recovered Henry VIII’s warship from the Solent seabed in October 1982.

More recently, he has been involved in making several television documentary films about archaeology, the Mary Rose and one exploring the events of the Battle of Hastings in 1066. He has worked on projects and exhibitions at The Tower of London and Hampton Court Historic Royal Palace. In conjunction with the Mary Rose Trust, he is currently researching the July 1545 “Battle of the Solent” in which the Mary Rose was lost.

He is also working on a project with The Royal Collection to better understand two of Henry VIII’s major historical paintings.

Dominic Fontana é professor no departamento de Geografia da Universidade de Portsmouth (Reino Unido), e é membro da Royal Geographic Society. Doutorado pela Universidade de Portsmouth. Trabalhou na British Railways no inicio da sua carreira, tendo depois participado no projecto de recuperação do navio Mary Rose.

Participou na produção de diversos documentários sobre arqueologia. Atualmente está a investigar a batalha de Solent (Julho de 1545), em que o navio Mary Rose se afundou, assim como um projeto da The Royal Collection sobre as pinturas históricas de Henrique VIII.

Abstract Resumo

Railways, and their associated industrial archaeology, are an area of immense fascination to me as I began my working career as a railwayman at the age of 16.

• 373

Railroads in historical context: construction, costs and consequences

Consequently, I must be one of very few academics who have used much of the 19th century railway technology in a real world context as part of a national railway system. After the railway, I worked on the maritime archaeological project to excavate and recover the remains of Henry VIII’s warship Mary Rose and subsequently, as an academic geographer specialising in the application of GIS technology to archaeological and historical studies. This unusual combination of career experience has provided considerable insight into the complexity of data, which arises from both archaeological and historical study within a spatial, geographical context. This paper considers some of the potential range of data which may be held within a GIS and offers some suggestions for innovative approaches for the utilisation of such data within historical, archaeological, cultural and railway projects. GIS technology is considered as an enabling technology which can assist researchers by providing an exploratory “tool for thought”.

374 •

Os caminhos de ferro, e a arqueologia industrial associada, são para mim uma área de grande fascínio, pois comecei aí a minha vida profissional, com a idade de dezasseis anos. Serei por isso um dos poucos académicos que usaram a tecnologia ferroviária do século dezanove no contexto real de um sistema nacional de ferrovias.

Depois do caminho de ferro, trabalhei num projeto de arqueologia marítima, para escavar e recuperar os destroços do navio de guerra Mary Rose, do Rei Henrique VII, e depois como académico em geografia, sobre aplicações de tecnologias SIG a estudos históricos e de arqueologia. Esta combinação pouco habitual de experiencias profissionais proporcionou uma visão aprofundada sobre a complexidade dos dados resultantes quer de estudos históricos como arqueológicos com um contexto geográfico e espacial.

Este trabalho trata os vários tipos de dados que podem ser integrados num SIG e sugere algumas abordagens inovadoras para a utilização desses dados em projetos de arqueologia, caminhos de ferro, história e cultura. As tecnologias SIG são consideradas como tecnologias de apoio à investigação, uma ferramenta para exploração de ideias.

Dominic Fontana

Railways: industrial and maritime archaeology, geographic information systems, history and culture Dominic Fontana

Railways are large and complex entities, which are fundamental to the efficient functioning of modern society. They have a long past, active present and significant future. In Britain many railways are usually at least 150 years old and consequently, have a very long history but they are also very actively used as modern metropolitan or intra-urban transport systems and they are viewed as having a bright future in a world where low emissions and energy efficiency are extremely important attributes. Railways are both substantial and ephemeral, they are expressions of man’s ingenuity developed in stone, concrete and iron and steel engineering. They have created and nurtured their own culture and have dramatically affected and altered the culture and lives of all that they touch. Railways, may be measured, recorded and understood through their archaeology, history, geography, art, science, politics and economics. Railways are unlike almost any other form of human technology because they have a very substantial existence in physical terms, represented by the routeways and architectural infrastructure which are placed within a widespread geographical context, but at the same time have such historical and cultural depth, which has been recorded and expressed in many different modes and forms. Aspects of railways can be seen as networks or points, which maybe viewed collectively or individually. Railways comprise of many different elements including; architecture, permanent way, signalling, locomotives, carnages and wagons, railway staff and folklore, railway passengers and their memories, railway companies

• 375

Railroads in historical context: construction, costs and consequences

376 •

and their economic relations with their investors and customers and a whole host of other elements of their existence. Some of these are tangible and persistent others are intangible and ephemeral, some permanently located and others mobile or transient. It is undoubtedly a complex mixture of components. Understanding the past present and future of railways requires a detailed appreciation of the interaction of these many facets. Achieving this holistic integration of data is by no means a straightforward task and in order to develop a fuller understanding of railways and their contextual setting, researchers need to be innovative in the research methods which they employ. A geographical approach to such data integration has much to recommend it in such circumstances. GIS technology provides a tool set, which can assist researchers, by providing an exploratory “tool for thought” as suggested by Eastman (1992, p. 32). Specifically, Eastman writes “With experience, GIS becomes simply an extension of one’s own analytical thinking... The system has no inherent answers, only those of the analyst. It is a tool, just like statistics is a tool. It is a tool for thought.”

When used in this way, the GIS becomes a tool for the exploration, investigation and interpretation of the data and therefore, the data itself becomes the most important part of the GIS. Its availability, form and mixture determines what can be done with the GIS. The specifics of the software and hardware system are of little consequence other than they should provide the necessary facilities to enable the researchers to store, access and display the requisite forms of data. Most current GIS software has the ability to use a very wide variety of different digital data types including text, facsimile documents, video and sound recordings as well as direct access to web-based resources. It is therefore, the data itself which becomes the all-important core for a study employing a GIS approach and, where possible, this should be at as high a temporal and spatial resolution as possible. In other words, the GIS hardware and software opens up the information contained in the data for a researcher, providing the mechanism through which a researcher may explore the data. In effect the GIS becomes an electronic, digital archive, which enables rapid access to the data contained within it and enables swift display of that data based on the relative geography. The GIS provides an exploration and integration mechanism with which a researcher may tap the potential informational value of the data more efficiently and effectively than might be the case with a traditional paper based archive.

Karl Popper, when considering the nature of scientific knowledge and understanding, (Popper, 1974) suggested that knowledge is finite whilst ignorance is infinite. It was his contention that it will never be possible to arrive at the definitive answer and that there are many possible answers to an infinite vari-

Dominic Fontana

ety of questions. Researchers should, if possible, have access to un­interpreted, and preferably un-aggregated, data in order to be able to address these many, as yet, un-asked questions. Easy, speedy access to un-interpreted data will be far more beneficial to the advancement of their understanding than traditional linear narrative outputs of research, which largely present interpreted information.

However, historical research is usually presented as such a linear narrative, derived from research attempting to develop an understanding of the past through records and evidence created or preserved at or nearer the time. These records themselves were rarely originally collected with this historical research purpose as their primary aim and such data is always incomplete, usually heavily filtered and is seldom more than a cipher for the reality it is attempting to capture.

Consequently, we must be acutely aware of the Popparian view (Popper, 1974) of finite knowledge and infinite ignorance because of the problems, which will always be inherent within the data that researchers may employ. History however, and especially the history of railways, occurs within a geographical context, which is nonlinear, multi-temporal, multi-faceted, and multi-scale thereby adding considerable complexity to the situations which the researchers need to understand and appreciate. Railways are often huge geographical undertakings and placing relevant historical data within its geographical setting can greatly enhance the inherent and possibly unrealised value, of many historical records.

Essentially, viewing the historical data within its geographical contextual setting may assist the researcher to arrive at a fuller understanding of its meaning and importance. It may also provide a researcher with greater insight, thereby encouraging the serendipitous discovery of new knowledge derived from the timely and efficient combination of datasets which may not otherwise be attempted or even envisaged. GIS offer researchers a tool set with which they can explore the possibility of integrating data from diverse and disparate data sources, through their geography, enabling the recognition, enhancement and extraction of greater levels of the informational value which all data contains.

This situation is experienced in maritime archaeological projects, such as the excavation of the Mary Rose (Rule, 1982), where immensely complex information can be recovered and recorded from the excavated remains of the sunken vessel.

Experience gained through research undertaken to better understand the events surrounding the sinking of the Mary Rose in 1545 can illustrate some of the potential of GIS technology when applied to the task of complex, geographically organised, research data.

The Mary Rose was vice-flagship of Henry VIII’s fleet, and on the 19th of

• 377

Railroads in historical context: construction, costs and consequences

378 •

July 1545 she was assembled together with the English fleet, in the Solent on the central south coast of England, to oppose the landing of a French army. This had been sent to invade England by Francois 1, King of France. The English fleet consisted of about 60 ships and the French 225. The French army was 30,000 strong; the English forces about 12,000 at the most. For King Henry VIII and the English this was a national emergency and the English were significantly outnumbered, both on the land and sea. Over a period of two or three days a series of engagements took place between the French and the English, which ultimately resulted in a stalemate position. Although the English lost the Mary Rose and the French lost a Mediterranean galley, Henry retained his crown, the French departed the shores of southern England to lay siege to Boulogne, which the English had captured the previous year, and the Mary Rose settled into the muddy seabed of the Solent where she remained for over 450 years.

During the 1970s and early 1980s the Mary Rose was excavated on the seabed, using painstaking archaeological techniques, to recover as much of the data from the wreck as was possible. Over 19,000 artefacts, the remains of 179 individual men, as well as the hull of the Mary Rose herself were recovered during the project. Much of this is now on display in the historic dockyard in Portsmouth where a new, permanent, museum is now under construction and is due to open in 2012. The data recovered during the archaeological excavation represents an extremely complex dataset in which the spatial relationships between the objects and their surroundings within the ship are of tremendous importance. These can tell, not only a rich story of daily Tudor life aboard a warship, but also of life more generally within Tudor England. They also hold evidence of the process by which the Mary Rose sank. The distribution and movement of material and objects within the ship have recorded elements of the sequence of events during the actual sinking itself. However, this data is immensely complex and requires not only an understanding of what each individual item of data is, but of how it interrelates with everything else. In addition to the archaeological data collected during these sites excavations there are also a number of contemporary written accounts, which describe the events on that fateful day in July 1545. These are traditional historical documents and present linear accounts of what happened, as it was perceived to be by their writers. Geographically, we also have the physical coastline and sea areas, which constitute the contextual setting within which these events were played out and within which the Tudor technologies of shipping and warfare came together to create the story. This geographical context might be considered as the “Theatre of War” which staged the events of the battle. Further to these data we also have contemporary and near contemporary maps of the town of Portsmouth itself and a fabulous and detailed

Dominic Fontana

image of the battle, which presents the events as a series of vignettes representing individual actions that took place over a number of days.

The data, therefore, by which we may examine and explore “The Battle of the Solent” of 1545, comprises an immensely complex and diverse set of data, which needs to be viewed in an integrated way and to be interpreted as part of a whole. A traditional historical documentary or archaeological approach for studying such material, and its consequent dissemination of its research findings, often produce views which largely ignore one of the most important elements; the geographical context within which all this took place. The topography of the Solent dictated the potential initial distributions of the fleets and the movements that could be subsequently undertaken by the ships of both the French and English fleets. The timing of the tides and the specific nature of individual tidal currents within the Solent determined the speed, direction and route that any individual ship could make. The combination of these factors and therefore dictated the tactics which could be employed by the soldiers and sailors of both sides and their commanders. (Fontana and Hildred, 2011), Consequently, having a good understanding of the geographical environment and context can significantly help researchers to develop a much fuller picture of what happened from the available evidence.

Such rich and diverse data needs integrating if it is to be fully utilised and geography is key to this integration. Geography can act as the mechanism by which these diverse sources of data that can be meaningfully joined or associated. GIS technology provides the tool set. Holding the data with geographical coordinates, for use in a GIS, will make the data much more useful to both present and future researchers. This is because GIS technology makes it possible to integrate documentary, archaeological, environmental, geographical and other diverse data in a way that is both accessible and potentially meaningfully organised.

Despite the potentially efficient data handling, integration and exploration provided by the technology, Historians have not so far greatly employed GIS in their research. Traditionally, it is normal practice for them to disseminate information and research findings by use of a linear historical narrative, which, it could be argued, has stood the test of time and provided an effective means of research dissemination. And yet, the question should be posed “can a researcher convey the full complexity of relationships inherent in the detailed and diverse data gathered and examined in projects such as the Mary Rose or research into the history of railways, through the mechanism of a linear narrative?” Certainly, traditional research publication outputs can provide some of the information but

• 379

Railroads in historical context: construction, costs and consequences

it is unlikely that they will be by any means as full and complete as they might.

GIS technology can make a valuable contribution to the enhanced transmission of historical research data making it more accessible and easier to explore by future researchers and thereby possibly effect the serendipitous discovery of new and deeper understanding. GIS technology presents current researchers with more effective means of holding, accessing and exploring complex geographically located data and it also offers future researchers new and better opportunities to explore the past than might otherwise be available to them.

380 •

Dominic Fontana

REFERENCES Eastman, J. R., 1992, Idrisi User’s Guide. Clark University, Worcester, Massachusetts., p.32.

Fontana, D and Hildred, A., 2011, “The Theatre of War: geographical evidence from the Cowdray engraving and GIS”, in Hildred, A, ed. Weapons of warre, the armaments of the Mary Rose. The Mary Rose Trust, Portsmouth. Pages 867 — 882.

Popper, K., 1974, Conjectures and Refutations, fifth edition, Routledge and Keegan Paul, London.

Rule, M., 1982, The Mary Rose, the Excavation and Raising of Henry VIII’s Flagship., Conway Maritime Press, Leicester, UK. • 381

Paulo B. Lourenço

DISMANTLING AN OLD RAILTRACK: OPPORTUNITIES IN TUA VALLEY DESMANTELAMENTO DE UMA LINHA ANTIGA: OPORTUNIDADES NO VALE DO TUA Paulo B. Lourenço (U. Minho, Portugal) Paulo Lourenço, professor, Department of Civil Engineering, University of Minho, Guimarães, Portugal and Head of the Institute in Sustainability and Innovation in Structural Engineering. He is experienced in the fields of NDT, advanced experimental and numerical techniques, innovative strengthening techniques and earthquake engineering. He is specialist in structural restoration, with consultancy in several World Heritage sites, such as Cathedral of Porto, Monastery of Jeronimos (Lisbon), Castle of Guimaraes or Qutb Minar (New Delhi). He has worked in more than 50 monuments in Portugal, Spain, Italy, Brazil, Cyprus, Iran, India and Morocco. He is Editor of the “International Journal of Architectural Heritage: Conservation, Analysis and Restoration”, Editor of the Conference Series “Structural Analysis of Historical Constructions” and Coordinator of the European Erasmus Mundus Master Course on “Structural Analysis of Monuments and Historical Constructions”. Supervisor of 40 PhD students (26 concluded) and coordinator of several national and international research projects. He is author or co-author of more than 700 technical and scientific publications in the fields of masonry, timber and concrete structures.

Paulo Lourenço é professor catedrático da Escola de Engenharia da Universidade do Minho, Departamento de Engenharia Civil. Dirige o ISISE, Instituto para a Sustentabilidade e Inovação em Engenharia Estrutural.

Especialista em restauro de estruturas, tendo sido consultor em diversos casos de Património Mundial.

Contribuiu para mais de 50 casos de restauro de monumentos em vários países. Editor do “International Journal of Architectural Heritage: Conservation, Analysis and Restoration”, editor da série de conferências “Structural Analysis of Historical Constructions” e coordenador do European Erasmus Mundus Master Course sobre “Structural Analysis of Monuments and Historical Constructions”. Publicou mais de 700 artigos técnicos e científicos nas áreas de estruturas em construções de pedra, madeira e cimento.

Abstract Resumo

The Tua rail track is a notorious technical achievement for Portuguese engineering due to the harsh landscape, particularly between S. Mamede do Tua and S. Lourenço (the parts of rail track to be dismantled). Besides gathering information about the construction and the technical achievements, the present project also aims at: (a) using the built heritage to obtain information about old building techniques and possibilities for strengthening, by testing on site selected buildings; (b) demonstration of the technical achievements for the general public, by developing exhibition materials at the Memory Nucleus.

• 383

Railroads in historical context: construction, costs and consequences

In the presentation, the built infrastructure will be presented (part to be submerged, part to be demolished and part to remain), together with the test set-ups and planned tests. For the development of a technical exhibition, a survey of the bridges of the Tua Valley will be presented, together with possible displays and hands-on experiences.

384 •

A linha do Tua foi um feito de tecnologia notório pela engenharia portuguesa, devido a um relevo difícil, em especial entre Foz Tua e São Lourenço – precisamente a secção da linha a ser desmantelada antes do enchimento da albufeira da nova barragem.

Para além de coligir informação sobre as técnicas de construção, o projeto em curso pretende ainda: (a) obter informação sobre o património construído, em especial sobre as tecnologias de construção, pelo recurso a ensaios estruturais sobre algumas obras selecionadas; (b) apresentação da capacidade técnica envolvida na linha para o publico em geral, através do desenvolvimento de materiais para exposição no futuro Núcleo da Memória.

Neste trabalho apresenta-se a infraestrutura construída, parte da qual será submergida, sendo parte demolida e outra não, assim como os ensaios previstos.

São também apresentados os trabalhos em curso de definição das experiências a colocar no Núcleo e sobre as diversas pontes sobre o rio Tua.

Paulo B. Lourenรงo

Dismantling an old rail track: Opportunities in the Tua Valley Paulo B. Lourenรงo

1. INTRODUCTION The importance of ancient constructions has been for long exclusively attributed to the use of the building, meaning that successive changes were made to the building in order to fulfill its new function. At such times, lack of use would condemn the building to ruin and often, the re-use of the stones elsewhere. Presently, modern societies understand built cultural heritage as a landmark of culture and diversity, which should last forever, being the task of the current generation to deliver the heritage in good shape for the generations to come. This act of culture poses high demands to engineers because deterioration is intrinsic to life (as an example, the expected life of a modern building is fifty years).

Only during the last decades has the idea that old and ancient buildings could be conserved and reused using a conservation approach become appealing for the market. In fact, the present policy is not only to preserve, but also to make buildings and the whole historic part of the cities alive, functioning and appealing to the inhabitants and to the tourists. It is the unique atmosphere of narrow streets and historic squares that provides a meaning to the cultural heritage of city centers, the everyday reality for the local population. The value and authenticity of the building heritage cannot be based on fixed criteria, because the respect due to all cultures and authenticity of all cultural properties also requires that its physical heritage be considered within the cultural context to which it belongs.

โ ข 385

Railroads in historical context: construction, costs and consequences

Due to the effects of aggressive environment (earthquakes, soil settlements, traffic vibrations, air pollution, microclimate, etc.) and to the fact that many old buildings and historic centers were not subjected to continuous maintenance, a large part of this heritage is now affected by structural problems which menace the safety of buildings and people.

European countries have developed throughout the years a valuable experience and knowledge in the field of conservation. In recent years, large investments have been concentrated in this field, leading to impressive developments in the areas of inspection, non-destructive testing, monitoring and structural analysis of historical constructions. These developments, and recent guidelines for reuse and conservation, allow for safer, economical and more adequate remedial measures. The dismantling of the Tua valley railways due to the construction of a new dam allow multiple opportunities in this context, addressed in the present paper.

2. ICOMOS RECOMMENDATIONS 386 •

Structures in cultural heritage buildings by their very nature and history (material and assembly) present a number of challenges in conservation, diagnosis, analysis, monitoring and strengthening that limit the application of modern legal codes and building standards. Recommendations are desirable and necessary to ensure rational methods of analysis and repair methods appropriate to the cultural context.

Therefore, an international committee has prepared recommendations intended to be useful to those involved in conservation problems, ICOMOS (2001).

These recommendations contain Principles, where the basic concepts of conservation are presented, and Guidelines, where the rules and methodology that a designer should follow are discussed.

The principles entail: General criteria, Research and diagnosis, and Remedial measures and controls. A multi-disciplinary approach is obviously required, and the peculiarity of heritage structures, with their complex history, requires the organization of studies and analysis in steps that are similar to those used in medicine: anamnesis, diagnosis, therapy and controls, corresponding respectively to the condition survey, identification of the causes of damage and decay, choice of the remedial measures and control of the efficiency of the interventions. Thus, no action should be undertaken without ascertaining the likely benefit and harm to cultural heritage buildings. A full understanding of

Paulo B. Lourenço

the structural behavior and material characteristics is essential for any project in cultural heritage buildings. Diagnosis is based on qualitative and quantitative approaches. The qualitative approach is based on direct observation of the structural damage and material decay as well as historical and archaeological research, while the quantitative approach requires material and structural tests, monitoring and structural analysis. Often, the application of the same safety levels used in the design of new buildings requires excessive, if not impossible, measures. In these cases, other methods appropriately justified, may allow different approaches to safety. Therapy should address root causes rather than symptoms. Each intervention should be in proportion to the safety objectives, keeping intervention to the minimum necessary to guarantee safety and durability and with the least damage to heritage values. The choice between “traditional” and “innovative” techniques should be determined on a case-by-case basis with preference given to those that are least invasive and most compatible with heritage values, consistent with the need for safety and durability. At times, the difficulty of evaluating both the safety levels and the possible benefits of interventions may suggest “an observational method”, i.e. an incremental approach, beginning with a minimum level of intervention with the possible adoption of subsequent supplementary or corrective measures. The characteristics of materials used in conservation work (in particular new materials) and their compatibility with existing materials should be fully established. This must include long-term effects, so that undesirable side effects are avoided.

A combination of scientific and cultural knowledge and experience is indispensable for the study of cultural heritage buildings. The purpose of all studies, research and interventions is to safeguard the cultural and historical value of the building as a whole and structural engineering provides scientific support necessary to obtain this result. The evaluation of a building frequently requires a holistic approach considering the building as a whole rather than just the assessment of individual elements. The investigation of the structure requires an interdisciplinary approach that goes beyond simple technical considerations because historical research can discover phenomena involving structural behavior, while historical questions may be answered by considering structural behavior.

Knowledge of the structure requires information on its conception, on its constructional techniques, on the processes of decay and damage, on changes that have been made and finally, on its present state.

In general, the process towards the definition of remedial measures should include the following steps: (a) Acquisition of data: Information and investigation; (b) Historical, structural and architectural investigations; (c) Survey of the structure; (d) Field research and laboratory testing; (e) Monitoring, see Figure

• 387

Railroads in historical context: construction, costs and consequences

1. Diagnosis and safety evaluation of the structure are two consecutive and related stages on the basis of which the effective need for and extent of treatment measures are determined. If these stages are performed incorrectly, the resulting decisions will be arbitrary: poor judgment may result in either conservative and therefore heavy-handed conservation measures or inadequate safety levels. Any assessment of safety is seriously affected by the uncertainty attached to data (actions, resistance, deformations, etc.), models, assumptions, etc. used in the research, and by the difficulty of representing real phenomena in a precise way.

Figure 1 - Icomos Methodology Historical Investigation (documents) Survey of structure = document DATA ACQUISITION Field Research and laboratory testing Monitoring STRUCTURAL BEHAVIOR

Structural scheme: Model Material characteristics Actions

DIAGNOSIS AND SAFETY

Historical analysis Qualitative analysis Quantitative analysis Experimental analysis

388 •

REMEDIAL MEASURES

Masonry Timber Iron and steel Reinforced concrete

Explanatory Report

Execution Documents

The proposed methodology is in clear opposition with the past understanding, in which “conservation” was warranted by the powerfulness of the intervention, blind confidence existed in modern materials and technologies, a mistrust was present towards original or ancient materials and original resisting resources of the building, the value of original / ancient structure and structural principles was not recognized, the importance of previous studies was not fully recognized and significant negative experience accumulated, see Figure 2a. The modern understanding is to respect the authenticity of the structure and structural principles governing its response, to believe that conservation should lie on knowledge and understanding of the nature of the structure and real causes of possible damage or alterations, to adopt respectful interventions (minimal, non-

Paulo B. Lourenço

intrusive and reversible), to give importance to knowledge (comprising historical, material and structural aspects) and to assume that studies prior to intervention and the intervention are multidisciplinary tasks requiring the cooperation of historians, architects, engineers, physicists, among others, see Figure 2b.

Figure 2 - Two different approaches towards conservation: (a) past understanding; (b) present understanding.

• 389

(a)

(b)

3. CHARACTERIZATION OF THE BUILT HERITAGE IN THE SUBMERGED AREA The analysis of historical masonry constructions is a complex task as usually geometry data is missing, information about the inner core of the structural elements is also missing, characterization of the mechanical properties of the materials used is difficult and expensive, large variability of mechanical properties is present, due to workmanship and use of natural materials, significant changes in the core and constitution of structural elements occurred, associated with long construction periods, construction sequence is unknown, existing damage in the structure is unknown, and regulations and codes are non-applicable.

The consideration of these aspects is complex and calls for qualified analysts that combine advanced knowledge in the field and engineering reasoning, as

Railroads in historical context: construction, costs and consequences

390 •

well as a careful, humble and usually, time-consuming approach. Several methods and computational tools are available for the assessment of the mechanical behavior of historical constructions, see e.g. Lourenço (2002, 2011). The methods resort to different theories or approaches resulting in: different levels of complexity (from simple graphical methods and hand calculations to complex mathematical formulations and large systems of non-linear equations), different availability for the practitioner (from readily available in any consulting engineer office to scarcely available in a few research oriented institutions and large consulting offices), different time requirements (from a few seconds of computer time to a few days of processing) and, of course, different costs.

The first approach followed in the works already carried out in the Foz Tua project was to identify and survey the built heritage in the area that will be submerged, see Paulino (2011), followed by the characterization of the bridges in the Tua Valley, see Lopes (2011). The Tua rail track between S. Mamede do Tua and Mirandela is 54.1 km, see Figure 3a. Fifteen train stations and stops exist in this stretch, the track being dismantled between S. Mamede do Tua and Brunheda, see Figure 3b. The built heritage of the railways in this area includes steel bridges, tunnels and several train stations and trains stops, being the latter buildings of smaller dimensions. The built heritage to be submerged in the flooded area by the dam includes four strain stations / stops, twenty-five houses and 33 other assets, see Figure 5.

The assets found in the region demonstrate the resilience of local populations and the attempts to take advantage of the natural resources of the valley.

The area has scarce human influence and the number of assets affected by the construction of the dam is relatively small. Over the years, these assets were neglected and abandoned, being today heavily deteriorated. With the construction of the new dam, the assets submerged or demolished include houses, shelters, barns, kilns, mills, retaining walls and wells, besides the railway infrastructure.

The stations and stops are an important component of the Tua railways, reflecting the construction techniques used at the time. A significant part of the work of Paulino (2011) focuses on those buildings and includes the geometric survey, the survey of the construction techniques and a photographic record, including the completion of a form for each building. The Tua railway line has seven train stops and eight train stations. The survey encompassed only stations / stops in the submerged area of the Tua dam, excluding therefore the station of S. Mamede do Tua. During the collection of field data, it was possible to observe that the buildings are abandoned but the condition is usually fair, with recent rehabilitation of many of the buildings. An example of the forms prepared is given in Figure 6.

Paulo B. Lourenรงo

Figure 3 - Tua rail track: (a) map between S. Mamede and Mirandela; (b) list of stations and stops.

(a)

(b)

โ ข 391

Figure 4 - Tua rail track: (a) map between S. Mamede and Mirandela; (b) list of stations and stops.

(a)

(b)

Railroads in historical context: construction, costs and consequences

Figure 5 - Affected built heritage in the Tua valley due to the construction of the dam: (a) train stations / stops; (b) houses; (c) other assets.

(a)

392 •

(b)

(c)

The survey allowed to confirm that the height of the stations are the same and the in plan areas are very similar, with the exception of the Cachão station (with two storeys) and Mirandela (with two storeys and an in plan area five times higher than the other stations). As described in the specification and justification of the railway, given the fact that the line would serve small villages only, with some exceptions, there was no need to build stations for a large number of passengers. At that time, there were great expectations in terms of the agricultural and industrial development for Cachão, and the station was conceived for a large industrial complex, being larger than the regular station typology. The station of Mirandela was at the end of the railways. It had higher population density and was expected to serve as a traffic node for future lines, so it was also a larger station. Apart from these two exceptions, there is a common pattern for the train stations with a height of 3.7 m, a length of 12.3 m and a width of 6.1 m. The width of the doors (three of them in the main façade) is 1.2 m and a gable roof is placed on top with an interlocking clay double roman tile, called Marseille tile in Portugal.

The typical train stop is similar to the station, with the only difference that the length of the building is 6.1 m and there are only two doors in the main façade.

Paulo B. Lourenรงo

Figure 6 - Examples of the forms prepared for each of the train stations and stops.

โ ข 393

Railroads in historical context: construction, costs and consequences

With the objective of selecting the assets for future study and testing, several other assets were characterized in detail, taking into account three main parameters: in-plan area, condition and relevance of the construction for the region. It was assumed that the minimum size of the buildings to be considered for further studies was a minimum in-plan area of 15 m2, a condition of Good, in a scale considering Good, Moderate Decay, Strong Decay and Ruin, and a typology of mills, houses and barns. Finally, sixteen assets were selected and characterized with a form similar to the form used for the train stations. These assets will be tested on site aiming at obtaining the mechanical and physical response of traditional buildings in Portugal, allowing for better assessment of existing masonry buildings and their proper conservation.

4. WORKS PLANNED AHEAD

394 •

Several works are planned ahead as other engineering contributions to the Nucleus of Memory of the Tua Valley. The first contribution is to develop an asset management system for the new Tua mobility system, including a funicular, a boat and a train in the remaining part of the rail track. This management system will include an inspection and diagnosis system, with paper-free forms, and a life cycle analysis cost model, including risk based knowledge approaches.

Over the last years, the scientific community has been directing substantial efforts to the investigation and development of tools to support the management of infrastructures. The main idea behind this is to improve the existing management systems to streamline the management process, with the overall objective to obtain significant efficiency earnings. Currently, this thematic is transversal to all the types of infrastructure and management models, being designated by Asset Management. In a broad sense, Asset Management is the application of a systematic process of operation, maintenance, and upgrade of assets in a sustainable way (considering the financial, social and environmental aspects) regarding its inherent life cycle.

The need to manage an infrastructure in an effective way led to the development of computerized tools. The Federal Highway Administration (USA) has developed the software PONTIS that comprises advanced functions for inspection, conservation, and management of bridges, Cambridge Systematics (2011), among many other bridge management systems. With the continuous developments in open computing technologies, the trend is to adopt computer management systems and geocoding of assets. Inspection processes and tech-

Paulo B. Lourenço

niques applied are quite different depending on the infrastructure. Today, there are operators that are applying cutting-edge technology in their inspections, for example, laser scanning for comparing subsequent layers of the structure to detect displacements, or the use of fiber optics to detect embankment failures.

Regarding prioritization of interventions, usually no automatized systems are available. After the inspection, if an anomaly is detected and justifies an intervention, the inspection team generates a report where the required intervention is proposed. Risk management and life cycle cost analysis of infrastructures are not normally considered by operators, so there is much room for improvement and it is expected that state-of-the-art approaches can be developed for the Tua Valley transport infrastructure, see Figure 7.

Figure 7 - Tua new transportation system asset management: (a) a possible inspection form; (b) a life cycle analysis model to be developed.

• 395

(a)

(b)

Another contribution is to promote engineering and technological knowledge by creating an open lab where visitors can learn about the construction and the railways in the Tua Valley. The first ideas, currently being developed and tested, include a railway scaled model with distinctive elements of railways (e.g. tunnels, bridges and stations), detailed history and technical description of a bridge construction, interactive demonstration of a dangerous curve and a tilting train, a masonry bridge that can be built by the visitors, demonstration of how different railway equipment works and interactive software, see also Figure 8.

Railroads in historical context: construction, costs and consequences

Figure 8 - Possible ideas for an open lab on engineering and construction technologies: (a) how to build a masonry arch (Science Museum, UK and BBC); (b) how to build a bridge (Kongregate software and K’nex); (c) build your own train station (Teifoc); (d) drive your own train (NRM, UK).

(a)

396 •

(b)

(c)

(d)

5. CONCLUSIONS The project of the Tua dam, involving University of Minho, MIT and EDP will allow to: learn from the history of buildings in the valley, the rail track and the local bridges; learn from testing the built heritage that will be demolished or submerged by the dam; use the new transportation infrastructure to validate / improve Life Cycle Assessment models and promote technical contents to the general public using hands-on and computer based technologies. The present paper presents a first contribution to the project from a civil engineering perspective.

Paulo B. Lourenรงo

REFERENCES Cambridge Systematics, PONTIS Release 4.5, AASHTO: Washington D.C.. Available from http://www.camsys.com/pro_inframan_pontis.htm (2011).

ICOMOS. Recommendations for the analysis, conservation and structural restoration of architectural heritage. Available from http://www.icomos.org/en/charters-and-texts (2001).

Lopes, L.A.V., Bridges and Tua railways: History, construction and valorization MSc dissertation, University of Minho. In Portuguese (2011).

Lourenรงo, P.B., Computations of historical masonry constructions, Progress in Structural Engineering and Materials, 4(3), p. 301-319 (2002).

Lourenรงo, P.B., Mendes, N., Ramos, L.F., Oliveira, D.V., Analysis of masonry structures without box behavior, International Journal of Architectural Heritage, 5, p. 369-382 (2011).

Paulino, R.F.C.P.S., Tua railways: History, construction and survey, MSc dissertation, University of Minho. In Portuguese (2011).

โ ข 397

Michel Cotte

TWO CASE STUDIES IN HERITAGE AND VALORIZATION OF ANCIENT MOUNTAIN RAILWAYS IN FRANCE DOIS CASOS DE VALORIZAÇÃO DE PATRIMÓNIO FERROVIÁRIO DE ANTIGAS LINHAS DE MONTANHA EM FRANÇA Michel Cotte (U. Nantes, France França) Michel COTTE is emeritus professor of History of Technology at the École polytechnique of the University of Nantes (France), where he was director of the Institut de l’Homme et de la Technologie.

His historical researches initially focused upon the introduction of the Industrial Revolution in France early 19th C, especially for civil engineering and transportation equipments and industries. He studied the concept of technology transfer and adaptation to new contexts that based sometime innovations. His PhD studied the major innovations of Marc Seguin during the 1820s for suspension bridges (cable by fine wires) and railways (line of mountain railways, tubular boiler). He studied the social context of rising for new professional categories like civil engineers and private contractors in French regional context, which based a critic of the classical model of the “state managed development” of industry in France. He pursued his researches by studying the circulation of technical ideas at large scale during 19th C as base of technical and industrial initiatives.

His teaching works for education of young engineers led him to have great interest in links between past and present technologies, especially for applying modern tools of engineering as design assisted by digital graphics to re-conception of machines of the past-times. It is now an important subject of joint-researches between mechanics, history of technology and digital sciences.

In parallel to his academic activities, he worked for the ICOMOS (International Council for Monuments and Sites) from the 1990s as expert for civil engineering heritage and hydraulic works in context of the World Heritage UNESCO Convention. He is currently advisor for the evaluation of applicant sites for the World Heritage List.

Michel Cotte é professor emérito de História da Tecnologia na Escola Politécnica da Universidade de Nantes, onde foi diretor do Institut de l’Homme et de la Technologie. O seu trabalho de investigação tem-se focado nas questões do advento da revolução industrial, especialmente nos domínios da engenharia civil, dos equipamentos industriais e de transporte, onde estudou as principais inovações de Marc Seguin nos anos de 1820 em pontes suspensas (com cabos de pequena espessura) e caminhos de ferro (linhas de montanhas, caldeiras tubulares), assim como a emergência de novas categorias profissionais como os engenheiros civis e empreiteiros no contexto regional francês (o que fundamentou uma crítica ao modelo clássico de desenvolvimento liderado pelo Estado em França). Estudou a circulação em grande escala de ideias técnicas, durante o século XIX, como fundamento de iniciativas técnicas e industriais.

Na educação de novos engenheiros interessou-se pelas ligações entre as novas e antigas tecnologias, especialmente na aplicação de modernas ferramentas digitais na reconcepção de máquinas antigas, tema que é atualmente objeto de investigação multidisciplinar entre a mecânica, a história da tecnologia e as ciências digitais.

Colaborador da ICOMOS, desde os anos 1990, como especialista em património de engenharia civil e obras

• 399

Railroads in historical context: construction, costs and consequences

de hidráulica na convenção do património mundial da UNESCO. Consultor para a avaliação de pedidos de classificação de património mundial pela UNESCO.

Abstract Resumo

400 •

The two examples I intend to present have similarities but also deep differences both for history, technology and tourism management. All together are coming from the early 20th Century and aimed to help local development by modern transportation railways. They share the metric gauge, a unique track and strong slopes. They also tried to bear local completion for the national network at normal gauge trough hills and mountainous countries. However, others issues are different and sometimes opposite. That bases possibilities of comparisons.

The first one is the “Train Jaune” line in the French Pyrenees close to the Spanish border. It reaches a summit level close from 1500 m and till today it remains the highest railway line in France. Civil engineering and technical choices were ambitious from the design times: innovative line design, exceptional bridges from great engineers (Séjourné, Gisclard), and a very interesting electrical powering solution by DC supplied by a hydraulic power station from the early 20th C. It was a reference project for the Midi Co, one of the major companies in France at that time, and that started Pyrenees tourism for European bourgeoisie by modern railway and “Grand Hotel” of Fond Romeu, built by the same Midi Co. Till today and through a complex social history, Train Jaune remains a line managed by the famous French National Co: SNCF, and the owner of the line is also a French National Co: RFF. It is now a major touristic line with important heritage values both for civil engineering and for rolling stock. Cultural landscapes of the railway in mountain context are exceptional.

Touristic development projects are based upon involvement of Department (Pyrénées-Orientales) and Region Languedoc-Roussillon, trough the regional railway traffic program with SNCF. Tourism organization also involves a Regional Park centered on the railway line, for cultural and natural heritage preservation and management of the surroundings. It is ruled by Region Languedoc-Roussillon, with involvement of departmental representatives and local municipalities. Train Jaune is on the French tentative List for the UNESCO World Heritage.

The second one is the “Chemin de fer du Vivarais”. It looks like a typical local project, at the same historical period of the Third Republic, using steam power and narrow gauge. The builder and ruler of the line was the local Department (Ardèche) gathering an exceptional financial effort to promote modernity trough railway transportation among the steep and mountainous region of the Eastern Massif central.

This departmental network made links from the Rhone Valley to the Plateau by around 130 km of tracks with two sections and a junction station at Le Cheylard.

Such public line management could not run beyond the years of automobiles, at the

Michel Cotte

turn of the 60s. At that time, a deal was contracted with an association of ancient railway workers and engineers from the region loving steam-engines and experts in their technical maintenance. Touristic development started there for a limited part of this local network (30 km), ruled by this non-profitable organization. This long and prudent touristic development program had an interesting history till difficulties happened during the 2000s. In financial difficulties and facing crucial maintenance problems, the Association stopped its activities (2008-2009). Comeback of local municipalities and department is now gathering financial solutions for maintenance, and that prepares reorganization and restart for the tourism exploitation line by a private society under public contract. Heritage attributes are also very interesting for the rolling stock, one of the most important in France for metric gauge, and for natural landscapes of “Gorges du Doux”, the local river.

Os dois casos têm semelhanças, mas também diferenças importantes, sob os pontos de vista de história, tecnologia e gestão de turismo. Ambos tiveram origem no início do século XX, e contribuíram para o desenvolvimento local com meios de transporte modernos. Ambos são vias métricas, de uma só via e com declives acentuados. Também ambos tentaram completar redes nacionais de via normal, através de regiões montanhosas. Mas outras características são diferentes e por vezes mesmo opostas.

O primeiro caso é o “Train Jaune” (comboio amarelo) nos Pirinéus franceses, próximo da fronteira espanhola. Atinge um nível próximo dos 1500 metros e continua a ser, até hoje, a linha de caminho de ferro a maior altitude em França. Am obras de engenharia civil e as escolhas técnicas eram ambiciosas, desde o seu projeto: um projeto de linha inovadora, pontes excepcionais por grandes engenheiros (Séjourné, Gisclard) e uma solução muito interessante de energia eléctrica, em modo de corrente contínua, fornecida por uma estação hidráulica de produção de energia do início do século XX. Foi um projeto de referencia para a Midi Co, uma das maiores empresas da altura em França, que iniciou o turismo nos Pirinéus, para a burguesia europeia, através de um caminho de ferro moderno e do Grande Hotel de Fond Romeu, construído pela mesma Midi Co. Hoje em dia, e depois de uma complexa história social, o Train Jaune continua a ser uma linha, operada pela famosa companhia nacional francesa de caminhos de ferro (SNCF). O dono da linha é também uma empresa francesa (RFF). É hoje uma importante linha turística com importantes valores de património, quer de engenharia civil como de material circulante. As paisagens culturais do caminho de ferro no contexto da montanha são excepcionais. Os projetos de desenvolvimento turístico envolvem o Departamento dos Pirinéus Oriental e a Região do Languedoc-Roussillon, com um programa de trafego

• 401

Railroads in historical context: construction, costs and consequences

402 •

regional pela SNCF. A organização do turismo inclui um parque regional centrado sobre a linha, para preservação do património natural e gestão do ambiente. É gerido pela Região do Languedoc-Roussillon, com o envolvimento dos departamentos representativos e de municípios. O Train Jaune está na lista preliminar da UNESCO para classificação como património mundial.

O segundo caso é o caminho de ferro de Vivarais. É um típico projeto local, do mesmo período histórico (Terceira República), movido a vapor e de linha estreita. O construtor e dono da linha foi o Departamento local (Ardèche) através de um esforço financeiro excepcional para promover a modernidade através do transporte ferroviário nas regiões montanhosas do Maçico Oriental. Esta rede departamental ligava o vale de Rhone ao planalto através de cerca de 130 quilómetros de linhas com duas secções e uma estação de entroncamento em Le Cheylard. Um linha pública desta natureza não pode resistir aos avanços do automóvel, por volta dos anos 60. Nessa altura foi feito um acordo com uma associação de antigos trabalhadores ferroviários e engenheiros da região interessados em máquinas a vapor e ainda técnicos de manutenção das linhas. Iniciou-se um desenvolvimento turístico numa parte limitada desta rede local (30 km), gerido por esta associação sem fins lucrativos. Este longo e prudente programa de desenvolvimento turístico teve uma história interessante, até aparecerem dificuldades por volta do ano 2000. Com dificuldades financeiras e enfrentando problemas cruciais de manutenção, a associação cessou as suas atividades (2009). Os municípios locais e o departamento estão agora a tentar uma solução financeira para a manutenção, e a preparar uma reorganização que permita recomeçar a exploração desta linha turística por uma nova sociedade privada sob contrato público. O material circulante tem um alto valor patrimonial, entre o melhor em França com via métrica, e as paisagens locais do Georges du Doux, o rio local, são muito interessantes.

Michel Cotte

Two case studies in heritage and valorization of ancient mountain railways in France Michel Cotte

The two case studies I present in this lecture concern Train Jaune in Eastern Pyrenees Mountains and Chemin de fer du Vivarais in Eastern Massif Central. They have important similarities as major heritages for mountain railway lines in France, with original rolling stocks, but also deep differences both for technology, history and today tourism management. All together are coming from the end of the 19th Century or early 20th Century. At that time, they aimed to open up marginal mountainous regions and to increase local development by modern transportation railways. They share similar mountainous contexts leading to metric gauge, a unique track and severe slopes. They also tried to bear regional junctions through hills and mountains to complete the national network at normal gauge. However, other issues are different and sometime opposite. That bases possibilities of comparisons.

1. TRAIN JAUNE IN FRENCH MEDITERRANEAN PYRENEES (PYRENEES-ORIENTALES): THE PRICE OF RAILWAY CONTINUITY Geographical and historical contexts The Cerdagne railway line (today named “Train Jaune”) climbs up the valley of La Tet River on the North-East side of Pyrenees, from Villefranche-de-Conflent to join the large upper plateau of Cerdagne, close from the Spanish boundary. It is a trans-Pyrenean line in the geographical sense, crossing the La Perche Pass (1592 m) to reach Cerdagne region giving its water to the South side of Pyrenees.

The main summits and picks of this country are around 3000 meters. But it is not a trans-Pyrenean line in historical and political terms: the line reaches the boundary at Bourg-Madame City but it remains in French territory going close

• 403

Railroads in historical context: construction, costs and consequences

404 •

to the boundary line to reach the trans-boundary station of La Tour de Carol.

This station is probably unique in terms of railway gauge with main line from Toulouse at European normal gauge, RENFE line to Barcelona at large Spanish gauge and Train Jaune at metric gauge!

At the turn of 19th – 20th Centuries, Train Jaune project issued through a strong lobbying by local elites gathering significant national support, both by the French Government itself and by one of the major railway companies in France.

Compagnie du Midi had monopoly on the main railway lines in South-Western France, from Bordeaux to the Rhone Valley. Purposes of Train Jaune were clearly local development by railway for a mountainous region in peripheral and isolated region. It is alone able to make a permanent and sure transportation link from mountain places, up to 1000 meters, to La Tet Valley and Mediterranean seashore. The project was presented as a perfect illustration for the concept of “equality” for the different territories inside the French Republic, with high symbolic value in French political sensibility. It was nicely advocated in the last decade of 19 Century by a remarkable local deputy at the National Assembly, Emmanuel Brousse (1890-1914)1, with strong echoes in republican parties.

Indeed, national context of the 3rd Republic was very favorable to the railway line development in general. It was the period of the “third network”, a kind of favorite political tool for regional and local development2. In 1877, the Freycinet Program planed a high number of local railways aiming to join the main lines together through the hinterlands of countryside and hilly regions. This densification of the national network was promulgated by the national law of 1879.

Unfortunately, severe geographical conditions of high valley of La Tet and upper plateau of Cerdagne didn’t allow having a credible project in technical terms at that time. The difference of altitude (more than 1000 meters) and slopes of the Pyrenees were too strong for steam powered trains at that time. The financial question was also delicate because the major railway company didn’t wish to make important investments for probably limited profits, and local company had limited funds...

However at the turn of the century, a series of specific and local issues were favorable for the Train Jaune project. Beyond general context for railways, different elements merged together. We had mentioned the strong local political lobby for creating a railway line to link upper plateau of Cerdagne to the lower valleys of Roussillon and to departmental prefecture of Perpignan. Roads were very steep and almost impracticable during winter with strong and sudden snow falls or rainy days making its very dangerous for transportation. Cerdagne remained isolated during long winter months and had easier facilities with Spain.

For the first time, it seemed possible to overcome these natural difficulties by 1 Jacques Churet, Le petit train jaune de Cerdagne, Ed. du Cabri, 1984 ; Michel Wienin, Le Train Jaune, un chemin de fer d’exception, Ed. du Patrimoine, 2000.

2 François Caron, Histoire des chemins de fer en France, Paris, Fayard 1997, tome 1.

Michel Cotte

the railway transportation solution. To complete these arguments of territorial link with a boundary country and local rural development, three specific economic issues reinforced interest in the project. First of all, there were were specific economic resources of the region: mountain wood of course but mainly the granite carriers of Capsir, of an exceptional quality, and discoveries of mine possibilities for iron oars and spath-fluor minerals in La Tet Valley. The second one was a serious hope to promote revival of the traditional spas of the region.

It was accompanied by a serious hope for development of Pyrenean tourism that started at that time supported by wealthy English visitors and by Parisian high bourgeoisie. The third issue was the hydroelectric potential of the mountains and possibility to implement simultaneously an innovative electric railway line and electrification of the valley. Railway appears as the unified solution to endeavor to all these potentialities.

For engineers the classical solution of steam engine was not able to climb up slopes of around 50 p mille in upper La Tet Valley. Only new technology of electric powering seemed able to answer such a challenge. Even with this solution, the route through the valley to reach the pass was very delicate to design and needing exceptional civil engineering with major bridges and numerous tunnels.

The Cerdagne chance was the interest of the Midi Co. both for experiments in electric powering and for development of tourism by creation of Grand Hotels in mountain countries like Cerdagne.

Figure 1 - Line of Train Jaune inside its Buffer Zone made of Regional Park © after Cotte – Muller, Le Train Jaune..., AME ed. Montpellier 2002

• 405

Railroads in historical context: construction, costs and consequences

406 •

Official authorization for the line construction was gained in 1903, for a narrow gauge of 1.00 m classical in continental Europe at that time for mountain railways. Combined with electric powering that allowed slopes till 60 p mille and short radius curve of 80 m. Electric traction was possible by remarkable capacities of the high valley of La Tet to produce hydroelectricity. But at that time high altitude dams and pressure pipes were very new solutions. The Midi Co. wished to develop competences in such new technical fields. It was based upon the development of an original electric railway solution in France. It was the DC power at low medium voltage (850 V) by a third rail with contactors for supply of the electric railcar engines3. This solution was first experimented with success for the metropolitan of Paris inaugurated at the 1900 Universal Exhibition.

It was also implemented in mountain traction by the PLM Co. at the Le Fayet - Chamonix line (1901)4. Indubitably it was and it remains a very robust electric solution for steep mountain trains in rainy and snowy situations and also for intensive use5. It also allowed reducing the height of the tunnel vault that meant serious cost cuts in tunneling (19 tunnels). Today’s electric heritage of the third rail is perfectly authentic with a complete integrity.

The Midi Co. constructed a first dam at Les Bouillouse (2010 m) achieved in 1906. It supported some important repairs along its history but it is now very similar to the original form. The initial power station of La Cassagne achieved in 1910 still exists, with a great external architectural authenticity. Pressure pipes, turbines and electric generator were changed, but produced the same alternative three phases current under 20 000 V inside the same places and buildings. Substations along the line make the transformation in DC 850 V and they were changed with improvements of technology6.

The line of Cerdagne was achieved in 1911 from Villefranche-de-Conflent to Bourg-Madame, later to La-Tour-de-Carol (1927). Its total length is close to 63 km. It involves numerous civil engineering features because design chose to be very direct in the steepest part of La Tet Valley within deep gorges. To the 19 tunnels already mentioned we must note 24 major bridges (more than 5m span) and numerous important retaining walls. But what was exceptional for bridges is the construction of two unique and innovative railway bridges as at international level.

From the departure station of Villefranche, the first is the Fondpedrouse Viaduc by the famous Ponts et Chaussees engineer Sejourne (1851-1939). It is an inclined and curve masonry bridge (length 237m, highness 65m) with a very 3 Christophe Bouneau, Modernisation et territoire, /’electrification du grand Sud-ouest..., Bordeaux, FHSO, Universite de Bordeaux III, 1997.

4 Yves Machefert-Tassin, Fernand Nouvion, Jean Woimant, Histoire de la traction electrique, La vie du Rail, 1980, tome 1.

5 Electric powering of the Parisian metro network is still operating today by this solution.

6 Machefert-Tassin and all, already mentioned, 1980; Michel Cotte et Caroline Muller, Le Train Jaune, dossier d’etude patrimoniale..., AME - Region Languedoc Roussillon, Montpellier 2002.

Michel Cotte

Figure 2 - Train Jaune running on the Séjourné masonry bridge © Humanities Dep., UTBM 2001

original design offering two levels of arches and an impressive central rib vault .The adaptation of the architecture to a natural gorge with a very specific shape is remarkable7. Higher in the valley, the second exceptional bridge is the rigid suspension bridge at La Cassagne designed by the military engineer Gisclard (1844-1909). It is an original metallic bridge combining classical suspension bridge by cables and cable stayed solution for reinforcing the structure. It was rigid enough for supporting trains8. It was structurally light allowing a large free span with an elegant shape inside landscape made of a deep and wild gorge (total length 234 m, main span 156m)9.

Presently, these two main bridges had a very good state of conservation and they are completely authentic. They were regularly visited, painted and restored for the metallic Gisclard Bridge by the owner along years (Midi Co, SNCF, RFF). We can tell similar appreciations for all the technical issues of the line.

The main stations were still used today for the passenger traffic and keep their original plans and designs. Restrictions concern some parts of the Villefranche station without present use. It also concerns the small and abandoned stations along the line. They were sold by the Co. to some individuals. Globally, the state of conservation of the Train Jaune is excellent, perhaps unique because of today’s use of traditional rolling stocks, with some electric railcars and wagons from the beginning of the line use10.

7 Cotte & Muller, already mentioned 2002.

8 The classical suspension bridge of the 19th C was not enough rigid to support railway lines, even with light electric trains.

9 Cotte & Muller, already mentioned 2002.

10 Two electric railcars and two wagons from the original rolling stock are listed as historical monuments. They were still operating on the line at the end of the 2000s.

• 407

Railroads in historical context: construction, costs and consequences

Figure 3 - Train Jaune crossing over the Upper La Têt Valley at Gisclard Suspension Bridge © Humanities Dep., UTBM 2001

408 •

Brief history of uses from the origins to the 80s Planned for 1910, the opening of the line by the Midi Co and local elites occurred finally in 1911, due to a major accident during the line tests in October 1909. Commandant Gisclard was one of the six human victims.

Initial traffic was as expected by the promoters of the line. It was regularly used by local passengers, for diversified goods transportation accompanying a real rural economic development of Cerdagne. It allowed a larger exploitation of forest woods and granite carriers in the plateau country, e.g. for pavement of Parisian streets. Opening of mines in the upper valley of La Tet also completed the income of the Co. It was also a period of strong development for tourism both by spas revival and by the Grand Hotel at Fond Romeu, erected by the Midi Co itself and who opened in 1911 the same year than the train. It developed a summer season quickly extended by a winter season with the beginning of ice sports and ski in Pyrenees. A special rapid train connected Paris to Fond Romeu during winter seasons, with unique transfer at Villefranche to the metric mountain line.

It was a clear and successful replication of the Swiss touristic model11. Development of Fond Romeu as touristic station around Grand Hotel accompanied the golden age of Train de Cerdagne until World War II. The benefits of the Midi Co. during the interwar period were reasonably satisfying by addition of the line exploitation, the Grand Hotel seasons and the electrical energy produced for local network in fast growing, with project of new dams and power stations in the valley inside a global regional hydroelectric program under Midi Co control12.

In 1937, railways of the Midi Co were nationalized and integrated inside the new national railway company, the SNCF. There were no major changes both 11 Laurent Tissot, Development of a Tourist Industry in the 19th and 20th Centuries. International Perspectives, Neuchatel, 2004.

12 Bouneau, already mentioned, 1997.

Michel Cotte

Figure 4 - Grand Hotel of Fond-­Romeu was created by the same Midi Co than Train Jaune Line and opened the same year: 1911 Š private collection

• 409

for local traffic and for employees of Train Jaune line, with technical specialties associated to the uniqueness of the line. Workers were mainly based at Villefranche with the depot for railway cars maintenance. Midi Co. remained owner of the electrical power stations and Grand Hotel. During this period it was also the well known economic crisis of the 30s and goods traffic decreased. Rural economy was less affected than industrial sectors. Of course, Train Jaune was concerned but it was not a major victim by its diversified activity as were pure industrial lines in Northern Europe. In some way, nationalization was a consolidation of activities and amplification of the Train Jaune public duties under political control more than ruled by economic forces. First deficits for the line exploitation occurred during this period but involved in the large asset of the

Railroads in historical context: construction, costs and consequences

410 •

Midi Co and after that in SNCF counts consolidated by the State.

The period of the World War II was a period of reduced traffic pursuing and amplifying the 30s economic crisis period. Already affected by the Spanish civil war at the end of 30s, the tourism and spas traffic decreased drastically. It was a period of military control of this boundary region and illegal emigration first from Spain to France during the civil war, second from France to Spain13.

The immediate post-war period was a clear revival period for all the economic sectors and Train Jaune traffic for goods and passengers restarted strongly.

But it was for a short moment and passenger traffic decreased at the beginning of the 1950s by the buses transportation benefiting from the amelioration of mountain roads. Good transportation was also affected by the mines closures during the same period14. The turn of the 50s and 60s clearly opened a crisis period. Even tourism practices changed drastically with abandon of Grand Hotel way and development of individual housing inside “Pyrenean tourism stations” and also the boom of individual cars for middle classes. Train Jaune was progressively reduced to only little local transportations, for students and people not having cars and for limited amount of diversified goods. At that time employees and workers for Train Jaune were around 120, a very high average per km because technical specificities of the line and of the rolling stock. Some voices started to speak about not only reducing traffic and employees but also to examine closure of the line. A series of reasons gave the feeling to citizens and rail workers that it was not a serious threat. First of all was a line ruled by the concept of “public service”, especially the aptitude of the train to link mountain with valleys during winter and very rainy periods. Second was the involvement of the Train Jaune inside the SNCF national network supported by State credits and an exceptional level of syndicalism ready to defend local interests. Train Jaune employees and citizens of the region remained confident for the maintaining of the line even it had chronically deficit. The 70s were years of management optimization and regular decreasing of employees and workers number. Real gains of productivity maintained both a basic traffic and a social peace between workers, inhabitants and SNCF managers.

At the beginning of the 80s, a deeper crisis situation started and the SNCF management officially announced the definitive closure of the line, due to its large deficits without hope to develop regular traffic in anyway. During these years, an exceptional conflict opposed not only rail workers, based upon CGT organization, a powerful syndicate linked with Communist Party, but also almost all the inhabitants and their deputies and representatives. It was like rebellion of the entire region with very strong strikes and operations of “dead region”.

At that time, the regional identity of the entire French Catalonia focused upon 13 Robert Belot, Auxfrontieres de la liberte, Vichy, Madrid, Alger, Londres, (1942-1944), Fayard, Paris 1998.

14 Churet, already mentioned, 1984.

Michel Cotte

Train Jaune, bearing the emblematic colors of the Region: yellow with red lines.

It was the pride and symbol of this region, like a romantic image of local forces among snow and winter times by the determination of rail workers. That forged a guarantee for future: “Facing natural difficulties Train Jaune never stopped, he will never stop”15!

From social conflict to value recognition and tourism development The conflict of the early 80s slowly promoted two discreet issues until that period. They rose slowly in parallel: First was a revival of tourism use but only for the pleasure of taking an old and famous train, from Villefranche to Cerdagne. Second was the quality of the heritage both for the train itself and for the natural surroundings of high valley of La Tet, with very nice villages and high plateau landscapes of Cerdagne. At that time, an increasing amount of people, beyond rail workers and living train amateurs, became aware of the Train Jaune’s value especially inhabitants and cultural authorities16. It was a particularly mountain train: the highest line in France kept in an exceptional state of conservation both for line and rolling stock. It had major technical issues at national and international level: bridges, design of the line, original railcars, exceptional natural landscape with many natural features, cultural heritage of Pyrenean Mountains, etc.

Having started very discreetly during the 70s, railway tourism increased as a paradox through social conflicts of the early 80s. The Train Jaune became a railway tourism reference in France at the end of 80s. For travelers, it was a really enjoying daily trip for individuals, groups or families with children. The trip itself is fascinating, crossing over two outstanding bridges, going through many tunnels and offering a series of diversified natural and cultural landscapes.

Comeback with going down by 60 per thousand slopes is also very impressive.

Special passenger cars with open roof were constructed by SNCF to enjoy more landscapes, and specific prices were applied: two times the normal km fees17! It was clearly a seasonal activity, from May to October. The typical tour program is climbing up at morning time, having choice for a day between pedestrian walk, visit of old villages, museums, cultural exhibitions, visit of solar research center and visit of citadel at Mont-Louis (World Heritage site from 2008), enjoying local gastronomy and shopping at Fond Romeu or Bourg Madame, etc., and return at the end of afternoon to the Villefranche station where cars and buses are parked. Under social and syndical pressure, tourism appeared as a compromise between stakeholders to not close the line and to reorient exploitation toward tourism.

At the turn of century, passengers travels were around 200 000 a year, main15 Cotte & Muller, already mentioned, 2002.

16 Wienin, already mentioned, 2000.

17 Inhabitants of the region have pass with normal SNCF fees.

• 411

Railroads in historical context: construction, costs and consequences

412 •

ly with tourists having two railway courses the same day18. That corresponds to a significant amount in term of touristic railway, around 100 000 rail tourists during the 2000s, perhaps the first in France. At that time, the line was saved but probably remained not profitable for the SNCF, count of the winter “public service” without any important traffic. Through its national interests, SNCF sought for a new policy aiming to share expenses of such local and not profitable lines but with strong cultural values and strong seasonal touristic impact. At the end of the 90s, it was a place for experimentation of shared responsibilities and expenses with local communities, especially the Region Languedoc - Roussillon.

Indeed it was one of the training cases for management of the future national policy of “TER”, finally implemented in 2002 at national scale19. Line and railway infrastructures are property of a specific national society (RFF), rolling stock management and line exploitation are done by SNCF and its rail workers, surroundings of the line (stations, car parks, access roads, etc.) are supported by local authorities, running expenses and railway investments are shared by Region and SNCF. For instance, Region helped to buy two new railcars for Train Jaune (2005), after long 8 years debates for design, utility and price count of the specificities of line.

To facilitate coordination between Train Jaune touristic use and local policy aiming to maintain natural and cultural heritage, Region Languedoc Roussillon implemented a project of Regional Natural Park (PNR) at the end of the 90s, with a long experimental period. This park is a classical park for keeping and managing mainly natural mountain heritage of la Tet high valley, Cerdagne plateau and surroundings of Pyrenean Mountain; but it was also very original because it was designed around Train Jaune as skeleton and access way. PNR was supported by Region of Languedoc - Roussillon and managed locally with involvement of Pyrenees-Orientales Department and local municipalities. It was finally implemented in 2004, with headquarter at Mont-Louis, inside the historical citadel. The final surface has 138 000 hectares for 64 municipalities for 21 000 inhabitants. The park is ruled by a chart signed by all the stakeholders and it contains different specific protection of natural areas and cultural sites like citadels of Villefranche and Mont-Louis now on the World Heritage List. One of the Park’s mission is the coordination of local policies, Region and railway line exploitation to offer a large array of services from practical facilities at the main stations to large possibilities of tours and visits from the stations to the natural and cultural surroundings, and touristic information in general. It also manages a specific chart with private owner to maintain the authenticity of old stations without use and sold by the SNCF dur18 Cotte & Muller, already mentioned, 2002. We can find on the web some fantastic numbers of passengers largely higher! Villefranche station alone gives a clear idea of real touristic use of the line because almost all the touristic users start the trip from this terminal station. In 1999 departures were 104 730 and arrival 97 754 at Villefranche. Local traffic among this amount remains little.

19 «Trains express regionaux», a national contractual policy for trains of regional interest with coasts shared by SNCF and regional authorities.

Michel Cotte

ing the 80s-90s to individuals. One of the initial duties of the Park was to promote a Train Jaune dossier for the World Heritage List, and it reached the French tentative list around 2000; but the inscription of two Vauban citadels of the country on this prestigious list in 2008 stopped the impulse for Train Jaune nomination among public authorities, and it seems now abandoned as short term project.

Figure 5 - Tourism success for Train Jaune with open passengers’ cars © Humanities Dep. UTBM 2001

• 413

II- THE COMPLEX SURVIVAL OF CHEMIN DEFERDU VIVARAIS TO ITS BRILLIANT PUBLIC SERVICES (ARDECHE AND HAUTE-LOIRE) Geographical and historical contexts Globally speaking, the eastern part of Massif Central is a large mountain ridge beside the North-South Rhone Valley, named Monts du Vivarais. It was lifted by the Alpine Rise during the Tertiary Geological Period. Inside the region of the railway initiative, it offers massifs of granite with geological vestiges of volcano eruptions forming summit and picks named “sues”. These massifs were dug by deep valleys and gorges mainly going down from West to East and offering some narrow basins and many plateaus for human settlement and development of agriculture. Valleys and basins were favorable for industrial development based on water energy through family or little private enterprises, e.g. for silk industry, sawmills, leather, paper fabrics, etc. To reach the Plateau of Velay, the pass of Saint-Agreve is around 1000 m; surrounding summits reach more than

Railroads in historical context: construction, costs and consequences

1500 m (Mont Mezenc 1753 m). Slopes are not so strong than for La Tet Valley in Pyrenees but significant and valleys and gorges offered important difficulties for civil engineering.

Eyrieux Valley and Le Doux Valley were concerned there, in the central part of the ridge, and will bear the railway line project. For a long time, this region was not a priority for developing transportation networks because the main national priority was for the North-South corridor of the Rhone Valley. However, we must notice this Vivarais’ region was formerly involved in industrial processes even it remained basically a rural region during the 19th century. It was the core of silk production for Lyon manufacturers from the end of 18th century, with famous engineers improving looms and mechanical techniques like Vaucanson (1709-1782). A bit later, it was a man from North Vivarais, Marc Seguin, who implemented railways with steam engines in France by the Saint-EtienneLyon construction (1824-1833) and who improved steam engines by innovation of tubular boiler20. This pioneer line was the first in the World to climb-up serious slopes in very similar context than Vivarais valleys, around 60-80 km north but with a lower pass.

Figure 6 - The present state of conservation of ancient Vivarais CFD network, © Michel Cotte 2011

414 •

Despite its diversified economic activities, the Vivarais region remained mainly a land-locked mountainous country with a relatively poor family and auto sufficient agriculture. Historians note a large emigration during the middle 20 Michel Cotte, Le choix de la révolution industrielle..., PUR, Rennes 2003.

Michel Cotte

of 19th century toward the wealthy surroundings of Rhone Valley, Lyon region and Provence21. Roots of the Vivarais railway projects rely upon these basic human and geographical features, aiming to open up the mountainous Vivarais both for development of agriculture and industry, and to open the rural population to market opportunities.

The first project for a departmental railway network appeared at early times of the French Third Republic, based upon a law coming from Empire (1865) offering possibilities to develop local lines22. First studies were done during middle of the 1870s, showing both important difficulties for civil engineering, severe slopes for steam engines and battle of interests for the line design between different possible ways23. The Freycinet Plan and Law at the end of the 1870s, we have evoked before, boosted this regional project already studied by engineers. Following their statement, it was possible to ascend slopes and to reach the Plateau by the main rivers valleys. Classical gauge lines seemed possible with direct junction to the North-South existing lines, in Rhone Valley and Massif Central. This approach was encouraged by the Freycinet Plan aiming both to reinforce the existing national network by local and regional junctions, and to involve the Majors railways Co in these processes. Local Company of Vivarais was promulgated in 1880; but the final cost evaluation of the project by engineers was enormous and the PLM Co24, withdrew quickly from the local Co. From the project to the construction of a interdepartmental railway network (1883-1903) Nevertheless, Department and local economic interests continued to desire opening railway lines among Vivarais valleys and plateaus. A second project for narrow gauge lines with less important civil engineering works restarted immediately, always in the national framework of Freycinet Plan. Slopes were evaluated for a maximum of 30 per thousand, close from the limit for steam engine traction and the curve radius limit was reduced. These diminished the maximal speed of trains to 55 km/h, but it was not a real problem count of the steep and twisty line design. The official convention with State and Department was established in 1886, and final contractor and successful bidder of the Vivarais network was the “Chemin de fer departementaux” Co (CFD), a French company devoted to this type of regional network, using narrow gauge with specific rolling stock powered by steam engine locomotives. For build21 Pierre Bozon, Histoire du peuple Vivarois, Imp. Reunies, Valence (26) 1974 ; Gerard Cholvy, Histoire du Vivarais, Privat, Toulouse 1988.

22 Caron, already mentioned, 1997.

23 Jean Arrivetz & P. Bejui, Les Chemins defer du Vivarais, Presses et Editions ferroviaires, Grenoble 1986; Pascal Bejui, Christophe Etievant & Vincent Piotti, Le reseau du Vivarais..., La Regordane, La Roche Blanche (63) 2008. Others books are mainly devoted to historical photos and anecdotes.

24 The Paris - Lyon - Mediterranee Co was one of the powerful and famous French railway private Co of that time.

• 415

Railroads in historical context: construction, costs and consequences

416 •

ing the Vivarais network, CFD got administrative and financial support from public authorities conform to “Plan Freycinet” and to the French philosophy for “economie mixte”, merging public and private interests. Historically, CFD ruled a maximum of 14 departmental companies in France, mainly at narrow gauge, from Corsica to Manche, from Indre-et-Loire to Yonne, etc. Technically speaking, Vivarais linking Rhone Valley to Velay Plateau was one the most ambitious of these regional metric lines, and one of the oldest.

The North-East terminal station was Tournon, the administrative centre for the North of Ardeche involving the high school of Department, diversified commercial activities and silk dying industry. The second terminal station in the Rhone Valley was planed at La Voulte, close to confluent of the Eyrieux River and also an ancient important industrial place for iron industry but declining quickly at that time. The Tournon and La Voulte stations were connected to the PLM railway line on the right side of the Rhone Valley. The Western terminal station of the Vivarais was planed at Dunieres (Haute-Loire) with railway connections to Saint-Etienne and Le Puy. Starting at Tournon main station, the railway used shortly the PLM line in the city, through an important tunnel and bridge, to reach the Le Doux Valley.

From there, the line goes inside the gorge of the Le Doux River and along its medium valley to the city of Lamastre, a regional centre with some industrial activities like sawing. Civil engineering in the Le Doux gorge was significant with curved viaduct and tunnel. It was inaugurated in 1891 for around 33 km length. The construction of the section from La Voulte to Le Cheylard along the Eyrieux Valley was done during the same period for around 50 km, involving also important bridges and viaducts. Le Cheylard was the center of a typical region of Vivarais: Les Boutieres, within notable activities in textile and leather industries at that time25.

The network was completed during the following decade. First junction was to link Lamastre to Le Cheylard by the Pass of Les Nonieres, with a summit tunnel around 600 m high. Second junction was the climbing up the high valley of Eyrieux to reach the ridge summit and the Plateau at Saint-Agreve (1060 m). The difference of level from La Voulte is close to 1000 m, similar to Train Jaune declivity between Villefranche and Le Perche Pass, but largely lengthy with around 75 km. Indeed, the serious slope is from the junction station of Le Cheylard to Saint-Agreve, within a gap more than 600 m for 25 km, offering the most difficult part of the line for the steam engines with declivities often close from maximum of 30 per thousand. The line pursued among the Velay Plateau (Haute-Loire) slowly going down to Dunieres. The network achievement happened in 1903, for around 160 km of lines26.

25 Bozon, already mentioned, 1974; Cholvy, Already mentioned, 1988.

26 Bejui and al, already mentioned, 2008.

Michel Cotte

History of uses from early times to the public line closure (1891-1968) Use of the line was progressive from initial opening of Tournon - Lamastre and La Voulte - Le Cheylard in 1891. It offered an excellent training for traffic organization and to appreciate technical difficulties for traction by steam engine on such mountain network at narrow gauge with slopes around 20 per thousand in Le Doux Gorge. It was also a period to observe the first economic and social effects of regular transportation by train both for passengers and goods. It was really an opening up for the two medium cities of Le Cheylard and Lamastre, offering possibilities to go easier to the Rhone Valley, e.g. at individual scale for temporary rural works during spring and summer and also for definitive emigration. At a larger scale, cities of Le Cheylard and Lamastre knew notable development till the World War I for markets, commercial activities and industrial development by little familial firms completing what exited. In counterpart, change and evolution in rural zones kept a slow pace count of rural crisis of the end of 19th century with disease of silk worm and pandemic phylloxera in vineyards27. What benefited mostly of the final opening of the line was forest industry by wood transportation, both to the Rhone Valley for construction and domestic fuel, and for the coalmines galleries of Saint-Etienne country.

Figure 7 - On of the famous Mallet Steam Engine 030-­030 powering a long CFV touristic train in Tournon terminal station Š Wikimedia Commons

What also appeared clearly during the partial opening period were possibilities and limits of the first rolling stock. The first steam engine locomotives had 27 Cholvy, already mentioned, 1988.

• 417

Railroads in historical context: construction, costs and consequences

418 •

not capacity to climb up the Nonieres Pass or the long Boutieres ascent. For the final Vivarais opening of 1903, CFD Co ordered a new series of locomotives from the famous Swiss engineer Anatole Mallet (1837-1919), a pioneer in mountain locomotives. He offered probably the best steam engines for mountain railways at that time, as he did for the Albula Line in Swiss Alps during this period and few years after for the coal transportation through Allegany Mountains in the US28. He proposed a compound engine using four cylinders with articulated bogies making in French norm a 020-020 steam locomotive. There was also a latter version, the 030-030 Mallet. Only this type of locomotive could climb up the long and severe slopes of Vivarais pulling a commercial train with normal loads.

Management of transportation led to a train composition specific to Vivarais, named “mix trains” with selected number of different type of wagons for passenger, freight and riddle cars. It was a real success and a kind of golden age for the Vivarais during the decades before and after the World War I, till the crisis of the 30s. For instance in 1913, the Vivarais network gained substantial benefits with around 600 000 travels, 11000 tons of fast speed goods, more than 160 000 cargo tons and thousand farm animals; it employed 380 rail workers, engineers and clerks and it managed almost 16 000 trains this year29. The most efficient locomotives were the 030-030 Mallet.

Just as it had for the Train Jaune, the crisis of the 30s slowly affected the Vivarais count of its rural bases and its peripheral role for industry. In parallel, technical modernization offered new diesel engines, adapted for little number of passengers. So, in 1935 CFD started to buy diesel railcars able to pull one small wagon or passenger car. Different French firms were specialized in such materials for metric gauge and mountainous regions: Billard, de Dion, Michelin, etc. Popular surname of these small railcars was “Micheline” in extension of the firm name well known for its pneumatics! Billard Co. from Tour was the main partner of CFD. It seems that diesel locomotive for cargo transport was not experimented on the Vivarais lines. Present diesel locomotives are coming from late purchases for touristic low cost uses.

Exactly like for Train Jaune, the Vivarais lines traffic suffered seriously at the end of the 30s and during the World War II, perhaps a bit less due to its geographical situation far from the Spanish boundary context we have seen. First deficits probably appeared at that time. However CFD was not involved into the 1937 nationalization and new SNCF, like all the independent metric gauge railway companies. It remained a little, isolated company with limited possibilities of help by departments, more aware to develop roads than railways at that time30. The post-war revival was short, perhaps with some exceptions like wood 28 Anatole Mallet, Evolution pratique de la machine a vapeur, Paris 1908.

29 Bejui and al, already mentioned, 2008.

30 It was a period of intensive improvement of ancient earth ways by tarring. Ardeche was with Corsica, one of the French department having the most dense mountain road network, with highest

Michel Cotte

transportation to the Rhone Valley. During the early 60s, bus companies and first massive development of individual car transportation in France lead to a sudden and irreversible fall of Vivarais passengers. Deficit of exploitation boomed and days of CFD without any public supports were announced to be short. Closure of the line was effective in 1968 with failure of the Vivarais section of CFD. No important public manifestations from workers or inhabitant accompanied such a situation, as had happened for Train Jaune some years after. That was due to different factors. First of all was the lack of any reasonable hope for maintaining a “public service” and a pragmatic attitude of inhabitants: nobody will accept to pay such expenses. Second was probably the very quick rise of alternative projects and ideas for reusing partially the line combining birth of rail tourism and residual wood transportation.

The touristic alternative projects and the volunteers’ efficiency (end of 60s to 2000).

To save what could be saved from the Vivarais network, a social dynamic was created merging different partners: mainly a private group of amateurs of steam engines and the departments of Ardeche and Haute-Loire. It was a challenge to maintain steam traction ongoing with exceptional Mallet locomotives, still valiant after more than 60 years of hard work. It was also a vision of environmental and landscape quality to develop railway tourism, a very new perspective in Continental Europe following the first British examples31. The reopening and pursuit of the line exploitation needed a revision of the national convention with CFD Co, promulgated at the origin for 99 years.

The new stakeholder was CFTM32, an association gathering steam engine amateurs from the Lyon region and lead by Jean Arrivetz. They had an interesting pioneer experience for tourism by train powered by steam engine at Meyzieux (Rhone). Technical capacity of the volunteers came generally from a former employment at SNCF for the last generation of steam locomotives or from the transportation companies of Lyon. This private little Company was a credible partner to get the public concession of the line with legal authorization to make steam engine transportation of passengers. Forced to stop its activity at Meyzieux for urban development questions, CFTM took the opportunity offered by the vacancy of Vivarais lines to pursue its activities in a marvelous natural place, without any urban pressure and at less than 100 km from Lyon for Tournon terminal station.

Evolution of projects and definitive institutionalizations were a bit complex at the turn of the 70s. Main facts could be summarized as follow33: A new conpercent in road maintaining expenses from the World War II to today.

31 Arrivetz & Bejui, already mentioned, 1986.

32 « Chemin de fer Touristique de Meyzieux ».

33 Bejui and al, already mentioned, 2008.

• 419

Railroads in historical context: construction, costs and consequences

420 •

cession was signed between State, Department and CFTM, but public authorities refused any subvention. In terms of rail junctions, the Eyrieux line (La Voulte - Saint-Agreve) was abandoned because its less touristic interest with a road parallel to the river and the sever and long ascent to Saint-Agreve. The junction from Lamastre to Le Cheylard by Les Nonieres Pass was also abandoned for the same strong slopes reason. These lines were dismantled to be reduced to simple walk or bike paths. Finally, in 1969-70, only two sections of the former Vivarais network reopened under the new regime, forming two distinct commercial and touristic organizations: first from Tournon to Lamastre for 33 km close from the Rhone Valley and offering the trip inside Le Doux Gorge, second from Saint-Agreve to Dunieres for around 36 km crossing the Velay Plateau from the Mountain Ridge of Ardeche. First line was managed by the CFTM and it was commercially named CFV34. It got the major part of the rolling stock and steam engines heritage. Second line was managed by CFR35 an association locally organized for this circumstance with some technical initial help from CFTM. The first idea was to keep some cargo transportation beside tourism; but tourism quickly based alone the activity of new Vivarais. Regularly, the travelers’ number increased during the railway season, from April to October, offering one of the most authentic steam engine railway experiences in a nice and charming natural environment. Reputation of the line increased first in South-Eastern France, but also at national and international level. Many British amateurs made the trip to Tournon to enjoy the climbing up to Lamastre with authentic Mallet 030-030 among perfect natural and well preserved Le Doux Gorge. The financial situation of the 70s-80s was positive count of the good state of the rolling stock well maintained by the high capacity of “amateurs”, indeed the last generation of steam power rail men in France. That was done by the not profitable association SGVA36, distinct from CFTM even it was the same human group. SGVA was the owner and the ruler of the Vivarais rolling stock. Touristic organization and exploitation was the field of the commercial Co CFV, with the free contribution of old and experimented drivers of SGVA. This secured development produced benefits and some financial reserves. That allowed an ambitious purchase policy aiming to develop the park of locomotives, diesel railcars and passengers’ cars at metric gauge, coming from others regional networks stopping their activities without new perspectives. Passengers were around 50 000 a year from the 80s to the end of century, with rise to 60 000 during the 2000s. Commercial organization increased and also the employers and young workers to replace steam amateurs from the first generation37.

34 « Chemin de fer du Vivarais », one of the former name of the railway network.

35 « Chemin de fer regional » 36 « Sauvegarde et gestion des vehicules anciens ».

37 Bejui and al, already mentioned, 2008.

Michel Cotte

Figure 8 - A Billard diesel railcar passing a cut inside the Le Doux River Gorge © Wikimedia Commons

From the golden age of tourism exploitation by volunteers to the structural changes of the 2000s The first important turning date awaited by all the stakeholders was 1985, with the end of public concession from the initial Vivarais CFD Co (1886). At that time, property of the line, stations and functional lands came back to public ownership. The two railway branches still exploited were in different position to negotiate this important step. CFTM, within a good exploitation period was able to buy the line in not so bad conditions. On the Plateau, the situation of CFR was not so favorable, with a reduced touristic traffic and limited means, lack of notoriety and no real funds. This little company was not able to pay the end of concession and it stopped its activity. The restarting of this line was a difficult challenge, first supported by the creation of a syndicate of municipalities along the line purchasing the line and its stations (1991). Regular reopening was effective only in 2002 by VFV38, a new commercial and touristic association supported by the Syndicate. This first financial intervention of public authorities must be noticed, with revival of “mixed economy” at local level, with direct public support for investments. Generally speaking, local and departmental public authorities were traditionally reluctant for such involvement during the last decades of the 20th century, because they had limited budgets with other priorities than tourism.

The group CFTM-CFV managing the Tournon -Lamastre line appeared as a 38 « Voie ferree du Velay »

• 421

Railroads in historical context: construction, costs and consequences

422 •

flourishing company during that period. For instance, at the turn of the century, SGVA was owner of around 80 locomotives, railcars, passengers’ cars and wagons generally in good maintenance condition, and around 25% were classified as Historical Monument by the Ministry of Culture, probably a unique situation for a local railway Co39. But indeed the situation was fragile, undermined by ineluctable trends. First already evoked was the change of generation with replacement of ancient professionals of steam engine, very capable and acting as volunteers, by new and young employees with normal salary. The average of volunteers versus employees evolved and increased fixed charges for the Co. Another issue was the maintenance of the line, needing important works for rail tracks, reinforcing ballast and changing rails, repairs of masonries, etc. That was undertaking by the CFTM-CFV on its own funds but creating important expenses. On the other hand, touristic policy was pushed to maximize trains size with great number of passengers’ cars. Daily intensive use of steam engines and their centennial age with normal decay led to some important maintenance problems, out of the Tournon depot capacity as refection of tubular boilers.

At the beginning of the 2000s, it was clear for an astute observer that such system of exploitation was running to close disrupts. A SEM (mixed economy society) was created in 2003 to bear some financial supports to CFTM, with involvement of Ardeche Department. Quickly SEM took majority and control of CFTM, mainly to insure the line maintenance; rolling stock remained under SGVA ownership and maintenance. Addition of technical problems for locomotives, volume of fixed expenses for CFV exploitation, need for deep tracks maintenance, and also a latent conflict with SNCF for the use of its Tournon tunnel40 led CFTM-CFV to stop its activity and to declare the failure of the exploitation (October 2008)41.

The present project for the Le Doux Line (or “Mastrou42”) At that time, some important conclusions became clear for restarting any form of sustainable exploitation. All the potential stakeholders must work together for a common project with shared responsibilities. It was not so obvious with an array of individual attitudes, sometime with political intentions. Ideas go from proposal to reopen the ancient public service with public funds to a lim39 http://trainduvivarais.org/spip.php7article5 40 CFTM-CFV had to pay an expansive annual toll for using the Tournon SNCF’ section, around 2 km mainly made of a long tunnel, and RFF-SNCF wished to stop metric traffic on this line normally devoted to intensive international cargo traffic.

41 CFTM was ruled by the public SEM, but if public authorities accepted to pay for investment and conservation of the property, it refused to pay functioning and salaries for such type of activity not considered as a “public service”.

42 “Mastrou” was the popular surname of the Le Doux Line coming from the ancient time of CFD. It is a contraction of something like “The small train coming from Lamastre” or Lamastrou or Mastrou in local language.

Michel Cotte

ited exploitation of around 15 km by a pure commercial company43. However, a large majority of departmental deputies, local councils and SGVA members progressively reached a common view for the future with clear roles devoted to the different partners. Technically speaking it would be better to have traffic out of the SNCF section. That meant abandon of the Tournon station and depot, creation of a new station at Saint-Jean, partial transfer of depot to Lamastre. The local Communities of communes44 along the line agreed and voted financial support for this change, mainly imaged by the project for a new station at SaintJean-de-Muzols. Department of Ardeche and Region Rhone - Alpes regained the line property through the CFTM-SEM heritage and eased the line maintenance. Association SGVA maintained its ownership upon the rolling stock but probably with public support, e.g. for locomotives and passengers’ car listed as historical monuments or for applying security rules for tourism transportation.

Nevertheless a difficult problem seemed not to be resolved, for the repairs of the most damaged steam engines, needing important works for tubular boilers.

No French firm had kept such boiler works competences at normal price. It was perhaps possible to have well fitted workplaces in Poland, but with important transportation fees. For commercial use a bid was organized to make a contract with a professional company of tourism or transportation. A local consortium of professionals in tourism transportation gained the bid.

At first a rail bike trip was proposed during summer 2011, to descend the Le Doux Gorge from the station of Boucieu to the station of Troye, for around 12 km among the most impressive part of the Valley with a regular slope around 20 per thousand. It is a real success, with number of echoes in newspaper and visit of some TV and movies stars. The site was indeed famous for movies, with more than 20 films using the Vivarais line and its landscapes from the end of 1960s. Coming up was done by diesel railcars picking up the bikers and pulling the rail bikes.

The project for the new terminal station was promulgated recently by the Community of Tournon region. It is a project evaluated around 3 million €, with land acquisition at the entrance of the gorge, involving station itself, technical facilities for rolling stock, a museum and communication hall, a car park for more than 300 cars, camping cars and buses. Others issues could be implemented in future: fast food, housing for guards, etc45. Complete opening will be done after the station construction, normally for spring 2013. It also depends of the steam engines repairs but it is secured by railcars and diesel locomotives; two were bought recently by the consortium coming from other metric gauge networks. Steam engine sustainable use remains today a question without an 43 Journal de Tournon-Tain from 2008 winter to today, it is a weekly local newspaper.

44 France is the European State with the smallest local administration by « communes », meaning that all the ancient villages have mayor and local council. Communities are a recent attempt to gather groups of communes to have common services, administration and financial management.

45 Journal de Tournon-Tain, 30-2010.

• 423

Railroads in historical context: construction, costs and consequences

Figure 9 - The most recent attempt for a revival is going down the Le Doux Gorge by rail bikes © http://velorailardeche.com

424 •

obvious solution46, but diesel traction and rail bikes seem to have a nice future.

Another pending question not so easy to solve is the exact length of the line for future exploitation. We have already seen the suppression of the Tournon - Saint-Jean section for practical reasons for around 4-5 km. On the other side, upper terminal station is not so clear today. For political and general economic reasons, all the stakeholders, mainly politicians, claim that Lamastre is the historical terminal station for the Doux Line or “Mastrou” and it must remain the upper terminal station. Technically, it offers a large station with important facilities to supply Tournon as depot. On the other hand, but meso voice, some people have reluctance to this solution as sustainable solution, and suggest an intermediate station as terminal point, like Boucieu at the upper end of Le Doux Gorge. Mixed solutions will probably experimented for next years with some activities starting or stopping at Boucieu, like rail bikes now, and others like “grand train” with steam or diesel locomotives will reach Lamastre. The future of the exploitation will indicate what is realistic and sustainable and what is not.

46 Intensive use with too heavy trains as it was done during the 2000s must be stopped for long conservation of steam engines; on the other hand a reasonable use of steam locomotives by capable drivers with regular maintenance could be operated compatibly with good conservation.

Michel Cotte

Figure 10 - The project for a new terminal station at the entrance of Le Doux Gorge by the le local Community of Communes © Communauté de communes du Tournonais, published by Journal de Tournon – Tain, n° 30, 2011.

• 425

Stefano Maggi

A NEW ROLE FOR AN OLD RAILWAY: TOURISM NOVOS USOS PARA UMA VELHA LINHA DE CAMINHO DE FERRO: TURISMO Stefano Maggi (U. Sienna, Italy Itália) Stefano Maggi is professor of Contemporary History in University of Siena (Italy). His research has focused on transportation issues, especially railways, as well as history of international trade and history of pharmaceutical industry. He teaches courses on history of communications, history of the regional development and contemporary history.

Stefano Maggi é professor de História Contemporânea na Universidade de Siena (Itália). Tem investigado temas relacionados com os transportes, em especial transportes ferroviários. Ensina cursos de história das comunicações, história do desenvolvimento regional e história contemporânea.

Abstract Resumo

This paper deals with cultural value and significance, conservation aspects, re-use and tourism.

All over Europe, the old branch lines, built when the train was the main means of transport, needed to find a new role in the age of mass motorization.

Quality tourists appreciate the cultural value of an old railway, when the railway is integrated on a territory, where the train was fundamental in the 19th and 20th century for connecting towns and villages.

The stations, depots and goods yards of the railway and the scenery visible from the train windows represent an important added value for making the most in the marketing of a territory.

Slow trains can give an historical interest to discover a territory, sometimes also for the residents of the zone.

In contrast, new tram-trains and old trains can give an innovative image of the railway.

In Italy the use of railway heritage is a very recent development, due to the fact that only the remains of the Roman and the Medieval Age were considered as History.

Historical trains on some branch lines have developed independent of the railway preservation movement which was establishing all over the world, leaving from Great

• 427

Railroads in historical context: construction, costs and consequences

Britain in 1951. We can say that it happened in spite of that movement, without the discussions of scholarly researches in comparative perspective.

In the 1990s, the demand increased for steam trains by groups who would charter them for special occasions, at times on the secondary railways, at other times on the national network and usually for travelling into the countryside.

My paper will analyse a railway line in Tuscany and the attempt to develop historical trains as a re-use with cultural value, on a line closed to normal traffic, the Orcia Valley Railway Asciano-Monte Antico, in Tuscany. This railway was closed in 1994 and tourist trains have until now prevented the track from the definitive abandon, while contributing to give value to the valley, which became UNESCO world heritage.

428 •

Este trabalho trata do valor cultural e seu significado, dos aspetos de conservação, re-utilização e turismo de antigas linhas de caminho de ferro. Em toda a Europa existem antigos troços de linhas de caminha de ferro, construídos em tempos em que o comboio era um meio de transporte de massas, que precisam de encontrar uma nova utilização nos tempos de motorização em massa.

O turismo de qualidade aprecia o valor cultural de uma antiga linha de caminho de ferro, quando a linha esta integrada num território em que o comboio era fundamental para unir vilas e cidades nos séculos XIX e XX. As estações, armazéns e depósitos do caminho de ferro, e o cenário visível de um janela de uma carruagem, representam um importante valor acrescentado para o marketing territorial de um território. Comboios a baixa velocidade têm um interesse especial para ajudar a descobrir um território, por vezes até mesmo para os residentes da zona. Pelo seu contraste, comboios antigos e tram-trains podem dar uma imagem inovadora do caminho de ferro.

O uso de linhas de caminho de ferro antigas é um desenvolvimento recente em Itália, devido ao facto de aí apenas os vestígios dos tempos de Roma ou da Idade Média serem habitualmente considerados como históricos. Alguns comboios históricos em certos troços de linhas desenvolveram-se de forma independente do movimento pela preservação de caminhos de ferro que se foi estabelecendo, especialmente no Reino Unido a partir de 1951. Foram desenvolvimentos feitos sem discussão nem investigação académica numa perspetiva comparativa.

Nos anos 90 aumentou a procura de comboios a vapor por grupos de entusiastas, capazes de os alugar para ocasiões especiais, por vezes em linhas secundarias, outras vezes na rede nacional, habitualmente para viajar nas zonas rurais.

Neste trabalho analisa-se o caso de uma linha na Toscânia e as tentativas para desenvolver comboios históricos como uma re-utilização com valor acrescentado, numa linha fechada ao trafego normal, linha do Vale de Orcia, em AscianoMonte Antico, na Toscânia. A linha tinha sido encerrada em 1994 e os comboios turísticos evitaram até agora o abandono definitivo da linha, contribuindo com valor acrescentado para o vale, que se tornou património mundial da UNESCO.

Stefano Maggi

A new role for old railways: tourism Stefano Maggi

1. THE ORIGIN OF TOURISM AND THE RAILWAYS In the nineteenth century in Italy, the thermal bath used at the time of the Roman Empire, after a long period of neglect re-established itself. However tourism became a considerable phenomenon only after the second half of that century, with a time lag in comparison to other European countries, because both the Risorgimento process and independence wars retarded the economical development. New tourist destinations spread out thanks to the interest towards the countryside; this was due to a return to nature, which is one of the typical subjects of Romanticism, a literary tendency which was becoming very popular. At first the beauty of mountains was appreciated, and like in other countries, the Alpine Club was founded. The Italian Alpine Club (CAI) was set up in Turin in 1863, with the aim to boost both knowledge and studies about mountains, in order to improve their accessibility both in northern regions and in the other parts of the Italian peninsula1. In the same period there was a development of bathing resorts with a consequent growth of beach tourism; this was a phenomenon which began in the second half of the nineteenth century. However at the beginning of the twentieth century only the upper-middle class and aristocracy were interested in tourism: above all they came from foreign countries, so this phenomenon was identified as a “foreigner’s movement”. A specific Italian association for the foreigner’s movement (Associazione Nazionale Italiana per il 1 Mila, M. (1965), Cento anni di alpinismo italiano. In C.E. Engel, Storia dell’alpinismo (p. 251), Torino: Einaudi.

• 429

Railroads in historical context: construction, costs and consequences

Movimento dei Forestieri) was set up in Rome in 1902.

As Italy lacked both a wide process of industrialisation and urbanisation, there wasn’t a middle class rich enough to spend part of its money on holidays.

For example, in 1903, hotels cost about 7-8 Italian liras per night with lunch and dinner included, yet workers earned 2-3 Italian liras every day; they spent their money on necessary articles2. However the most perspicacious observers affirmed that these resources could be open to significant developments. Maggiorino Ferraris, editor of the most important Italian review, Nuova Antologia, stated: this is one of the biggest sources of wealth and profit of our country; however it will be even larger in the future… As everyone appreciates English people, who are able to exploit their iron and coal treasures in the subsoil, why shouldn’t Italy use its own sun, weather, sky and art resources with greater energy ?3.

430 •

Most of the Italian railway network was constituted after the accomplishment of the national unification in 1861. Actually, in that year, Italian railways measured only 2700 km; in 1881 they measured 9500 km and in 1901 16,400 km. In 1921 the railway network reached an expansion of 20,500 km: a greater expanse than its current length in kilometres4.

Before 1865 the management of the railway-sections was not uniform, as it derived from different pre-unification experiences. Some railways belonging to the State in Piedmont, existed at the same time together with the lines owned by various private companies, such as Strade Ferrate del Mediterraneo, Strade Ferrate Romane, Strade Ferrate Meridionali. Every company received considerably different concessions from the other ones.

In 1865 it was decided to entrust five companies with private investment, which had to be spent for the construction and exercise of the railways. However, this system almost immediately went through a crisis, so the State had to spend a large sum of money for debt control and the balancing of budgets.

Consequently a long debate arose to find a new order, which was finally put into practice in 1885. Three big companies were entrusted with this operation, with State ownership of lines and permanent installations, which were outsourced to the private companies, while the rolling stock was privately owned.

The peninsula was divided longitudinally with the intention of encouraging traffic between the north and the south of Italy, in order to overcome the considerable differences between the two areas of Italy. The Società per le Strade Fer2 Battilani, P. (2001), Vacanze di pochi, vacanze di tutti: L’evoluzione del turismo europeo, Bologna: Il Mulino.

3 Ferraris, M. (1912), Per le industrie termali e climatiche d’Italia, Nuova Antologia, fasc. n. 968, pp.

697-698.

4 Maggi, S. (2003), Le ferrovie, Bologna: Il Mulino, p. 126.

Stefano Maggi

rate Meridionali was entrusted with the eastern network and managed in total 4300 km of railways, while the Società per le Strade Ferrate del Mediterraneo was entrusted with the western network and managed 4100 km. The third and final grantee company was the Società per le Strade Ferrate della Sicilia which had to manage 1100 km of lines on the island. The grants of 1885 lasted for a period of sixty years, divided into three periods of twenty years.

This system also failed and the State was forced to offer repeated financial handouts, until the nationalisation of 13,000 km of railways – the majority of the existing network – was achieved after a heated debate in April 1905, relatively early in comparison with the other European countries. A State Railway administration (Ferrovie dello Stato - FS) was created.

2. TRAVELS BY TRAIN FOR TOURISM While the railway network developed, in Italy the Grand tour tradition exhausted itself. The travel by the European upper classes towards foreign countries, such as visits to the Italian peninsula and its towns in order to learn both the language and to find traces of the Roman past ended in the railway time. The last romantic travellers, such as John Ruskin and Henry James, who both used stage coaches and railways for their trip in the mainland, showed a certain sense of regret toward the passing of the slow postal stage coach and individual travel based on great care5. A new standardized tourist movement would have taken its place, with the origin of a real industry strictly connected to it.

The train and then the extra-urban tram, of which a wide network developed in Italy, were decisive in encouraging a previous form of popular tourism, the so called “gita fuori porta” (an excursion outside the town): this was a day or half-day trip beyond the town with the purpose to discover ancient archaeological remnants or admire the beauty of nature. Thanks to the train, village fairs developed; in these occasions special trains with very low rates were promoted.

However the main tourist movement remained the foreigners’ one, and starting the 1870s the Italian railway companies introduced facilities with the aim to promote pleasure travel, such as cut-rate tickets for tourist itineraries. Tourist tickets were also activated on long distances; they allowed travellers to perform a real, complete tour of Italy, across lines which belonged to different railway companies.

5 Brilli, A (1986), Viaggiatori stranieri in terra di Siena, Roma: De Luca, pp. 67-68.

• 431

Railroads in historical context: construction, costs and consequences

432 •

Moreover, for the first time cheap return tickets among many stations were suggested. The validity of this kind of ticket ran from the last train of the day before the holiday until the second train of the day following the holiday. These tickets provided for a reduction from 25% to 35% according to the distances; however they usually had a limited range of action (short and middle length itineraries)6.

Besides, the government suggested companies adopt differential rates based on a diminishing rate of fare increase over distance. Such rates would have had to replace flat rates, which were absolutely prohibitive for long distances. However companies didn’t agree with this proposal; they feared that a strong decrease in incomes could compromise budgets. The differential rate was very important also to develop tourism and to encourage new travels; in Italy it was brought into effect only in 1906, thanks to the State management of railways.

Also special competitions named “stazioni fiorite” (stations in bloom) were closely connected to tourism; the first one was in 1911, on the occasion of the fiftieth anniversary of the Italian unity. These competitions had the purpose to embellish stations with flower beds, awarding prizes to the best stations. The Great War 1914-18 interrupted these competitions. They started again in 1925 after the coming of Fascism, in cooperation with tourist boards, such as the Italian Touring Club (TCI), the Italian Tourist Board (ENIT), and the Italian Federation of Agrarian Consortiums (FCA)7.

Moreover it is important to remember that a lot of railway lines on the Alps and on the Apennine mountains had a marked attraction and potential for tourism, as they linked medieval towns to the railway network through wonderful landscapes. In 1919, a monthly magazine published by the Italian Touring Club, Le Vie d’Italia, stated: We all know that nature is one of the fundamental factors which cause the tourist movement: however we also have to be well-organized. The salutary effects of climate, the influence exerted by altitude, the whispering woods and winds, the rushing waters and the beauty of landscapes are all precious elements: nevertheless if we don’t exploit these elements, they remain almost entirely sterile. We have to do our best to make the places – where nature more generously spread its gifts – accessible, comfortable and smooth. We have to make the customers’ stay here conform to their high standards of living, as these people are complex and may have different mentalities... Here we would like… to deal with a very particular aspect of the means of communication: we refer to the small accessory railway lines which are joined to railway networks and which 6 Guadagno, W. (1991), Ferrovie ed economia nell’Ottocento postunitario, Roma: Cafi, p. 283.

7 Tedeschini-Lalli, E. (1927), Per l’estetica delle stazioni ferroviarie, Le Vie d’Italia, XXXIII, n. 2, p. 153.

Stefano Maggi

precisely serve to approach the most interesting stations on the mountains: simple adherence railways of mountain or funicular railways, rack railways, cableways. There are lines which leave the bottom of the valleys and reach the places with the highest panoramas, sometimes they reach passes and summits …8.

Even if shorter in comparison to Switzerland, the mountain tourist railway network of Italy was not modest. These infrastructures were built between the end of the nineteenth century and the first thirty years of the twentieth century, as the train reached several Alps and Apennine mountain resorts. These railway lines were real masterpieces of engineering: nowadays only few of them are still working, because of the lack of awareness towards their value.

3. THE TOURIST INSTITUTIONS After the Italian Alpine Club (CAI), the Italian Cyclists’ Touring Club (TCCI) was the first national association set up with tourist purposes; it was established in Milan in 1894 and it was based on the model of the English Cyclists’ Touring Club, with the aim to promote the use of bicycle both for pleasure and travels.

In 1900 the name was changed into Italian Touring Club (TCI) because of the increasing tourist activities of cars and trains. A cooperation between Italian Railways and Italian Touring Club began: in 1903, railway lines guide books, edited by Ottone Brentani, were printed for the first time: as we can read in the introduction, they wanted to be “not a guide for towns and countries where you pass through, just the railway line guide book, in order to show and explain, in a simple way, only what can be watched by train”9. Then, between 1907 and 1921 a series of regional monographs for tourism and railways were published; they were dedicated to Apulia, Umbria, Abruzzo, Sicily, Piedmont, Latium and Emilia. Moreover, a particular space was reserved to the railways in the first Italian guide published between 1914 and 1929 by the Italian Touring Club: information in it was so accurate that they even suggested the passenger on which side of the carriage it was better to sit in order to enjoy the sights at their best10.

Between the end of the nineteenth and the beginning of the twentieth century, railways had such an obvious, well-known and fundamental importance in the Italian society, that several observers suggested to entrust Ferrovie dello Stato with the responsibility of a national tourist bureau, with the purpose to promote and coordinate this growing phenomenon; the public administration 8 Gerelli, A. (1919), Le ferrovie di montagna nei dintorni di Bolzano, Le Vie d’Italia, III, n. 11, p.

662.

9 Touring Club Italiano (1905), Guide di linee ferroviarie. Milano-Genova, Milano, p. 2.

10 Vota G. (Ed.) (1954), I sessant’anni del Touring Club Italiano, Milano, p. 149.

• 433

Railroads in historical context: construction, costs and consequences

decided not to take on this project.

In 1913 Maggiorino Ferraris proposed a public bureau which could deal with tourist matters, which was to be put in action by the management of the State Railway.

434 •

Its own activity should... be realized in two ways: acting towards foreign countries and acting internally. Abroad the foreigner’s bureau should act especially thanks to a thick network of shipping and railway agencies...

These agencies, in their turn, would take advantage of three means to perform their action: advertisements, free information, ticket selling…

In-country, the tourist railway bureau should take charge of all the factors which contribute to promote, enlarge and to make the foreigner’s and travellers’ movement easier, such as railway and postal utilities, hotels, museums, public security. So, I suggest the institution of a Tourists’ Movement General Council in care of the general management of the State Railway, on which Ministries could be represented, together with the greatest national associations, the Chambers of Commerce, the municipalities which have a wide touristic movement... Only the Ferrovie dello Stato have or can have this wide network of executive agencies abroad and in-country, which is essential for a practical and profitable work11.

Filippo Tajani, one of the most well-known railway scholars, university professor in Milan at Politecnico, observed that it was inopportune to entrust the Ferrovie dello Stato with a national tourist board. They – he wrote – «have in their hand the basic mean to promote tourism... A journey brings with it, not only great attractions around it, but also an amount of discomfort: railways have to reduce them to a minimum». In order to obtain this result, it was necessary for the main stations to find always an interpreter who could give information to the passengers in different foreign languages; the train crew had to be “provided with a suitable, clean and neat uniform”; they had to learn to look after their customers, in the same way a shop assistant looks after his regular customer; they had also to check tickets, «being strict with cheats and understanding towards people unaware of the difficult fare rules». In this way – so thought Tajani – “the railway will bring an invaluable contribution, with a result otherwise unimaginable even spending a lot of money for advertising”. Regarding this, he also proposed a change of politics accepted till then: Nowadays railway advertising is meant as a sort of advertising of our 11 Ferraris, M. (1913), Di un ufficio di Stato per il movimento dei forestieri, Nuova Antologia, fasc.

985, pp. 149-150.

Stefano Maggi

country’s beauties. This is based on an apparently persuasive argument.

Spreading the knowledge of Italian natural and artistic beauties, foreigners are so tempted to come to Italy: so railway travelers increase and the aim of advertising is achieved. However this way of thinking is too reductive. Our countries’ beauties are known even in Lappony; even hotelkeepers in Switzerland agree on this. As it is a long travel to Italy, we must tell them that they can see all those extolled Italian beauties without a great amount of money and, above all, with no great discomfort at all.

We must tell them that when they come to Italy… they will not find there Stendhal’s stage coach, on the contrary they’ll find there a great amount of comfortable and express trains, wagon lits and restaurants, provided with electric lighting, with favorable prices12.

The debate on the necessity of a national office for tourism and on the advisability to entrust Ferrovie dello Stato with it had no result at all for many years. However in the meantime some important prescriptive agreements were brought about. The first law on this subject was issued in December 1910 (law 11 December 1910 n. 863). This provided the power to some communes to impose a visitor’s tax in their municipal territory to those tourists staying there more than five days; of course these communes had to be marked out by their importance in being health-resorts or seaside resorts. This tax had only to be used to finance necessary works, in order to improve tourist facilities. The tax for staying was then extended by an ordinance on May 1920.

After the First World War Italy realized the economic importance of tourism and it started more active politics. A committee of studying was set up. Its aim was to work out proposals about the development of hotel trade and tourism; moreover it had to think about the opportunity of setting up a public organization of tourism. It was so passed an ordinance based on the committee report: this approved the establishment of the Italian Tourist Board (Ente Nazionale Industrie Turistiche - ENIT) in 1919. ENIT’s main tasks were: the integration of private enterprises for a cultural and an advertising promotion both in Italy and in foreign countries, the supervision on hotel activities, the gathering and the elaboration of statistic-economic information.

ENIT cooperated with Ferrovie dello Stato and Italian Touring Club, offering advertising brochures and taking part in tourist exhibitions with its own stands. ENIT cooperated above all with Ferrovie dello Stato. On December 1923 both administrations drew up an agreement: this had to last nine years and it was «about the management and running of Offices for Travels and Tourism in Italy and in foreign countries…, about information service to the public, about 12 Tajani, F. (1917), Le ferrovie dello Stato e il movimento dei forestieri, Le Vie d’Italia, I, n. 3, p.

153.

• 435

Railroads in historical context: construction, costs and consequences

the selling of railway tickets and about the issuing of some common advertising material». After a few years, in August 1926, ENIT and FS constituted the CIT (Italian Tourism Company). ENIT gave it the management of offices for travels and tourism, of information offices in stations and harbors. Ferrovie dello Stato on the contrary gave it the management and running (both in Italy and in foreign countries) of offices for travels’ development, for information to the public and the selling of railway tickets. Moreover, CIT had the task to spread advertising materials issued together by ENIT and FS, such as booklets translated in many languages and the Rivista Mensile di Propaganda, whose circulation was more than 30,000 copies (after, their circulation increased up to 50,000 copies) for the English, the French and the German edition. The first copy was printed in January 193313.

Many other initiatives were taken in this field by Fascist Regime: travels of Italian people and foreigners were encouraged by authorizing favourable fares, and above all by establishing special trains for tourists only.

4. TOURIST TICKETS AND “POPULAR TRAINS” 436 •

In the early 1930s Ferrovie dello Stato reported a great decrease in the flow of travellers: this was due both to the economic crisis spreading through all over the world and to the increasing competition of cars. In order to face the crisis, a lot of express trains on long distances were set up. Starting in 1932 very short trains consisting of 2 or 3 carriages and no luggage van were set up, too; so it was possible to increase the average speed thanks to prompt starts and brakings.

Moreover, starting in the Spring of 1931, Ministry of Communications granted special discounts for some destinations. These discounts concerned families going to thermal or seaside resorts, foreigners from other countries and going to Adriatic seaside or to some thermal resorts, and Italian people going to the mountains in the Dolomites14.

Ferrovie dello Stato also set up trains to the mountains for skiing. So it was set up one of the first special tourist train from Rome to the Abruzzi Appennine: it left Rome on Saturday at 1.00 p.m. and came back on Sunday shortly after midnight. This train had a luggage van in its composition; this van was equipped with ski supports, whose freight was free15.

In 1931 circular tickets for foreign tourists were issued: they were valid on 13 Ministero delle Comunicazioni. Ferrovie dello Stato, Relazioni per gli anni finanziari 1927/29, p.

9; 1932/33, p. 15; 1933/34, p. 12; 1934/35, p. 12.

14 (1932) Facilitazioni di viaggio per le stazioni balneari e termali, e per l’Alto Adige e il Cadore - I viaggi delle famiglie, Le Vie d’Italia: Notizie ed Echi, XXXVIII, n. 7, p. 274.

15 (1931) L’Istituzione di un treno speciale per sciatori, Le Vie d’Italia: Notizie ed Echi, XXXVII, n.

3, p. 113.

Stefano Maggi

different routes on the FS railway network16. Moreover Sundays and holidays return tickets came into operation, with a reduced fare of 40% for some important tourist or historical places. These tickets were valid from Saturday till Monday17.

However our weekends didn’t exist. Until 1935 Saturdays were not holiday; as a matter of fact people worked eight hours on Saturday and even the institution of the so called “fascist Saturday”, the second day of rest, didn’t substantially change habits. However the fascist Regime together with Opera Nazionale Dopolavoro (the Fascist institution for workers’ free-time activities) organized successfully mass entertainments on Sundays and on holidays; inside this institution there were people of all ranks. The membership card of dopolavoro allowed discounts on travels, magazine subscriptions, tickets for cinema, theatre, dance-halls and football matches18.

The trend of the Fascist Regime, aimed to the amusement of people in order to create consent, allowed the establishment of “very cheap” trains. Starting in August 1931 and running above all in the following years, the so called “popular trains” began to leave the most important Italian railway stations: these were special trains with only third class tickets, with a discount up to 80% on the standard rate, a promotion with no precedent which allowed thousands Italian people to travel by train for the first time.

Le Vie d’Italia, briefly explained its accomplishment rules: Special trains with only third class carriages, leaving the main cities, can be organized on every holiday: these are trains for pleasure trips to some resorts chosen between the most interesting ones from a tourist, historic or folkloristic point of view.

Departure is usually between 5.00 a.m. and 7.00 a.m. and return is within midnight of the same day, in order to give travelers the opportunity to spend the whole day at the chosen resort… The communication of the trip is given a week before, on Monday, and tickets go on sale the same day…

Cut price tickets are valid only for this special train or this pleasure trip: this means, mind you, that if traveller misses that train for any reason while going or coming back, he cannot get onto any other train even paying the difference, and he cannot have any refund at all: these rules are explainable if you think that transport is made with very special conditions and cut prices…

Special trains are always express trains19.

16 (1931) Nuovi biglietti circolari turistici, Le Vie d’Italia: Notizie ed Echi, XXXVII, n. 8, p. 342.

17 (1931) Biglietti di andata-ritorno domenicali e festivi, Le Vie d’Italia. Notizie ed Echi, XXXVII, n.

8, p. 342.

18 Venè, G.F. (1988), Mille lire al mese. Vita quotidiana della famiglia nell’Italia fascista, Milano: Mondadori, pp. 221-223.

19 (1931) Importanti innovazioni in materia di facilitazioni ferroviarie adottate sulla rete statale, Le

• 437

Railroads in historical context: construction, costs and consequences

438 •

These trains were carried out with a political and social aim; it was so that “popular trains” succeeded in creating a new traffic of travelers towards climatic, seaside, historical or artistic resorts. Results were very good: for 8 festivities on August-September 1931, 415 round trips were made on special trains and 80 round trips on customary trains, transporting 459,000 travelers20.

In 1931, the season of “popular trains” began on 5 June and ended on 18 September; during this time, there were 17 days of pleasure trips, with a 58 trip average per day. In comparison with the previous year, there was a wider development of mainline trains; moreover a second class service for night trains was established. Almost all trains left stations full with travelers. In all there were 948 round-trip special trains, 834,000 travelers and an average distance of 485 km (going and coming back). Then there were 152 round trips on ordinary trains: these were used when people weren’t enough and special trains weren’t necessary21.

The success of “popular trains” also continued in the following years and railway traffic increased a lot. All of this stopped when the Fascist Regime decided to declare war to Ethiopia: 864,000 travelers were transported in 1933 during 20 festivities and 1,030,000 travelers were transported in 1934 in 22 festivities.

Popular trains were important because they permitted the first spreading of mass tourism in Italy, allowing thousands people to reach holiday resorts for the first time. However this phenomenon regarded a small part of people, i.e. those people living in big cities and working in factories and offices. People living in the country and in small cities, on the contrary, had to wait till after the second postwar period to gain the “right to vacation”22.

In the period of the “golden age” (1955-1965), most part of the Italian people became tourists: however the train lost the social and economic role it had for over a century and it was replaced by cars. The car was going to be a consumer good; a lot of railway lines were definitively closed. Even the most beautiful railway lines located where tourism was increasing couldn’t be saved.

Only at the end of the twentieth century, a movement for the safeguard of tourist railways began to spread.

Vie d’Italia. Notizie ed Echi, XXXVII, n. 9, pp. 383-384.

20 Ministero delle Comunicazioni. Amministrazione delle Ferrovie dello Stato (1932), Relazione per l’anno finanziario 1931-32, Roma: Istituto Poligrafico dello Stato, p. 123.

21 (1932) Il magnifico risultato dei treni popolari di quest’anno, Le Vie d’Italia. Notizie ed Echi, XXXVIII, n. 11, pp. 417-418.

22 Degl’Innocenti, M. (1993), La villeggiatura: un diritto per tutti. In Il Risorgimento, XLV, n. 2, pp.

205-222.

Stefano Maggi

5. TOURIST RAILWAYS AND VINTAGE TRAINS: THE NATURE TRAIN IN TUSCANY In Italy historical trains developed independently of the railway preservation movement which was establishing in some countries of Europe. That movement was little known. During the 1980s and the 1990s the demand grew for steam trains by groups who would charter them for special occasions, at times on the secondary railways, at other times on the national network and usually for travelling into the countryside.

This lack of knowledge of what was happening abroad led to the demolition of a great deal of historical trains, in particular after the asbestos scandal during the mid ‘90s, when it became known that this mineral fibre provoked cancer to workers of the factories where it was manufactured; given that this material was used as an insulator, a great deal of the old rolling stock was demolished.

Above all, however, in Italy there has been no culture of industrial archaeology until recent years, because of the fact that the roman and the medieval age were considered the most important Italian history. Consequently, the vintage train was not seen as a relic of history, but rather as an old scrap, just as the old railways that were being closed to the traffic have been considered ruins23.

These are the reasons why both railwaymen and local administrators as well as railway enthusiasts have a tepid concept of a preserved railway. In fact, attempts to build a preserved railway are very rare all over Italy; in reality, rather than preserved railways they can be qualified as tourist railways; railways closed to the ordinary traffic, on which tourist trains run. A few of these lines belong to a local railway company, the Sardinian Railways, where the Trenino Verde della Sardegna runs usually during the summer, operated by normal rolling stock and for charter trains by old carriages and steam locomotives on the lines closed to normal traffic Mandas-Arbatax, Isili-Sorgono, Macomer-Bosa, Nulvi-Palau.

Other local companies sometimes run historical trains on rail tracks open to the ordinary traffic, such as the Domodossola-Locarno railway near the Swiss border, the Genua-Casella company, the LFI on the lines Arezzo-Stia and Arezzo-Sinalunga.

The Ferrovie dello Stato (by their company Trenitalia) run some vintage trains all over the peninsula on lines opened to normal traffic, both charter trains and trains realized every year on some occasions, for example village festivals.

Moreover the Ferrovie dello Stato operate two tourist lines, closed to normal traffic, where a tourist service runs for a few days a year with the support of two voluntary associations. These associations are joined in the Italian Tourist Railway Association (Ferrovie Turistiche Italiane). The two lines are called: The 23 S. Maggi, In treno per diporto. Dal turismo ferroviario alle ferrovie turistiche. Esperienze e prospettive, Siena, 1997, p. 34.

• 439

Railroads in historical context: construction, costs and consequences

440 •

Railway of Basso Sebino (ancient name of the Lake of Iseo) located near Milan in the northern part of Italy, 10 km long; and the Val d’Orcia Railway, located in Tuscany, 52 km long from Asciano to Monte Antico.

In the latter, the Val d’Orcia Railway in Tuscany, the “Nature Train” runs.

The line was opened in 1872 and closed to the normal traffic in 1994.

The starting point to accomplish a tourist train service on the Siena-AscianoMonte Antico-Buonconvento-Siena line dates back to 1990, when on this route a rail ring of particular interest was discovered, in a period when an environmental sensitivity began to spread. The tracks run right into the woods and along river embankments into areas that cannot be reached with roads in the place where the Natural, Artistic and Cultural Park of the Val d’Orcia had already started to be designed. The Siena Provincial Council resolved to finance what appeared as an innovative enterprise in the Italian panorama; by implementing a new way of combined travelling: train and trekking, the so-called “Nature Train”.

Unfortunately, soon after the opening ceremony in September 1991, the project was abandoned, until the most interesting section of the rail ring between Asciano and Monte Antico was closed to the ordinary rail service in September 1994. The protests that were aroused by the closure of the railway line moved the Siena Provincial Council and the State Railways to re-launch the Nature Train project, which originally was a charter train, managed by a travel agency.

The system that was achieved without a plan or an international comparison did not, however, meet the coveted success and therefore in 1996 began its activity the voluntary association called Val d’Orcia Railway (Ferrovia Val d’Orcia); by means of an agreement with the State Railways it began running a service similar to the one of the Basso Sebino Railway in northern Italy. In Siena was put together a group of volunteers, who were neither environmentalists nor railway enthusiasts like those of the Basso Sebino Railway. They were, instead, retired railwaymen, actually, to be precise, railwaymen in early retirement owing to the State Railways restructuring process. Therefore, since 1996 the Nature Train was characterized by the implementation of a regular service in some nonworking days during the months of May, June, September, October and November on the Asciano-Monte Antico route, called Val d’Orcia Railway24.

Ever since the service has been operating by 50 year old diesel-rail cars, and occasionally by steam engines with vintage wagons dating back to 1910s and 1920s. The timetables have been arranged in order to make it possible to arrive by ordinary trains at Siena, Asciano and Monte Antico, so that the tourist train can be reached from the large urban areas of Florence and Rome, and from the seaside around Grosseto.

The cost of the trains is partly borne by the Siena Provincial Council with 24 Travelling on the Nature Train. An Historical and landscape guide with eight travelling itineraries, edited by Stefano Natura, Siena, Nuova Immagine, 2004, pp. 9-10.

Stefano Maggi

a daily grant, because the service is operated by normal State railwaymen, and therefore it is impossible to cover the costs with the takings. However the Nature Train covers about a 50% of its costs by its fares, which is much more than the branch lines percentage, usually limited to a 30-35%.

The project’s philosophy that is also stated by the name given to the line, “Val d’Orcia Railway”, and not only to the trains, is to give life to a proper tourist and preserved railway, instead of running some trains with no specific connection to the infrastructure.

The Nature Train is still operating in the year 2012, with passengers also from abroad, after seventeen years.

• 441

Railroads in historical context: construction, costs and consequences

REFERENCES Journal article Maggi, S. (2011), Treni d’epoca. Ritorno al futuro, in Rivista del Turismo, n. 4, pp. 10-16.

Degl’Innocenti, M. (1993), La villeggiatura: un diritto per tutti. In Il Risorgimento, XLV, n.

2, pp. 205-222.

(1932) Il magnifico risultato dei treni popolari di quest’anno, Le Vie d’Italia. Notizie ed Echi, XXXVIII, n. 11, pp. 417-418.

(1931) Importanti innovazioni in materia di facilitazioni ferroviarie adottate sulla rete statale, Le Vie d’Italia. Notizie ed Echi, XXXVII, n. 9, pp. 383-386.

(1931) Biglietti di andata-ritorno domenicali e festivi, Le Vie d’Italia. Notizie ed Echi, XXXVII, n. 8, p. 342.

442 •

(1931) Nuovi biglietti circolari turistici, Le Vie d’Italia: Notizie ed Echi, XXXVII, n. 8, p.

342.

(1931) L’istituzione di un treno speciale per sciatori, Le Vie d’Italia: Notizie ed Echi, XXXVII, n. 3, p. 113.

(1932) Facilitazioni di viaggio per le stazioni balneari e termali, e per l’Alto Adige e il Cadore - I viaggi delle famiglie, Le Vie d’Italia: Notizie ed Echi, XXXVIII, n. 7, pp.

274-280.

Tajani, F. (1917), Le ferrovie dello Stato e il movimento dei forestieri, Le Vie d’Italia, I, n.

3, pp. 149-154.

Ferraris, M. (1913), Di un ufficio di Stato per il movimento dei forestieri, Nuova Antologia, fasc. 985, pp. 146-151.

Gerelli, A. (1919), Le ferrovie di montagna nei dintorni di Bolzano, Le Vie d’Italia, III, n.

11, pp. 661-668.

Tedeschini-Lalli, E. (1927), Per l’estetica delle stazioni ferroviarie, Le Vie d’Italia, XXXIII, n. 2, pp. 153-158.

Ferraris, M. (1912), Per le industrie termali e climatiche d’Italia, Nuova Antologia, fasc.

968, pp. 697-703.

Stefano Maggi

Book Maggi, S. (edited by) (2004), Travelling on the Nature Train. An historical and landscape guide with 8 travelling itineraries, Siena: Nuova Immagine.

Maggi, S. (2003), Le ferrovie, Bologna: Il Mulino (new edition 2007).

Battilani, P. (2001), Vacanze di pochi, vacanze di tutti: L’evoluzione del turismo europeo, Bologna: Il Mulino (new edition 2009).

Maggi, S. (1997), In treno per diporto. Dal turismo ferroviario alle ferrovie turistiche.

Esperienze e prospettive, Siena.

Guadagno, W. (1991), Ferrovie ed economia nell’Ottocento postunitario, Roma: Cafi.

Venè, G.F. (1988), Mille lire al mese. Vita quotidiana della famiglia nell’Italia fascista, Milano: Mondadori.

Brilli, A (1986), Viaggiatori stranieri in terra di Siena, Roma: De Luca.

Vota, G. (Ed.) (1954), I sessant’anni del Touring Club Italiano, Milano.

Touring Club Italiano (1905), Guide di linee ferroviarie. Milano-Genova, Milano.

Chapter Di Mauro, L. (1985), L’Italia e le sue guide turistiche dall’Unità a oggi. In Storia d’Italia, Annali, V, Il Paesaggio (pp. 369-428), Torino: Einaudi.

Mila, M. (1965), Cento anni di alpinismo italiano. In C.E. Engel, Storia dell’alpinismo (pp.

249-353), Torino: Einaudi.

Report Ministero delle Comunicazioni. Amministrazione delle Ferrovie dello Stato (1932), Relazione per l’anno finanziario 1931-32, Roma: Istituto Poligrafico dello Stato.

Ministero delle Comunicazioni. Ferrovie dello Stato, Relazioni per gli anni finanziari 1927/29, p. 9; 1932/33, p. 15; 1933/34, p. 12; 1934/35, p. 12.

• 443

Günter Dinhobl

RAILWAYS HERITAGE. AN OVERVIEW PATRIMÓNIO FERROVIÁRIO: UMA VISÃO GLOBAL. Günter Dinhobl (ÖBB-Infrastruktur AG, Austria Áustria) Guenter Dinholb works in the department of R&D of the Austrian Federal Railways, the infrastructure company. He published several books about financing and cultural aspects of railways, as well as papers related to railways and heritage preservation issues.

He is a member ICOMOS, TICCIH, International Association for the History of Transport, Traffic and Mobility (T2M) and International Railway History Association (IRHA).

Gunther Dinholb trabalha no departamento de I&D da OBB, a empresa federal de infraestruturas de caminhos de ferro na Áustria. Publicou diversos livros sobre financiamento e aspetos culturais do caminho de ferro, assim como numerosos artigos sobre património ferroviário e sua preservação e reutilização. É membro do ICOMOS, TICCIH, International Association for the History of Transport, Traffic and Mobility (T2M) e International Railway History Association (IRHA).

Abstract Resumo

The paper introduces “Railway Heritage” and gives an overview of the most discussed and most important thematic areas – and (open) questions. A first statement on “what is railway heritage” leads to the guiding question of the session.

Railways were (and are) built for daily use since more than 170 years. They have supported the industrial revolution like no other transportation technology. In this time period rolling stock, buildings and railway routes were built and used, but also maintained and renewed. If this framework changes, railways are in danger of being closed, dismantled or destroyed. This situation is why we can look at railway heritage today.

The paper gives an overview of the concept of ‘using, reusing and preserving’ which was first presented at ICOHTEC-TICCIH-conference in summer of 2010 in Tampere/ Finland. It was discussed there and in spring 2011 presented as an extended version at VDI in Berlin/Germany. The backbone of the concept of ‘using, reusing and preserving’ is to identify the different actors and their different interests in railway heritage. This leads towards a ‘railway heritage matrix’ as a tool to identify, to clarify – and to develop how to handle railway heritage.

This structural approach of the paper is illustrated with examples of UNESCO World Heritage Railways, Museum railways and museums which show us the rich railway heritage, but also with examples of protected railway heritage and everyday rail operation. Special focus will be given on the World Heritage site ‘Semmering

• 445

Railroads in historical context: construction, costs and consequences

Railway’ in Austria, a prototype railway of the 1850s which is still in use as main line with heavy traffic.

446 •

Estra trabalho trata de património ferroviário e apresenta um sumário das áreas temáticas mais importantes e mais discutidas – assim como das questões em aberto. Começa-se por discutir o próprio conceito de património ou herança ferroviária.

Os caminhos de ferro foram construídos para uso diário à mais de 170 anos.

Suportaram a revolução industrial mais do que qualquer outra tecnologia de transporte. Durante esse período o equipamento ferroviário, os edifícios e as linhas foram construídas e usadas, mas também sofreram manutenção e renovações.

Se este quadro se altera, os caminhos de ferro correm o risco de serem fechados, desmantelados ou destruídos. É por isso que hoje em dia se tem uma preocupação com o património ferroviário.

Discutem-se os conceitos de “uso, re-utilização e preservação”, antes apresentados numa conferencia ICOHTEC-TICCIH no verão de 2010 em Tampere (Finlândia).

Uma nova versão mais desenvolvida foi apresentada no VDI, em Berlim (Alemanha). No centro do conceito de “uso, re-utilização e preservação” esta a identificação dos diferentes atores e os seus diferentes interesses específicos no património ferroviário. Define-se uma matriz de património ferroviário, como uma ferramenta para identificar e para clarificar – assim como também tratar esse património.

Este modelo teórico é ilustrado com exemplos de caminhos de ferro classificados como património mundial pela UNESCO, museus ferroviários, e museus com um rico conteúdo ferroviário, mas também com exemplos de património ferroviário protegido em operação diária. Trata-se em especial o caso, classificado como património mundial, relativo ao caminho de ferro de Semmering, na Austria, uma linha protótipo de caminho de ferro dos anos 1850, que ainda hoje continua em exploração como linha principal e com um elevado trafego.

Günter Dinhobl

Railways Heritage. An overview Günter Dinhobl

INTRODUCTION This paper will give an input to the discussion on the history of railways and its heritage. The aim of the paper is manifold: on one hand it is to identify the characteristics of railways – depending of the time period they have been built – and on the other hand to establish a structure to make it easier to handle railway heritage.

ABOUT RAILWAYS It was in 1830, when railways had their breakthrough: in that year the LiverpoolManchester-Railway was inaugurated as the first railway in the modern sense: iron (steel) rails guide flanged wheels of iron, and finally mechanical power to move these carriages. Until today this principle remains the same. During the 19th century the railways became the most important industrialised land transport technology: cities and regions were connected and countries were unified by the help of railways. Technology was relatively stable during a lot of years, and in the mid-20th century – at the time when motorcars and aeroplanes came into daily life – the railways began its decline.

• 447

Railroads in historical context: construction, costs and consequences

At that time railways were something like an old technology from yesterday and only good for heritage but not for actual needs of society. But not at all: just at that time the opening of the high-speed-rail line Shinkansen (Japan) in 1964 started the ‘renaissance of railways’ all over the world. Impressive development took place until then – and the last milestone of this development was the trial run in France with a speed of 574 km/h! – several years ago it was not believed that a conventional rail system would be able to go so fast.

448 •

This aspect of the beginning of the railway age and today illustrate quite well what happened during the last century: it was a permanent development for more than 170 years of railways.

Steam power was replaced by electro- and Diesel-powered engines, bodies of the carriages changed from wood to steel. Also in the area of railway infrastructure a change of materials can be identified: for example, the early bridges were made of stone or brick and in the second half of the 19th century often also of steel. More traffic resulted in higher loads and therefore bridges were strengthened and/or renewed. Finally, today’s bridges are built of concrete or steel - or using both materials. Station buildings were made of stone and brick, while nowadays they are a mix of concrete, steel and glass architecture. At the track itself, wooden sleepers are no longer used in new lines anymore – concrete sleepers or ballast-less systems for high speed rail are now the state of the art in railway infrastructure technology. And in the field of signalling – essential to guarantee the high safety of railways – started with basket signals in the 1830s and changed to form- and later to light signals. Finally the future will bring in Europe the ETCS-System which shall be implemented at least in the main European corridors.

Günter Dinhobl

Beside technical issues there have been always a political, economical and/ or cultural frame which brought other demands to the railways. These issues resulted in the need for adoptions of existing buildings or completely new buildings and lines – or the closing and destruction of railway lines. At existing lines changes also happened – for example see a detail of the Semmering-Railway panorama which was published in 2010: when the line was opened in the beginnings of the 1850s there was no railway station at the village of Küb. The building of a train stop with a simple waiting room was done around 50 years after opening of the Semmering line in 1899. This waiting room was made of wood and 60 years later this waiting room was renewed: the proportions stay near the same but the building was built now of concrete.

All these examples illustrate what railways are: railways are a complex system for the transport of people and goods which were – and are – developed constantly.

More than 170 years of railways leave numerous remains which are worth to be preserved: Future needs origin. But it was a long way that railway heritage became what it is today. The Venice Charta – which is an important document

• 449

Railroads in historical context: construction, costs and consequences

for heritage – declared in 1964: “the conservation of monuments is always facilitated by making use of them for some socially useful purpose”.

Under these circumstances railway Heritage can be distinguished into three areas: > ‘using‘: useful for travelling and freight transport – eco-efficient > ‘re-using‘: useful for getting experience and feeling > ‘preserving‘: useful for education and identity of people Finally, handling the railway heritage is linked up with different actors and the following examples should illustrate these three areas.

USING, …

450 •

The Semmering-Railway in Austria is in operation since 1854 and since 1998 it is listed as a UNESCO-World Heritage Site. But what is the worldwide importance of that railway line? – lets have a look back: in the 1830s Vienna was the capital of the Habsburg empire in middle and eastern Europe. The transport from and to Vienna was done by boat in the east-West-direction on the river Danube. In the north-south-direction it was more difficult, because the roads have had limited capacity. Even in the late 1790s an inland waterway was planned from Vienna to Trieste, the only harbour of the Habsburg Empire. The first 50 km were built and then, in the 1830s, a new land transport technology emerged: the railway. So it does not surprise that the first railway have been built in northern and southern directions: the Kaiser-Ferdinands Nordbahn made accessible the northern parts of the empire since 1838, and the Wien-Gloggnitzer Bahn made the same for the southern parts since 1841/1842. The southern route was extended by the state beginning in 1841 with the final destination Trieste.

And the Semmering line was part of this long-distance railway line from the capital Vienna to the port of Trieste. In the case of Semmering, it was the first time a railway passed a high-mountains area (Alps). The chief engineer was Carl Ghega. He was born in Venice and after university he got experimental knowledge in mountain road building. In 1842 he was sent to England and North America by the state to identify the latest news in railway technology

Günter Dinhobl

Ghega combined his personal knowledge and the knowledge gathered together at the study tour and planed a railway line between Gloggnitz and Mürzzuschlag which doubles the length compared with the direct way: instead of 21 km Ghega used side valleys to build a railway to climb up the Semmering pass: serpentines guarantee to make possible the exclusively use of steam locomotives – while the usual technology at that time were cable inclines with fixed steam engines. The line was a big success and still today the main rail traffic from Vienna southwards – to the Austrian provinces Styria and Carinthia – runs over the Semmering. That means that today approx. 10 Mio. freight tons a year (netto-tons, ALPINFO / CH) passes the Semmering line. Freight and passenger traffic together leads to 170 trains a day in average and therefore the Semmering route is one of the most frequented railway lines that passes the Alps.

• 451

Railroads in historical context: construction, costs and consequences

But what is the importance of this line, in particular why was it listed as UNESCO-World Heritage? – the aspects of route design and the exclusively use of steam locomotives were mentioned above; this resulted in a locomotive competition in 1851 to find the most efficient engine for the Semmering line. Four locomotive factories participated – from Belgium, Germany and Austria. The winner was the company Maffei with its engine named ‘Bavaria’ and Ghega summarised: ‘now we have locomotives which are more powerful than we really need’. In the end, none of these engines was good enough for the daily railway operation and that is why a special construction style was done by Wilhelm Engerth: the so called ‘supporting tender’ was the invention of the time and at first introduced at Semmering. In later times locomotives with this construction style were in use all over Europe.

452 •

Another outstanding value of the Semmering line are huge viaducts – overall there are 16 which have a length up to 228 m, a height up to 46 m and are mostly situated in curves which was at that time a novelty. Also tunnels had been built, there are 15 with a maximum length of 1.500 m – also here there are some of them with curves inside which was difficult to build because of the need of higher precision in surveying methods.

Günter Dinhobl

All this resulted in a railway line in a surrounding which was recognised more and more with every train ride: finally the railway line generated views onto the region, onto the landscape. Also ‘standardised’ views of the building of the railway line embedded in the landscape were produced – the most important is the so called ’20-Schilling’- view which recalls a view printed on an Austrian banknote which was in use between the late 1960s and the 1980s. The railway company recognised this potential and established a special touristic traffic on the line. Touristic infrastructure like hotels and hiking trails, and later a golfcourse and an alpine open-air swimming bath were built to attract tourists. And the high-society took Semmering as the summer destination, where numerous villas were built and this resulted in the nickname ‘High-Vienna’ for the Semmering village. Overall, the railway enabled the touristic development of the region – in short: the railway produced a cultural landscape.

The railway itself was operated by steam engines since the beginning in 1854 for more than 100 years in daily use. In 1959 the electrification was finished and

• 453

Railroads in historical context: construction, costs and consequences

454 •

today all trains are operated without smoke. The capacity of the line has become higher and higher and today the heavy freight trains need up to three engines – up to 18.000 HP – that pull and push the freight trains up the incline to pass the Semmering line.

Günter Dinhobl

Beside these express trains and local trains pass the route with nearly the same speed – 50 to 70 km/h. On some days of the year special trains – nostalgic steam trains or some with more modern, but also historic diesel or electric engines are set into operation. These offer special experience and these trains are nowadays well accepted.

Another example of ‘using’ a railway heritage is the so called ‘Vorortelinie’ (suburban-line) in Vienna. It was one of three routes of the ‘Stadtbahn’ (city lines) which was designed by the architect Otto Wagner in last years of the 19th century. The station buildings were done in the secessionistic style and the ‘Vorortelinie’ had to pass mountainous terrain and that is why there are several tunnels, dams and bridges at this line. After the second World War the ‘Vorortelinie’ was closed for passenger transport and in 1987 the Austrian Federal Railways renovated the line and reopened it. Today there are trains every 15 minutes during the day and the line is quite well accepted by the people. In the night time there are some freight trains that pass the ‘Vorortelinie’ – but with the problem of railway noise in living areas.

• 455

Railroads in historical context: construction, costs and consequences

… RE-USING, …

In contrast to the ‘Vorortelinie’-example the re-use of railways took place when railways, e.g. rolling stock (steam/diesel/electric or carriages) were re-used for different tasks. While rolling stock had its original function in a daily use, today several of the historic rolling stock is re-used only some days a year. One example in Austria is the Museumstramway Mariazell: the railway line was built completely new in the late 1980s from a famous (and touristic) place of pilgrimage to a small lake nearby. The rolling stock is put together from different locations in Austria. Electric and steam tramways are in use, but also diesel engines of branch lines to factories. Overall, at the Museumstramway Mariazell there is a re-use of historic light-railways on a completely new line only for touristic purpose on weekends and during holidays.

456 •

In the case of the Heizhaus Strasshof there is a listed locomotive shed – the biggest still remaining in Austria – where a association works on renovation and restoration of historic locomotives. The main focus is still on steam engines, but there are also diesel and electric engines which can be set in operation. But on the other hand, there is numerous historic rolling stock which is still not renovated – and the damages are quite high that a renovation will be unrealistic. The Heizhaus Strasshof offers with its locomotives also one-day train rides in the eastern parts of Austria and organises special days for families or locomotive meetings in the area of the locomotive shed.

Günter Dinhobl

Another example of re-using is the railway which passes the UNESCO World Heritage Wachau valley in Austria. Until 2010 this line was in daily operation – so it was used – but at the end of 2010 it was transferred from Austrian Federal Railway to the province of Lower Austria. The province had no railway infrastructure concession and that is why the line has to be closed before this transfer.

Today it is only a touristic railway which is now operated only on weekends in summer – while the daily transport of children to school has to be done by bus.

And only one month after the closing of the daily railway operation a high water of the Danube river flooded the street but not the railway – but this was closed because of the missing infrastructure concession.

• 457

…AND PRESERVING… The third way to handle railway heritage is ‘preserving’: this is the ‘classical’ way and every museum is doing that: for example the Technical Museum Vienna preserves several old rolling stock – locomotives of the early 1840s, a restored – and in some parts rebuilt – passenger carriage of the ‘Vororteline’ of Otto Wagner, or the saloon car of empress Elisabeth.

Railroads in historical context: construction, costs and consequences

… - IN COMBINATION: Other museums – like the Südbahn-museum in Mürzzuschlag at the Semmering line – link re-using and preserving: while historic buildings, which are also listed, are adopted to the new use, other items are preserved: the huge collection of gang cars is in a quite new preserved condition while other items are preserved in the condition of their last state: so you also can see a ‘flagged’ – and rusty steam locomotive with the flag of former Yugoslavia on the wind shields.

458 •

Sometimes using, re-using & preserving come together in one place. So, for example in Payerbach, also along the Semmering line: while new ‘city shuttle’ trains go every half hour to and from Vienna, a narrow gauge railway line starts its way to a side valley from one station end. It was built for freight issues, but passenger transport started in 1926 and ended in 1963. Freight transport took

Günter Dinhobl

place until 1982 and since that time a ‘re-use’ touristic railway operation started on weekends. Also within the area of the Payerbach railway station there is a museum park with a Semmering steam locomotive, an electric locomotive of the narrow gauge branch line, a historic bus and a cab of the Rax mountain cable car.

• 459

PROBLEMS & CHANCES OF USING Using of Heritage Railways is always linked with problems and chances: problems because of the daily use, which result in wear, renewing and today’s standards for building and railway operation. But there are also sometimes chances, e.g. to make detailed surveys of infrastructure buildings: The example to show

Railroads in historical context: construction, costs and consequences

460 •

this is taken from the Semmering Railway: the arched bridge ‘Adlitzgraben’ was built in 1851-1853 and the ‘problem’ is to get a detailed surveying to guarantee today’s safety, while the ‘chance’ is to include construction history of this building. In 2010 a surveying by laserscanning was done by the Technical University of Vienna on behalf of Austrian Federal Railways.

Also a research was done to compile the history of that building with the help of plans and other documents which are available at the Austrian State Archive. With that information backward-editing was done to show the permanent

Günter Dinhobl

changes of this building – lets go back in history: in 1959 the overhead wire was mounted, in 1938 a final strengthening of the viaduct was done, in particular at the arches. In 1911 one final side pillar was build, 1873 two side pillars on the left side were built and in 1855 the three middle side pillars and the strengthening inlays of the arches were built.

The construction history, and in particular the documents still available of the Austrian State Archive gave light to this interesting building: even in 1854, the year when the daily operation began, damage was recognised: damage documentation can be found in the State Archive which show for each arch the cracks and done in a very precise way. Further documents, which are still available, are the plans of variants: for example, one suggestion was to rebuild the viaduct to a simple double storey viaduct – but this suggestion was never realised. It only exists as memory of the Semmering line which never exists physically.

• 461

Railroads in historical context: construction, costs and consequences

CONCLUSION All these examples illustrate that there are only three aspects which are very important when someone works with railway heritage: knowledge, knowledge and knowledge!

At first, it is the knowledge of the history of the railway objects itself; these can be implemented e.g. as an inventory. The second dimension of knowledge is the specific knowledge of the techniques to handle the Railway Heritage; this can be knowledge of maintenance, but also the knowledge of daily railway operation.

And the final ‘knowledge’-aspect is dedicated to the meaning and perception of railway objects, e.g. as economic/social/cultural history of railways.

Finally, it is important to distinguish between these three levels of railway heritage knowledge which forms the basis of our living with the railway heritage – when it is in use, when it is re-used and/or when it is preserved.

462 •