Urban form at the edge : proceedings from ISUF2013. Volume 2 by Linda Carroli

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Urban Form at the Edge: Proceedings from ISUF 2013 is published in two volumes. This volume is the second. Volume I was published in 2014. Editors: Paul Sanders, Mirko Guaralda and Linda Carroli ÂŠ Contributing authors, 2016 This publication is copyright. Apart from fair dealing for the purpose of private study, research, criticism or review, as permitted under the Copyright Act 1968, no part may be reproduced by any process without written permissions. Enquiries should be made to individual authors. All contributing authors assert their moral rights. First published in 2016 by: Queensland University of Technology (QUT) Creative Industries Faculty â&#x20AC;&#x201C; School of Design in conjunction with International Seminar on Urban Form GPO Box 2434 Brisbane, QLD 4001 Australia QUT School of Design: https://www.qut.edu.au/creativeindustries/about/about-the-faculty/school-of-design ISUF: http://www.urbanform.org ISBN 978-1-925553-01-7 National Library of Australia Cataloguing-in-Publication entry Creator: International Seminar on Urban Form (20th : 2013 : Brisbane, Queensland) Title: Urban form at the edge : proceedings from ISUF2013. Volume 2 / Paul Sanders, Mirko Guaralda, Linda Carroli editors. ISBN: 9781925553017 (ebook : volume 2) Subjects: Cities and towns--Congresses. Sociology, Urban--Congresses. Urban anthropology--Congresses. Other Creators/Contributors: Sanders, Paul, editor. Guaralda, Mirko, editor. Carroli, Linda, editor. Queensland University of Technology. School of Design. Dewey Number: 307.76 All opinions expressed in this publication are those of the authors and do not reflect views of the publishers and their associates. Permission to republish images is the responsibility of authors, who have confirmed they have secured appropriate permissions to republish any images appearing in their essay.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Contents Acknowledgements

Introduction Paul Sanders, Mirko Guaralda, Kai Gu, Tony Hall and Leigh Shutter

CITIES ON THE EDGE | cities on edge conditions Morphological study of a hilly urban fringe landscape: the case of Bilbao Maider Uriarte Idiazabal

Dwellings in the borderland: a study of traditional hill settlements in the Chittagong Hill Tracts of Bangladesh Monojit Chakma, Mirko Guaralda and Paul Sanders

Planned towns in the ‘Conquered Territory’ between Basutoland (Lesotho) and the Boer Republic of the Orange Free State (Free State Province of South Africa), 1867: balancing security with inherited cultural traditions and townscapes Walter Peters

OFF CENTRE | urban form in emerging economies and postcolonial countries Urban plot characteristics study: Casing Centre District in Nanjing, China Lina Zhang and Wowo Ding

The Denied City: how the crisis is leading cities to the edge Nicola Marzot

ON THE EDGE OF THE CITY | peripheral areas and urban form in suburbia Looking for the mall: public life in the city of dispersal Luisa Bravo

The impact of social housing on the urban form in the suburbs of Shanghai Dong Yijia and Li Zhenyu

The urban village code in Guangzhou: morphological self-evolution on the edge of the metropolis Quanle Huang and Tao Li

The evolution of urban tissue and building types in the Guangfunan area of Guangzhou, from 1840 onward Ye Li and Pierre Gauthier

108

Xi Guan: edged community in the very heart of the city, the decline of Guangzhou’s traditional centre Tao Li and Quanle Huang

125

The compact city neighbourhood: an emerging new stereotype and model to redevelop the industrial fringe of the historical European city and develop new sustainable suburbs Todor Stojanovski

135

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

REGIONAL CENTRES | cities and towns with local importance, at the edge of national or regional urban networks The evolution of the urban grid of Indus Civilisation: case studies of Mehar Garh, Mohan Jo Daro, Cave City ‘Rakas Jo Rohro’ at Rohri and ‘Sat Gharhi’ at Sehwan, Sindh, Pakistan Javeria Shaikh

156

PUSHING THE EDGE | new technologies and new techniques A quantitative approach linking street spatial configuration to street pattern Wu Zhouyan

170

Shape grammar as a method for describing order of Chinese traditional urban pattern Guoxin Ke

187

Biographies

200

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Acknowledgements ISUF 2013 Conference Organising Committee Paul Sanders, QUT Mirko Guaralda, QUT Kai Gu, University of Auckland Tony Hall, Griffith University Leigh Shutter, Griffith University Scientific Committee Paul Sanders, Mirko Guaralda, Kai Gu, Tony Hall, Leigh Shutter, Teresa Marat-Mendes, Vitor Oliveira, Michael Barke, Jeremy Whitehand, Marco Pompili, Nicola Marzot, Ivor Samuels, Peter Larkham, Gordon Holden, Rosie Kennedy, Anoma Kumarasuriyar, Kathi Holt-Damant, Claudia Taborda, Emilio Garcia, Milica Muminovic, Gillian Lawson, Ian Weir, Philip Crowther, Veronica Garcia Hansen, Debra Cushing, Luisa Bravo, Damrongsak Rinchumphu, Jaz Choi, Marcus White, Nicola Marzot. Graphic Design and Cover Art Manuela Taboada Conference Administration Emma White, Jenny Greder, Maree Francis

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

[This page is intentionally left blank]

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Introduction Urban morphology as a field of study has developed primarily in Europe and North America, and more recently emerging as a recurrent topic in China and South America. As a counterpoint to this centric view, the ISUF 2013 conference explored aspects of ‘urban form at the edge’. In particular the conference examined ‘off centre areas’ such as India, Africa, Middle East, Central Asia and Australasia which require innovative approaches to the study of traditional, as well as post-colonial and contemporary, morphologies. Broader interpretations of urban form at the edge focus on minor centres and suburbia, with their developing and transilient character; edge cities and regional centres; and new technologies and approaches that are developing alongside established methods, tools and theories of urban morphology. Sub-themes for the conference, which comprise the sections of volumes 1 and 2 of the ISUF2013 proceedings, were:      

Cities on the Edge – cities on edge conditions, such as natural limits or political boundaries Off centre – urban form in emerging economies and postcolonial countries On the Edge of the City – peripheral areas and urban form in suburbia Edge Cities – new urban conditions Regional centres – cities and towns with local importance, but at the edge of national or regional urban networks Pushing the Edge – new technologies and new techniques.

Although Australia has historically been considered at the edge of the world due to its location, the conference will take advantage of its relative proximity to Africa, India and South East Asia, especially targeting the seminar to these geographical areas, and directly addressing the challenge for ISUF to develop into these continents. The South East Queensland region incorporates both Brisbane and its neighbour the Gold Coast City. It is the fastest growing metropolitan region in Australia. The rapid processes of urban transformation have brought about challenges that are comparable with the experience of many developing nations. We would like to thank QUT students and staff who volunteered their time before and during the conference; without their support the event would not have been possible. Conference Organising Committee Paul Sanders, QUT Mirko Guaralda, QUT Kai Gu, University of Auckland Tony Hall, Griffith University Leigh Shutter, Griffith University

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

[This page is intentionally left blank]

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

CITIES ON THE EDGE cities on edge conditions

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Morphological study of a hilly urban fringe landscape: the case of Bilbao Maider Uriarte Idiazabal School of Architecture of Donostia-San Sebastian, University of the Basque Country, Spain Email: maider.uriarte@ehu.eus

Abstract. The paper examines whether the form and character of urban fringes located on mountain slopes can be understood by using an urban form analysis method. The objective is to locate homogeneous areas within the fringe that might respond to a morphogenetic explanation taking into account that these areas represent a specific character of their own, not resembling a consolidated urban area, but rather a hybrid landscape. An in-between area from Bilbaoâ&#x20AC;&#x2122;s mountainous urban fringes is analysed by employing urban form analysis elements such as building, plot, open space and street, along with plot/building, street, city and region resolution levels, as well as the time variable. The findings reveal that it is possible to demarcate four different areas: Clusters, Isolated Plots, Infrastructure and Open Space. However, these do not seem to fit formally, and are mostly embedded in open space and structured by the highway. Nevertheless, several problems have been found that impede a comprehensive analysis. In particular, since several built elements are omitted from the process of analysis, the urban/rural edge is perpetuated. Also the effects of the mountain slope on the overall character of the urban fringe are not considered. Keywords: Bilbao, urban fringe, infrastructure, open space, landscape

The urban edge has always existed (Thomas, 1990) and its study involving many approaches and nominalisations as reviewed in the works of Meeus and Gulinck (2008), Thomas (1990), and Addell (2000). These reviews point out the variety of perspectives addressing the topic of the urban edge. This can also be interpreted as a sign of their complexity, rather than a liability, as well as indicating how multiple urban situations shape just as many urban edges. Simultaneous to this differentiation and diversification of the urban condition and its edge, discussions in urban studies have claimed the end of the urban-rural dichotomy by coining a variety of concepts to describe, quantify and identify the urban edge and new urbanities. Such conceptualisations can be grouped into three main categories. First, there are claims for the emancipation and dispersal of new urban centres such as Soja (2000), Garreau (1992) and Indovina (2005). Second, there are claims for the existence of an in-between landscape, such as Rowe (1992) and Sieverts (2003). Third, there are those working on a global scale of totalities, namely Brenner and Schmidt (2011). The urban fringe is no longer embodied or represented by a definite form, but is formed by an array of heterogeneous elements and land uses (Ipsen and Weichler, 2005), and has become a landscape with its own character. If urban fringes are more various and diverse than ever in the light of a global urbanisation, and present a landscape characterisation of their own, are traditional morphological analysis tools capable of identifying the unique form and character of the urban fringe? Is it possible to identify specific morphological areas and consider a morphogenetic study within the fringe landscape? Could other analysis methods be useful for the purpose of understanding these pervasive and ubiquitous, but at the same time specific landscape character and forms? In the text that follows, a morphologic analysis is carried out on the urban fringe of Enekuri area located on one of the mountain slopes that structures the Greater Area of Bilbao (Basque Country, Spain). The

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

study forwards the limits of the morphological method based on built form and the sequence of spatial resolution analysis, suggesting the adaptation of the method of analysis and variables in order to understand the character and form of the urban fringe landscapes.

Methodology Case study Bilbao The greater area of Bilbao, with its 35 municipalities, is clearly structured by two main geographical features; first the estuary and the sea act as the commercial and industrial backbone, and second the three parallel mountain ranges divide the area in two main valleys. The southern feature contains the estuary and the main urban centre of Bilbao, while the northern one houses the less dense Txorierri area. These mountains, while not of alpine character or altitudes, do comprise topography of more than hills with altitudes ranging from 200m to 900m. The estuary has known numerous transformations; ranging from the earliest project in the 19th century to channel its course so as to guarantee navigation, to the most recent post-industrial reconversion. That is, the water edge has been defined and retraced frequently and each new transformation has been communicated and celebrated publicly. The estuary is, and presumably will always be, the most meaningful urban scene of Bilbao. Green open space, agricultural fields and forests, along with buildings and infrastructures are necessary elements for analysing the form of the urban edge and their inner structures. Morphological basic cells such as pastures, forests, quarries and mines belong â&#x20AC;&#x153;naturallyâ&#x20AC;? to the slopes of Bilbao. While other elements, such as antennae, landfills and reservoirs find opportunities that the geomorphology creates to enhance their function. The slopes of Bilbao house an array of elements that is functionally, formally, and historically diverse and heterogeneous. This set of elements characterises any given urban fringe landscape, but nevertheless shows signs of specificity and local characteristics articulated by the mountain slopes. The objective has been to apply the urban morphological analysis tools to the urban fringe of Bilbaoâ&#x20AC;&#x2122;s mountain slopes in order to understand their logics of formation and character. This study particularly focuses on the case of Enekuri (Fig. 1) as an example of a common morphological growth area of the urban fringe of Bilbao.

Figure 1. Map showing the analysed Enekuri area Source: Eusko Jaurlaritza / Gobierno Vasco. geoEuskadi with changes made by the author

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Enekuri is an in-between area located around the administrative boundaries between Bilbao and Erandio municipalities on a relatively high altitude of approximately 90m considering its close proximity to the sea. Its transition character is marked from long ago since the area has acted as an entryway to Bilbao from the Txorierri Valley, and it still serves as a regionally important road transportation interchange point. Geomorphologic variety embodied by the Asua River tributary to Bilbao’s estuary, and the Altzaga and Enekuri-San Blas hills structure the area’s natural character. The strong interaction among industrial, residential, infrastructural and rural land-uses structure a case study that is morphologically diverse. Morphological analysis of fringe landscapes Moudon (1997) lists three elements that are fundamental to morphological analysis of the urban form: form, resolution and time. Of form, she mentions three elements: buildings and related open space, plots and streets. The second element is resolution and it covers four levels: building/lot, street/block, the city and the region. As for the last aspect, time, she claims that morphological analysis considers the relationship between history and urban form in order to understand and differentiate periods of morphological homogeneity (Moudon, 1997). In addition to these three basic elements, morphological analyses also focus on plan units; these units are homogenous and cohesive since they were built and/or transformed at the same time or in a similar way (Moudon, 1997). In short, traditional morphological methods build on the continuity factor or on the way that things “fit together”. The data used for such analysis is a combination of aerial photographs and GIS database analysis (Gobierno Vasco, 2015) combined with field trips using private and public transportation means. Buildings and open space, plots and streets are studied in order to identify homogenous morphological spaces. A morphological analysis considers the scales of analysis of plots, the street/block, the city and the region (Moudon, 1997) following a gradation of scales, but in this case, the work has chiefly focused on the lot and plot/street. It must be pointed out that the analysis takes place only at a plan level, that is, without taking into account any kind of section through the terrain. Therefore, the slope is intentionally not included in the method as a variable. The objective is to ascertain whether the use of form, resolution and time is sufficient to understand the specific character of the urban fringe on a mountain slope. The field work has shown that the following sets of functions can be found on the hills of Bilbao:     

local customs and cultural landscape related elements – farmhouses and chapels transportation infrastructure and service infrastructures economic and industrial activity related buildings open natural space residential and everyday life related elements – housing areas, education and sport facilities.

At the building and open space level of analysis many buildings of various sizes and forms are found, depending on their activity and land use: industrial, commercial and residential. In addition, schools and urban service related buildings can also be found. Various combinations of building typologies and related open space can be identified, although some repetition occurs such as in the case of industrial pavilions or single family homes. The size of an economic activity related building varies from a 22,400m 2 shopping mall to the smallest 1000m2 building. Detached single family housing ranges from 100 to 250m2, and residential blocks of 210 or 435m2 have been identified. However, the field work signals that there are further kinds of activities, such as service (a 746m 2 electrical substation building) and transport infrastructure, such as several gas stations of about 435m 2 distributed in three small pavilions. There are also educational facilities (a school structured by three buildings amounting to 5467m2), economic facilities (a junkyard with four buildings totalling 1148m 2 of construction), and religious buildings (an 897m2 church). These elements have also been included in the

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

analysis due to their prominent visual presence and seemingly spatial-structuring capacity. A total of 13 plots have been analysed (Fig. 2).

Figure 2. Overview of the 13 plots that have been included in the analysis Source: Eusko Jaurlaritza / Gobierno Vasco. geoEuskadi with changes made by the author

Following the building level analysis, the open space within the limits of the building plot is also analysed. Focusing on the three main activities, findings show that in each case the open space form and size is closely related to its functional needs and the size of buildings and plot. Bigger buildings, such as factories and commercial buildings, might have adjoining open space in the form of ornamental or residual green space, parking lots or access points. Residential buildings are also surrounded by open space such as gardens or squares and road and pedestrian access points that can be gated or not, depending if it is an isolated house or housing block. Expressed in percentages (Table 1), it is revealed that industrial and commercial plots tend to exhaust most of the capacity of the plot with buildings that take almost half of the plot’s space. Yet, in most stances, the plots are formed by open space with various buildings placed in various ways:     

placed in the centre – industrial and economic plots displaced to one side from the centre – single family housing against one side – church – or the corner of the plot – electrical substation scattered around the plot – gas station and junkyard clustered irregularly – school – or symmetrically – residential plot.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Along with the open space related to buildings, thereâ&#x20AC;&#x2122;s another kind of open space that is not directly related to a specific building or buildings of a significant size or volume. In addition to open space in the form of parks, other forms of open space have been identified including: agricultural land use, at times adjacent to a residential unit; open space used in economic activity such as slag sorting centres; open space of ex-landfills; and other smaller residual or large open space that are seemingly abandoned and covered in shrubs. These open spaces are significant and differentiated formal elements of the urban fringe. In addition, the open space might include other kinds of constructions such as high tension electricity pylons or the viaductâ&#x20AC;&#x2122;s supporting pillars. All of these elements are located in plots that are apparently pastures. Other than those constructions, the open space plots seem to be covered by thickets, shrubs, and overgrown small groups of trees. The spaces are sometimes crossed by dirt roads, and in other cases some terraced land formation can be interpreted from the aerial views. The open spaces also contain land uses other than seemingly agricultural uses, such as informal open air warehouses, small allotments and slag sorting infrastructure. Table 1. Comparative figures of plot sizes, built-up area and percentage of open space Example plots

Size of building (m2)

Size of plot (m2)

Percentage of open space (%)

Residential plot Enekuri Single family housing 1 Single family housing 2 Industrial pavilion Enekuri Industrial pavilion Asua Supermarket Car seller Junkyard Educational facility Electrical substation Gas station Water reservoir Church Enekuri

440m2 x 4 buildings = 1760 454 105 1835 16538 22379 5859 1447.8 5467 746 435 (three buildings) 227 897

5550 3765 1126 3100 38246 64246 15850 10258,08 70011 10165 2028 227 2987

70% 88% 90% 40% 57% 65% 63% 86% 92% 93% 78.5% 0% 70%

The following phase studies the way that independent plots are related to the streets or transit elements as well as to the adjacent plots. In this process several elements have been identified as plot formalising features: roads and green open space mostly, but also streets, railways, waterways and adjoining similar plots. In addition to the formalising elements, the way that the plot can be accessed has also been taken into account. Most plots are made accessible to vehicles through roads and streets via single entrance points except for the case of the gas station which has an entry and exit point from the highway. Pedestrian access is only formalised in cases where the plots have adjoining streets that include sidewalks, such as residential plot, industrial pavilion and church in Enekuri, where the road section includes a sidewalk as in the case of the supermarket, and where the plot is served by a footbridge over the highway as it happens with the school plot. As for the elements that limit the plots, each analysed unit has its own specific condition showing a variety of combinations: streets, secondary road and vacant plot, green open space and similar plot, industrial plot, river and railway, green open space, urbanised street and green open space, etc. It is interesting to find that mostly roads and the highway make up for the transit element usually exemplified by the street in more urban settings. Further analysis in order to identify the homogeneous areas covers the aspect of time. For that purpose, the Geo-Euskadi data base and earlier aerial views have been consulted at the Spanish National Geographic Instituteâ&#x20AC;&#x2122;s photograph library, where aerial views as old as 1946, 1956 (the so-called

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

“American flights”) and 1973 can be obtained (Instituto Geografico Nacional). After observing said images, the analysed plots can be organised in four different periods of development: 1940-1960s, 19701980s, 1990s and 2000s. Most plots were developed in the first period or earlier: single family housing plot 2 along with the adjacent houses, industrial pavilion in Asua, the water reservoir, the church and the electrical substation. The next period saw the construction of the gas stations, the junkyard, the school and the car seller. In the 1990s only the single family housing plot 1 was developed in an existing rural area with similar morphology, and the last period included the construction of the supermarket as well as the industrial pavilion and residential plot in Enekuri. In addition to the individual plot development, the construction and development of the main transportation element must be mentioned as well: the highway which traverses the neighbourhood. Although the road already existed as a way to connect the Txorierri valley and Bilbao, it was first widened and enhanced in the 1970-1980s, while its current state as a rapid transportation road was established in 2001. Homogeneous heterogeneous areas connected by a highway After analysing Enekuri’s plots, buildings, open space and street structures, several observations can be made. First, the building/plot resolution shows that there’s a huge variety of building/plot sizes and forms, and that only in one instance is there a sense of homogeneity: the case of the residential and the industrial plots in Enekuri (Fig. 3). These are rectangular lots including square or rectangular buildings within.

Figure 3. Detailed view of the newer homogeneous clusters Source: Eusko Jaurlaritza / Gobierno Vasco. geoEuskadi with changes made by the author

Second, the combinations among irregular plots and irregular shaped buildings represent the largest part of the case. Nevertheless, these form their own type of separate clusters of plots (Fig. 4) structured around roads that interconnect them in several ways. Third, at the same time, there are several building plots that remain isolated and do not participate in any cluster of any sort; such is the case of the water reservoir, the school, the gas station, and the junkyard. The latter three are only accessed from the highway, or by an additional road as in the case of the school grounds. Fourth, the irregularly shaped clusters, the isolated plots and the regularly clustered area are interconnected roughly only using the highway.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Finally, most of the land in the analysed area is of open character and unrelated to a building. This area represents almost 80% of the area of study (Table 2) and by relying on the current state aerial views and field work observation, not all green areas house neither an activity nor the same kind of land use and land cover.

Figure 4. Overview of the two older irregular clusters Source: Eusko Jaurlaritza / Gobierno Vasco. geoEuskadi with changes made by the author Table 2. The study area’s open space in figures

Total built (including, highway, open areas that are not considered productive or rural but are 462015,73 in an "urban" plot) approximate value Total Area (m2)

2.186.776,327

Total Open space approximately (m2)

1.724.760,597

Total Open approximate %

78,9

In addition, it must be said, that the time-based analysis of Enekuri-Bilbao has shown that some plots were developed and built at the same time and that this could indicate a kind of planned unity among them. For instance, some of the isolated plots and vehicle transportation related elements – gas stations, junkyard, and car seller – seem to have been developed at the same time the highway was first enhanced in the 1970s-1980s. Therefore, it is possible to discern four main elements that shape the urban fringe landscape of EnekuriBilbao’s mountain slope. While some can be considered homogeneous parts, other elements are homogeneous areas in themselves and their character can also be considered structural. 1.

Clusters: there are two types according to their form: regular and irregular and include residential or industrial plots structured by side roads or streets. Other kinds of plots and building types are also present in them: infrastructure related or industrial plots. The newest

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

2. 3. 4.

instances are the most homogeneous in terms of form, while the irregular ones are older developments. Isolated Plots: there are several examples in Enekuri and are surrounded by open green space and accessed through the highway: the school, gas station and junkyard and the water reservoir. Infrastructures: large scale (highway) or middle scale (roads) transportation infrastructure that serves as main and secondary structuring element among the different clusters and plots. Open Space rural land uses, green land cover and open space represents the largest amount of land in the analysed area and is usually unrelated to a specific building.

Conclusions It can be said that the morphological study can identify the unique form and character of the fringe and locate some specific morphological areas. An alternative method is not specifically necessary, but rather an adaptation that builds on the specific forms and elements found at the urban fringe. The overall character is that of morphological heterogeneity where it is possible to demarcate four types of elements: Clusters, Isolated Plots, Infrastructures, and Open Space. However, the four elements do not seem to fit together; Clusters, Open Space and Isolated Plots seem to be self-homogeneous and unrelated to each other were it not for the highway connectivity. As to the form of the Open Space plots, it is not possible to understand their logic of formation and relationship to the rest of elements by following the same urban morphology analysis steps. Neither can other infrastructural elements such as electrical lines be understood using the same rationale. However, these are necessary elements in order to apprehend the form of the urban fringe landscape due to their notable visual and spatial presence, but mostly due to their contiguity. By not including un-built open space and infrastructural elements within the analysis process, this method creates a differentiating tension among elements that typically are considered urban – some included in the analysis – or rural, and therefore neglects the rural-urban hybrid character of the fringe. Curiously, the scales of plot/street, city and region and elements that belong to each of these coexist sometimes in Open Space plots. It can be said that on the plot/street analysis level within the Clusters and Isolated Plots there’s usually a dominating building which embodies the land-use and completely takes over the whole space of the plot while also structuring the character and form of the connection to the transit element depending on its function – industrial, commercial-economic, service residential, educational. However, the un-built plots represent a different situation where several scales of analysis and elements coexist by combining for instance an electrical pylon and a pasture, or a pillar and viaduct of the highway that occupies seemingly rural open space. Also, the direct connection amongst the Isolated Plots to the large scale infrastructure that overcomes the city level indicates a discontinuous transition from one resolution scale to the other. In other words, whereas the Clusters follow the plot-building-street block-city scale continuum and then are linked to the wider urban form through various connecting elements such as infrastructure and open space, the Isolated Plots and Infrastructures, including the electrical lines perhaps, through their supporting constructions and elements directly jump from plot to region. Therefore, the main problems encountered in the morphological analysis based on morphological areas are the following: 



Many elements of the urban fringe, such as infrastructures, agricultural structures, free open space and other urban service related elements, cannot be analysed based on morphologically homogeneous areas structured by plots, streets and buildings. By leaving out said elements, the method accentuates a conceptual separation between so-called urban and rural elements, and therefore neglects the unique hybrid character of the urban fringe

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2



There is no sign of the slope acting as a formalising element, other than what can be guessed from observing the land form of several open space plots.

In the face of this separation and juxtaposition of two worlds, and also of the distinction of continuous and discontinuous sequence of analysis among the levels of resolution, it is possible to speculate that on the smaller scale of analysis by taking into account the presence of elements that belong to different analysis resolution levels there are signs of some levels of interaction. For instance, the coexistence between service structures or infrastructure constructions and open space is a significant combined form that would need further analysis since it seems that it is precisely on the plot level that both urban and rural worlds, along with multi-scalar and multi-resolution analysis levels collide and produce hybrid formalities. As for the lack of analysis addressing the possible correspondence among the form and functions of the area with the slope that might also explain the presence of certain types of open space, and in general add to the understanding of the specific character of the urban fringe, probably terrain sections should also be put in use to further explore the particularities of urban fringes on mountain slopes. In order to understand the form of the urban fringe, its special formal units and discontinuity of scales and contiguity of elements must be taken into account. The specific case of urban fringes on mountain slopes requires at least the use of sectioned views combined with aerial views.

References Brenner, N. & Schmid, C. (2011) ‘Planetary Urbanisation’, in Gandy, M. (ed.) Urban Constellations (Jovis Verlag, Berlin) 10-13. Gallent, N., Andersson, J. & Bianconi, M. (2006) Planning on the edge: the context for planning at the rural-urban fringe (Routledge, London). Garreau, J. (1992) Edge city: life on the new frontier (Anchor Books, New York). Gobierno Vasco (n.d.). GeoEuskadi Country Spatial Data Structure. http://www.geo.euskadi.eus/. Accessed November 2015. Indovina, F., (2005) ‘La ciudad difusa’, in Ramos, A. M. (ed.) Lo Urbano en 20 autores contemporáneos (Universidad Politécnica de Catalunya, Barcelona) 49–59. Instituto Geografico Nacional (n.d). Fototeca Digital. Gobierno de España. http://fototeca.cnig.es/. Accessed November 2015. Ipsen, D. & Holger, W. (2005) ‘Landscape Urbanism’ Monu-magazine on urbanism, 39-47. Meeus, S. J. & Gulinck, H. (2008) ‘Semi-Urban Areas in Landscape Research: A Review’ Living Reviews in Landscape Research 2. Moudon, A.V. (1997) ‘Urban morphology as an emerging interdisciplinary field’ Urban Morphology, 1, 3–10. Rowe, P. G. (1992) Making a Middle Landscape (MIT Press, Cambridge). Sieverts, T. (2003) Cities without cities: an interpretation of the Zwischenstadt (Spon Press, London). Soja, E. W. (2000) Postmetropolis: critical studies of cities and regions (Wiley-Blackwell, Malden, MA). Thomas, D. (1990) ‘The Edge of the City’ Transactions of the Institute of British Geographers, New Series 15, 131 -138.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Dwellings in the borderland: a study of traditional hill settlements in the Chittagong Hill Tracts of Bangladesh Monojit Chakma Primeasia University, Dhaka, Bangladesh Email: miki_buet@yahoo.com

Mirko Guaralda School of Design, Creative Industries Faculty, Queensland University of Technology, Brisbane, Australia Email: m.guaralda@qut.edu.au

Paul Sanders School of Design, Creative Industries Faculty, Queensland University of Technology, Brisbane, Australia Email: ps.sanders@qut.edu.au

Abstract. Bangladesh, like other developing Asian countries, is undergoing rapid urbanisation and deep socio-economic changes. The Chittagong Hill Tracts (CHT) region in the eastern part of the country has always been a unique area. It is characterised by the presence of diverse highland groups, each with its own language, culture and settlement patterns. The region, once isolated and not well researched, is now experiencing increasing transmigrations and urbanisation. The urban form of local traditional towns, affected by socio-economic changes, is at a risk of losing some of the spatial patterns that provide their sense of place. This paper discusses the background and complexities of CHT, at the edge of different cultural and political regions, and thus provides preliminary understanding of the traditional settlements in this region. A spatial analysis of selected urban areas has been based on qualitative data. This paper aims to discuss a methodological framework to inform a data collection process about morphological patterns in culturally diverse regions. It suggests that this methodological framework could have useful applications in investigating geographic areas with similar social and ethnic complexities. Keywords: highland groups, hill town, town-plan analysis, typological process, urban morphology, Zomia, Chittagong Hill Tracts

As Asia is experiencing rapid urbanisation, its cities are characterised by a complex mix of settlement patterns and built form types, mirroring the interplay between a great variety of societies and diverse influences (Dutt, 1994). Rapid development is challenging the traditional urban form of many settlements as well as threatening cultural and social continuity in these towns. Morphological analyses can be a strategic approach to gaining a better understanding of these cities and their complex cultural backgrounds. Additionally, in most parts of Asia, traditional sources of data for morphological analyses are not available or hardly accessible. In this region there is an endemic lack of historical town plans, maps, cadastral survey, and aerial survey photographs. Given such a situation, this study has developed a method to select case studies in a variegated socio-cultural context in order to inform subsequent field survey. The proposed approach is based on the use of different qualitative, not representational, data sources, such as archival records, census data, incidental ethnography, and colonial reports about a specific region in Bangladesh in South Asia. The aim of the paper is to guide the selection of specific cases of urban form that can then be investigated through field work to collect detailed physical data about these settlements.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

The Chittagong Hill Tracts (CHT) in Bangladesh is a particular case of this situation, with relatively limited cartographic data available, albeit being inhabited by diverse cultures â&#x20AC;&#x201C; highland people of various ethnicities as well as the majority Bengali population. These multiple cultural identities result in controls and adjustments among diverse culture groups, and shapes a mix of space types even within the same region. By examining CHT, in view of such cross-cultural setting and cartographic data constraints, this study developed a framework to identify the highlander neighbourhoods with diverse spatial layout. Mugavin (1999) outlined a general framework for understanding place in relation to its physical form. The framework suggested a three phase analysis in order to understand a city or town. The first phase involves and inductive description of an urban area acknowledging its socio-spatial practices. The second phase involves identification of patterns between physical fabric and its occupants along with sociospatial codes that integrate physical, mental and social concepts of a place. The third phase involves the discovery of place through historical review (Mugavin, 1999). Similarly, Kostof (1999) stressed the importance of conducting a deductive (historical/archival) review of documents and maps to develop an understanding of the cultural conditions of a settlement form, following inductive field observation methods. While this type of urban analysis is quite instructive, its application in Asian contexts would require some adjustment due to limited historical and cartographic data sources. Unlike the West, in most parts of Asia there was no consolidated tradition of developing documentation of settlement forms. With the absence of historical town plans showing plots and built forms in detail, morphological analysis at a micro level can be an alternative means of investigation (Chen, 2012). Consequently, typological understanding of neighbourhoods can be the primary step in undertaking morphological study in most parts of Asia. Aggregations of these micro scale investigations can contribute to the discussion about emergent patterns in settlement morphology of an entire city or town.

Typological process: Cultural model Morphological studies attempt to explain the physiognomy of a built landscape, its patterns and processes of change. The term, morphology, was originally conceived by philosopher Goethe as a method of investigation for studying the change and transformation process of all physical forms (Wilkinson and Willoughby, 1962). Later Sauer (1925) conceptualised it as morphology of landscape â&#x20AC;&#x201C; a synthesis of both social and natural features. Urban morphology is generally understood as the study of urban form (Whitehand, 2012) referring primarily to townscape (Smailes, 1955) or in more broadly to human settlement (Kropf, 2009). Among urban morphology research traditions, the theory of typological process had been developed based on Muratoriâ&#x20AC;&#x2122;s concept of operational history (Moudon, 1994, 1997). At the same time, Conzen theorised town plan analysis in his seminal work of Alnwick (1960) which involved a detailed morphological study of a market town in England. In discussing morphological research methods, Kropf (2009) contends that process typological (typological process) and historic-geographical process (town plan analysis) are the major theories. On transformation of types, drawing parallel to Darwinian view of evolution, Kropf (2001) also asserts that the typological process represents an example of phylogenetic change, featuring reproduction, variation and viability. In Europe, studies on vernacular settlements test the hypothesis of typological process (Cataldi, Maffei and Vaccaro, 2002). These studies were more in line with Caniggia and Maffei's (2001) earlier study on the base type of dwellings, and more elaborate studies on typological transformation of row and apartment houses in Rome. Continuing from these studies, other comparative studies were undertaken on small and medium sized towns in the context of cultural identity and the distinctive character of subregional areas (Cataldi et al., 2002). Similarly, there is a growing need for comparative and cross-cultural studies on the typologies of traditional neighbourhoods in Asian context. The CHT region offers the opportunity to study an area where vernacular built form types of various highland groups are undergoing a transformative process, demonstrating a constant dialectic between urbanisation forces and local built form types.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

In the highlands of Southern Asia, particularly in CHT region, the stilt dwelling is the base type built form, adopted by all the hill groups. Currently urbanisation is exerting rapid changes in spatial practices as well as built form types in the hills. With regard to external social factors that influence the built form transformation, Habraken (1998) pointed out that control causes transformation. In discussing the conceptualisation of type and typological process in built environment, Gauthier (2005) describes how the complex dialectic between type (cultural model) and external social factors requires recodifications of it. The following conceptual diagram (Fig. 1) based on Gauthier’s concept (2005), attempts to illustrate the relationship between built form types and the urbanisation process in present CHT. This diagram conceptualises that the traditional model of settlement forms and types have been built upon from a base type. The effects of urbanisation interact with the traditional model and produce recodified types in urban areas.

Urbanisation: external model

Recodified types Traditional dwelling type: local model

Base type

Figure 1: The relationship between the built form types and urbanisation process in CHT

Zomia: Highlands of Asia Recently a conceptual framework was developed in ethnological area studies introducing the concept of Zomia (Van Schendel, 2002), which is also known by other terms such as Southeast Asian Massif (Michaud, 2006). According to Van Schendel (2002), this highland region – Zomia (Fig. 2) – qualifies as a distinct study area, a cultural landscape characterised by diversity in ‘shared ideas, related lifeways, and long-standing cultural ties’ (Lewis and Wigen, cited in Van Schendel, 2002, p. 649). A transnational space, the Zomia geography sprawls across several nation states – ‘China, Vietnam, Laos, Cambodia, Thailand, Burma, India, Bangladesh, Bhutan, and Nepal’ (Van Schendel, 2002) and covers ‘an area roughly the size of Europe ... populations alone at around eighty million to one hundred million’ (Scott, 2009, p. 14). Once hilly shatter zones in pre-modern times (Scott, 2009), in this age of globalisation, the erstwhile peripheral regions are becoming linking spaces for the exchange of people, culture and trade among neighbouring states, with increasing access to transport networks and communication technologies (Brenner, Marcuse, and Mayer, 2012; Michaud, 2010). Against the backdrop of urbanisation, the highland settlements, at the edge of dominant nation state cultures, are developing particular settlement types.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 2. Map of Zomia Source: Creative Commons Zomia by United Kingdom Government, 2011 Image retrieved from https://commons.wikimedia.org/wiki/File:Zomia.jpg#

The borderland regions of various nation states are becoming increasingly relevant to the socio-political situation of South, Central and Southeast Asia. In addition to typical nation-state viewpoints, it is necessary to adopt theoretical frameworks from a transnational perspective to compare and investigate cities and towns in this complex culture area. Commenting on the necessity of comparative study of Asian traditional settlements, Knapp (2003, p. 10) argues for developing an understanding about ‘the nature of adaptation and acculturation of cultural material within and across contemporary borders’. The Zomia concept provides the spatial framework to examine this particular area, its people and settlements.

Chittagong Hill Tracts: past and present The CHT landscape, along with its people, fits the description of a Zomia culture area (Scott, 2009; Van Schendel, 2002). In Bangladesh, the urbanisation of CHT represents in a limited scale the cultural complexity and ambiguities of wider Zomia geography. To develop a pilot method, CHT is adopted as model area to identify specific cases for neighbourhood studies in a diverse socio-cultural context where different ethnicities coexist and there is a considerable lack of cartographic information. In a deltaic country with a seemingly homogenous population, the CHT represents physiographic variety as well as diverse highland groups. It comprises a surface area of 13,189km2 (5,089 square miles) with a population of 1.6 million (Fig. 3). Its three administrative districts cover seven towns and surrounding rural areas. In addition to central administration, a ministry specific to CHT, titled Ministry of Chittagong Hill Tracts Affairs (MoCHTA), has been set up to supervise its development goals. The traditional system through its hierarchy of community chiefs is nominally recognised to manage community affairs. The history of CHT can be divided into three broad phases: pre-colonial, colonial and post-colonial or nationstate (current Bangladesh) period. The pre-colonial historical references of CHT refer it as being part of Arakan kingdom (present Rakhain state of Burma) until late 1600. The earliest mention of one of the tribes – ‘Mranma’ – is found in a 1102 Mon inscription (Hall, 1981, p. 158). The region is mentioned in western narratives and maps as the Arakan and Burma employed Portuguese soldiers (Campos, 1919; Hall, 1981). In late 1600, the CHT became part of the Mughal Empire which captured a large portion of Arakan. However, imperial administration was minimal due to its remote location as well as gradual weakening of the empire which was drawn into chronic civil wars. Historical data from revenue records and hill chiefs’ chronicles supports the fact that the region remained in a tributary status (Bessaignet, 1958).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 3: Current Chittagong Hill Tracts Source: Creative Commons Chittagong Hill Tracts by Armanaziz adapted by User:AnonyLog â&#x20AC;&#x201C; Self-made based on File:BD Districts LOC.svg Retrieved from https://en.wikipedia.org/wiki/Mro_people#/media/File:Chittagong_Hill_Tracts.PNG

The colonial period started with the British Raj formally annexing CHT region in 1860 (Hutchinson, 1906; Lewin, 1869). It was a single administrative region, later divided into three administrative areas. Among colonial ethnographic and travel descriptions from this period, the most cited data source is the incidental ethnography of Lewin (1869). As a regional colonial chief, he lived in the hills for a long period, lived closely with hill groups and travelled far into remote parts of the region. His narratives about the CHT hill groups were the most detailed and provided vivid description of the geographic area and its diverse groups, their possible origins, languages and customs. Based on his descriptions, this research has drawn a route map (Fig. 4) indicating possible migration movements of the hill groups at the pre-colonial period.

Figure 4. Movement of CHT hill groups in pre-colonial period (pre 1860). Source: Author based on Lewin, 1869

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Another colonial data source, Hutchinson’s (1906) gazetteer report is also widely cited, as it provides useful data about the geographic area, location of the hill groups, their culture and the first colonial census. However, it did not provide much information about the minor groups and had number of debatable assumptions. While most data sources inform about the locale from a micro scale viewpoint, Francis Buchanan’s (1992) travel description of Southeast Asia offers a relatively macro scale comparison of the subject as he journeyed through many remote areas and communicated with local highlanders of both South and Southeast Asia (Van Schendel, 1992). The post-colonial period started with the end of British rule in South Asia in 1947 and CHT became part of the independent nation state Pakistan. In 1960, part of the region was inundated to make way for a hydroelectric project, resulting into mass displacement and deforestation and had a massive impact on the landscape and its people. Later in 1971, Bangladesh (former East Pakistan) gained independence from Pakistan and CHT came under new rule. However, under the new state, the local situation was marked with ethnic conflicts and insurgency that continued until 1997, when the signing of a peace accord formally ended the armed struggle. The post-independence period records several social science studies of notable western scholars on the study area. Geographer Sopher (1964) investigated the living style of the hill groups in relation to dwelling types and economy. His classification of hill groups based on their settlement topography is similar to Lewin’s observations (Fig. 5). Additionally, the study showed some hill groups have changed to valley settlement in a hundred year period. The kinship system of major highland groups had been investigated by Levi-Strauss (1952). Later, Bessaignet’s (1958) study involved a comparative review of the major hill groups including some settlement studies. The studies done by Loffler (1986) were focused on particular hill group as a whole – their living style, customs and settlements – while van Schendel (1992) completed a review of the socio-political situation of the region. Historical data sources indicate that diverse ethnicities settled and coexisted in this region prior to any colonial incursion. The hill groups, due to particular cultural or topographic contexts, settled at a range of altitudes in a variety of spatial arrangements. This apparently averted the possibility of any major conflict among them.

Figure 5. Topographic distribution of hill groups Source: Author based on Lewin, 1869 and Sopher, 1964

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Urbanisation in the hills Current urbanisation in CHT is now focused on its three major city areas. The major hub of the region is Rangamati, with an urbanisation rate of 67%, while the other two major towns, Bandarban (47%) and Khagrachari (62%), are not far behind (BBS, 2011). All three towns had low growth rate before 1990. Since mid 1990-2000, the growth rate accelerated, showing apparent correlation to the peace accord in 1997. Starting from about 1985, the Khagrachari municipality had the highest growth rate in the region until 2000, and the high growth rate continued (2000 to 2010) into the recent decade. After 2000, it has been eclipsed by Rangamati. While Rangamati was shadowing Khagrachari for 15 years (1980 to 1995), it has been experiencing a sharp rise in growth rate. There is apparently no primacy issue as all three are secondary cities and the difference in growth is minimal, after the peace accord, it can be argued that Rangamati is steadily shaping into a regional centre.

Methods With the backdrop of urbanisation, CHT characteristically shows, at a small scale, the diversity and complexity of the Zomia cultural landscape, its people and settlements. Similar to other peripheral regions of Asia, there is minimal cartographic data available. This study explores methods to identify case studies in a diversified socio-cultural context to provide information for further investigations. The proposed approach is based on the use of available archival data sources, such as census data, incidental ethnography, colonial records and social science studies about CHT. The region, characterised by its diverse cultural context and limitations in cartographic data, can be the study area for developing a pilot method to select and identify specific cases for further morphological investigations. For typological process, Caniggia and Maffei (2001) postulated a set of abstract hierarchical order of subdivisions (Fig. 6) which can be applied to individual buildings or to city or town. The selection framework (Fig. 7) â&#x20AC;&#x201C; beginning from regional scale (macro level) to the building scale (micro level) â&#x20AC;&#x201C; is based on the typological process framework (Caniggia and Maffei, 2001) and consists of spatial structure at different scales. In this framework, the study employs several methods such as ethnographic mapping, historical timeline, census data table comparisons and statistical modelling using relevant demographic and archival data in different scales of settlements to identify the most diverse ethnic neighbourhoods.

Figure 6. Caniggiaâ&#x20AC;&#x2122;s Typological process Source: Diagram redrawn by Author based on Caniggia and Maffei, 2001 and Kropf, 1993

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 7. Selection framework

Ethnographic mapping As a research tool, ethnographic mapping provides the starting point to determine spatial distribution of any particular cultural phenomena within a reasonable timeframe. The study uses ethnographic maps (Fig. 8) from the available data sources to gain an understanding about the geographic locations, distributions and numbers of the hill groups in different time periods. The following series of ethnographic maps are analysed in chronological order: 1900, 1950, 1960, 1965 and 2009.

Figure 8. Ethnographic map of CHT, 1901 Source: Author, based on Hutchinson, 1909

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Timeline table Based on the ethnographic maps, a timeline table (Table 1) is prepared in order to gain an historical overview about the locations, numbers of the hill groups at district level. Using the timeline table, the research selects 11 hill groups, who are identified as distinct culture group by available data sources and currently living in the region. In the table, the study uses an abbreviated letter-form prepared for each group; the first letter in upper case denotes the ethno-linguistic grouping, for instance, A for the Arakanese, M for the Mizo-Kuki-Chin and T for the Tripura group. The second letter in lower case denotes individual group name.

Current demography After selecting the hill groups, a map of their distributions at the sub-district level is prepared in table format (Table 2). In the table, data shows that the three sub-districts cover the most diversity given the number of hill groups. While two of them are urban sub-district, the third is mostly rural settlement. The table also shows that in another major urban sub-district, along with its surrounding sub-districts (both urban and rural), have very low diversity with just three hill groups. The previous timeline table (Table 1) supports this data; historically, the whole district (with all sub-districts) always featured less diversity. This can be due to the less hilly nature of the area or the minor groups not moving that far. The three sub-districts are spread in two districts. Among three sub-districts, Bandarban sub-district covers one town (Bandarban), nine wards and 69 neighbourhoods of urban area and a rural area with five unions, 16 mauza and 225 villages. Next, Rangamati sub-district covers one town (Rangamati), nine wards and 56 neighbourhoods of urban area and a rural area with six unions, 21 mauzas and 178 villages. Finally, Ruma sub-district covers just one urban neighbourhood and a rural area with four unions, 14 mauzas and 225 villages.

Data table comparison Selection of towns leads to the next step of identifying the neighbourhoods. The recent census (BBS, 2011) is used as a major data source for this selection process. From the census data, several data tables are used in combination to trace the neighbourhoods. The first community data table C12 provides the number of ethnic household, population and major communities at ward and union level. This data table does not provide data at neighbourhood and village level. To supplement this gap in data, the data table C13 and C15 are used. The community data table C15 provides the total household number and tenancy status of the household. This data table has information up to neighbourhood and village level. But, it does not have any category that provides specific data about the hill communities. Another data table C13 provides data on religious affiliations. This data table also has information up to neighbourhood and village level. This data table is useful for identifying ethnic neighbourhoods. This process is illustrated in Fig. 9. At the beginning, the first two factors, data table C12 and data table C15 are used to derive a percentage of ethnic household. This percentage is compared with the population percentage (Religious affiliation) derived from the third factor, data table C13. The corresponding percentages, both at ward or union level, validate the data from data table C13. Based on this comparison, potential wards and unions are identified. The identified wards and unions are further investigated at neighbourhood and village level using the data from Factor 1 (data table C15) and Factor 3 (data table C13). The same data are used to identify other neighbourhoods or villages. This produces three sets of data at neighbourhood/village level: potential wards and unions which have high percentage of ethnic household and community, other identified high number of ethnic communities in low diversity and ethnic community ward/union and high level of household ownership status. The comparison leads to the selection of neighbourhoods in urban and rural area.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

3000 25980 3000 1500 2000 ?

Khyang (My) Tripura (Tt) Murung (Tm)

? 15000

Reeang (Tr)

Marma (Am) Chakma (Ac) Tangchangya (At)

Identified as subgroup of Tripura Identified as subgroup of Tripura Not mentioned

Not mentioned

Khagrachari

Rangamati

Khagrachari

Rangamati

Banderban

District

977 3341 627 16121 1951 Not mentioned

416 23341

Not mentioned

Identified as sub-group of Tripura and location data Identified as sub-group of Tripura and location data

Identified as Mru

1300 38257

Not mentioned

Identified as Mru

Identified as a distinct group

Not mentioned

Identified as a distinct group but no location data Not mentioned

Identified as a distinct group

31906 44329

UNDP Survey (2010) District Banderban

Loffler (1965)

Khagrachari

Banderban

Rangamati

Sopher (1960) District

Khagrachari

Banderban 696 1615 144 10540 1469

Not mentioned ? 25000 2500

Rangamati

Bessaignet (1950) District

Khagrachari

Rangamati

Banderban

Khagrachari

Rangamati

Hutchinson (1900) District

Bawm (Mb) Lushai (Ml) Pangkhua (Mp) Mru (Mm) Khumi (Mu) Shendoo (Ms)

Chak (Ak) Arakanese group

Lewin (1860) District Banderban

Ethno-linguistic group Mizo/ Kuki/ Chin group Tripura group

Hill group

Identified as sub-group of Tripura

65889 124762 8313

Table 1. Timeline of ethnic distribution of hill groups

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Table 2. Distribution of hill groups at sub-district level

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Results The main purpose of this research is to explore methods to identify case studies in a diversified sociocultural context to provide information for further field research. The proposed approach is based on the use of available archival data sources, such as census data, incidental ethnography, colonial records and social science studies about CHT. The adopted neighbourhood selection process (Fig. 9) uses current census data of the following three sub-districts to identify the case study neighbourhoods. Factor 1: Neighbourhood/ village level data

Factor 2: Ward/ union level data

Factor 3: Neighbourhood/ village level data

Table C-15 General Household and ownership

Table C-12 Ethnic population and Household ↓ Ethnic Household number

Table C-13 Population by Religion (B/ C/ O)

↓ Total Household number

→

→ ↓ Identification of other neighbourhood/village

→

Percentage ↕ Percentage

Identification of ward/ union ↓ Comparison of neighbourhood/village ↓

←

↓ Total population ↓ Targeted Population

Ward/ union level

←

↓ Identification of other neighbourhood/village

Neighbourhood/ village level

Selection of neighbourhood/village

Figure 9. Neighbourhood selection process

Sub-district A: Bandarban The Bandarban sub-district covers and area of 501.99km2 with a population of 88,282, density of 176 per km2 and urbanisation rate of 46.93% (national: 23.30%). Its urban area includes one town having nine wards and 69 neighbourhoods, and the rural area has five unions with 16 mauzas and 225 villages. In Bandarban sub-district, the first two factors, data table C12 and data table C15 are used to derive a percentage table (Table 3) and chart (Fig. 10) illustrating the situation of ethnic household at ward level. The data on unions shows less diversity. This percentage is compared with the population percentage table (Table 4) and chart (Fig. 11) of religious affiliation derived from the third factor, data table C13. The corresponding percentage tables are compared at ward level. As both data are at the same level, the corresponding results also validate the data from data table C13. The data identifies some wards with higher percentage of hill groups thus having better possibility of diversity. The identified wards are further investigated at neighbourhood and village level using the data from Factor 1 (data table C15) and Factor 3 (data table C13). The same data are used to identify other neighbourhoods. Finally, five neighbourhoods are selected (Table 5) distributed across two wards.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Table 3. Hill group household percentage, Bandarban Percentage C-12 (Factor 2) C-15 (Factor1)

Ward number

Hill group Household number 177

Total Household number

Ward No-1

Hill group Household percentage 26%

Ward No-2

15%

130

850

Ward No-3

19%

132

699

Ward No-4

26%

182

688

Ward No-5

71%

812

1146

Ward No-6

12%

178

1479

Ward No-7

801

Ward No-8

766

Ward No-9

12%

151

1211

686

Source: Bangladesh Bureau of Statistics, 2011

Hill group Household percentage

71%

26% 15%

19%

26% 12% 1%

12%

Ward Ward Ward Ward Ward Ward Ward Ward Ward No-1 No-2 No-3 No-4 No-5 No-6 No-7 No-8 No-9 Figure 10. Hill group Household percentage, Bandarban Source: Author based on Bangladesh Bureau of Statistics, 2011 Table 4. Community percentage (by Religion, Buddhist/Christianity/Others), Bandarban Ward number

Percentage: C-13 (Factor 3) Community percentage Community population (by Religion, B/ C/ O) (by Religion, B/ C/ O)

Total Population

Ward No-1

31%

1243

3996

Ward No-2

14%

635

4511

Ward No-3

28%

982

3561

Ward No-4

33%

1145

3507

Ward No-5

77%

4150

5395

Ward No-6

15%

1112

7424

Ward No-7

3656

Ward No-8

155

3654

Ward No-9

12%

683 5730 Source: Bangladesh Bureau of Statistics, 2011

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Community percentage (by Religion, B/ C/ O) 77% 31%

14%

28% 33%

15%

12%

Ward Ward Ward Ward Ward Ward Ward Ward Ward No-1 No-2 No-3 No-4 No-5 No-6 No-7 No-8 No-9 Figure 11. Community percentage (by Religion, Buddhist/ Christianity/ Others), Bandarban Source: Author based on Bangladesh Bureau of Statistics, 2011

Ward number

Ward No-3 Ward No-5

Neighbourhood

Tripura para Kayang Area Madhyam para Rajbari area Ujani para

Table 5. Selected urban neighbourhoods, Bandarban Percentage: C-13 (Factor 3) Community Community Total percentage (by population (by Population Religion, B/ C/ O) Religion, B/ C/ O) 97% 169 174

C-15 (Factor1) Total Ownership Household number 36 92%

96%

120

125

63%

82%

1955

2376

551

56%

59%

159

268

55%

84%

1257

1490

303

53%

Source: Bangladesh Bureau of Statistics, 2011

Sub-district B: Ruma The Ruma sub-district, with a population of 29, 098, density of 59 per km2 and urbanisation rate of 1.25%, is a predominantly rural area. It covers an area of 492.09km2 and includes one urban neighbourhood, and four unions in the rural area with 14 mauzas and 225 villages. In this area, the data table C12 shows that there is no hill population in the only urban neighbourhood. It also shows except for one union there is very low diversity. Only the Ruma union has the most number of hill populations with considerable diversity. Therefore, concerning this sub-district the selection process can begin with identification of the union. The Ruma union has 89 villages. For these villages, the population percentages of religious affiliation are derived from the data table C13 (Factor 3). This percentage is compared with ownership level and infrastructure situation from data table C15 (Factor 1). The table (Table 6) shows five selected rural neighbourhoods having high level in all the factors.

Village

Betel para Asrom para Eden para Lon zhiri para Eden road para

Table 6. Selected rural neighbourhoods, Ruma Percentage: C-13 (Factor 3) C-15 (Factor1) Community Community Total Total Ownership percentage (by population (by Population Household Religion, B/ C/ O) Religion, B/ C/ O) number

Electricity

100% 100% 100% 100%

100% 72% 100% 100%

98% 97% 100% 92%

84%

97%

646 294 153 97

120 60 30 25

138 142 32 Source: Bangladesh Bureau of Statistics, 2011

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Sub-district C: Rangamati The Rangamati sub-district covers an area of 546.48km2, with a population of 124,728, density of 228 per km2 and an urbanisation rate of 67.35%. It includes one town with nine wards and 55 neighbourhoods of urban area, and six unions with 21 mauzas and 178 villages of rural area. In this sub-district, the data on the rural area from data table C12 shows only one union includes a considerably large hill population while having a reasonable number in diversity. As an urban area, the wards of Rangamati sub-district show considerable diversity. Therefore, considering this sub-district, both urban wards and the rural union are identified for further investigations. Beginning with the urban area, the first two factors, data table C12 and data table C15 are used to derive a percentage table (Table 7) and chart (Fig. 12) of ethnic household at ward level. Following that, the population percentage table (Table 8) and chart (Fig. 13) of religious affiliation derived from the third factor, data table C13 is compared with the previous percentage. Comparison of both percentage tables, as they are at the same level, and corresponding results validate the data from data table C13. The data shows number of wards having high percentage of hill communities indicating better possibility of diversity. Next, ten urban neighbourhoods are identified across five wards for further investigations. The identified wards are further investigated at neighbourhood and village level using the data from Factor 1 (data table C15) and Factor 3 (data table C13). The same data are used to identify other neighbourhoods. Finally, considering all available factors, five neighbourhoods (Table 9) distributed across two wards are selected for field study. In the rural area, the Balukhali union has 31 villages. On these villages, the population percentages of religious affiliation are derived from the data table C13 (Factor 3). This percentage is compared with ownership level and infrastructure situation from data table C15 (Factor 1). The table (Table 10) shows selected five rural neighbourhoods having high level in all the factors. Ward number

Table 7. Hill group household percentage, Rangamati Percentage C-12 (Factor 2) C-15 (Factor1) Hill group Household percentage

Hill group Household number

Total Household number

Ward No-1

1561

Ward No-2

1881

Ward No-3

22%

312

1402

Ward No-4

108

1782

Ward No-5

38%

503

1338

Ward No-6

51%

975

1905

Ward No-7

19%

438

2317

Ward No-8

58%

1918

3297

Ward No-9

49%

1103 2264 Source: Bangladesh Bureau of Statistics, 2011

Hill group Household percentage

22%

38% 51%

19%

58% 49%

Ward Ward Ward Ward Ward Ward Ward Ward Ward No-1 No-2 No-3 No-4 No-5 No-6 No-7 No-8 No-9 Figure 12. Hill group household percentage, Rangamati Source: Author based on Bangladesh Bureau of Statistics, 2011

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Table 8. Community percentage (by Religion, Buddhist/Christianity/Others), Rangamati Ward number

Percentage: C-13 (Factor 3) Community percentage Community population (by Religion, B/ C/ O) (by Religion, B/ C/ O)

Total Population

Ward No-1

369

8222

Ward No-2

574

9258

Ward No-3

27%

1747

6357

Ward No-4

541

8420

Ward No-5

38%

2281

6036

Ward No-6

55%

5062

9186

Ward No-7

21%

2198

10675

Ward No-8

61%

9158

14988

Ward No-9

46%

4947 10858 Source: Bangladesh Bureau of Statistics, 2011

Community percentage (by Religion, B/ C/ O) 61%

55%

46%

38% 27% 4%

21% 6%

Ward Ward Ward Ward Ward Ward Ward Ward Ward No-1 No-2 No-3 No-4 No-5 No-6 No-7 No-8 No-9 Figure 13. Community percentage (by Religion, Buddhist/Christianity/Others), Rangamati Source: Author based on Bangladesh Bureau of Statistics, 2011 Table 9. Selected urban neighbourhoods, Rangamati Percentage: C-13 (Factor 3) C-15 (Factor1) Ward number

Ward No-5

Ward No-8

Neighbourhood Community percentage (by Religion, B/ C/ O)

Total Household number

Ownership

Tangchanga para

85%

215

84%

Belai chari

99%

Tribal Adam Purba Tribal Adam

97%

185

65%

79%

161

76%

Rajbari

89% 546 Source: Bangladesh Bureau of Statistics, 2011

72%

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Table 10. Selected rural neighbourhoods, Rangamati Percentage: C-13 (Factor 3) Village

Total Popul ation

C-15 (Factor1) Owner Total ship Household number

Electri city

Community percentage (by Religion, B/ C/ O)

Community population (by Religion, B/ C/ O)

99%

431

433

100%

Mog Para Lokkana Karbari para

99%

94%

100%

122

100%

20%

Kuki para Jarul Chari para

100%

184

100%

50%

100%

133

56%

Balukhali Chakma Para

Source: Bangladesh Bureau of Statistics, 2011

From three sub-districts, a total of 10 urban neighbourhoods and 10 rural neighbourhoods are identified as potential high diversity ethnic neighbourhoods. Among these neighbourhoods, Bandarban sub-district contains five neighbourhoods, Ruma sub-district covers five rural neighbourhoods and Rangamati subdistrict includes ten neighbourhoods in urban and rural area. Following that, in Bandarban town the urban wards are identified (Fig. 14), showing that the targeted neighbourhoods are located in a cluster.

Figure 14. Bandarban municipality map: identified neighbourhood Source: Google Maps

Conclusion This study proposes a framework of methods that provides an alternative approach to the neighbourhood selection process of settlements in cross-cultural settings, albeit with limited cartographic and survey data. The framework uses minimal forms of data available for the area, such as, ethnographic maps, gazetteer records, census data and social studies. It identifies potential case study neighbourhoods facilitating further field investigations. For any morphological study of a less researched area such as CHT, a micro scale analysis of neighbourhoods can be the primary step in initiating a typological discussion. In the absence of historical plans showing plots or building plans, field investigations can provide the necessary primary data. Prior

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

to embarking on fieldwork, identifying the neighbourhoods with the most ethnic diversity, can be the starting point for any micro level settlement study. This can effectively lead to findings of diverse built form types in a time efficient way. Typological discussion of a particular landscape can begin with simply identifying the neighbourhoods which contain the targeted cultural groups and their dwellings. Then field data can validate the types and patterns of the built forms. In the case of diversity (or lack thereof) in built form types, the findings from field data might explain whether the arrangement of multiple ethnicities has resulted into diverse types of spatiality or the effects of urbanisation producing a recodified types in the urban area. The lack of comprehensive cartographic data or survey (cadastral) records is endemic to many developing countries in Asia. In some cases, even the census data is not available in proper detail. Therefore, in a possible scenario of minimal data availability, the proposed framework might be useful to identify targeted neighbourhoods or settlements to initiate any micro level morphological study. The process can be an iterative and complementary one and its methods and steps can be adjusted to any particular situation. Therefore, the framework developed in this study provides an alternative or corresponding means to explore an uncharted landscape that awaits scholarly attention.

References Adnan, S., & Dastidar, R. (2011) Alienation of the Lands of Indigenous Peoples of the Chittagong Hill Tracts of Bangladesh. Retrieved from http://www.iwgia.org/iwgia_files_publications_files/0507_Alienation_of_the_lands_of_IPs_in_the_CHT_of_Banglad esh.pdf. Bangladesh Bureau of Statistics (2011) Bangladesh Population and Housing Census 2011: Community report. Retrieved from http://www.bbs.gov.bd/home.aspx#. Brenner, N., Marcuse, P., & Mayer, M. (2012) Cities for people, not for profit: critical urban theory and the right to the city (Routledge, Abingdon, Oxon). Caniggia, G., & Maffei, G. L. (2001) Architectural composition and building typology: interpreting basic building. (Alinea, Firenze). Cataldi, G., Maffei, G. L., & Vaccaro, P. (2002) Saverio Muratori and the Italian school of planning typology. Urban Morphology, 6(1), 3-14. Conzen, M. R. G. (1960) Alnwick, Northumberland: A Study in Town-Plan Analysis. Transactions and Papers (Institute of British Geographers)(27), iii-122. Retrieved from http://www.jstor.org/stable/621094. Dutt, A. K. (1994) The Asian city: processes of development, characteristics, and planning (Vol. 30.) (Kluwer Academic Publishers, Dordrecht). Gauthier, P. (2005) Conceptualizing the social construction of urban and architectural forms through the typological process. Urban Morphology, 9(2), 83-93. Retrieved from http://qut.summon.serialssolutions.com/link/0/eLvHCXMwY2BQSDFKTklKSktMSUoEln0GyZYmZqmJaWZGacag w-_A1yYgBtuQSnM3UQZtN9cQZw9d5CogvgBy7EI8-GwWUIMfPCRtZCjGwJsIWgWeVwLeLZbCp7N995Haq-4ukLCRRU9m7IBsJ0nNA Hutchinson, R. (1909) Chittagong hill tracts. Knapp, R. G. (2003) Asia's old dwellings : tradition, resilience, and change (Oxford University Press, Oxford) Kropf, K. (1993) The definition of built form in urban morphology (University of Birmingham). Kropf, K. (2001) Conceptions of change in the built environment. Urban Morphology, 5(1), 29-46. Kropf, K. (2009) Aspects of urban form. Urban Morphology, 13(2), 105-120. Retrieved from http://qut.summon.serialssolutions.com/link/0/eLvHCXMwY2BQSDFKTklKSktMSUoEln0GyZYmZqmJaWZGacag w-_A1yYgBtuQSnM3UQZtN9cQZw9d5CogvgBy7EI8eJYMVOWYgxu2hmIMvImgVeB5JeDdYil82xadDLz3w6Nq7HJcVyMe4IBwUco2Q. Lewin, T. H. (1869) The Hill Tracts of Chittagong and the dwellers therein: Bengal printing company. Michaud, J. (2006) Historical dictionary of the peoples of the Southeast Asian massif (Scarecrow Press, Lanham, Md). Michaud, J. (2010) Editorial â&#x20AC;&#x201C; Zomia and beyond. Journal of Global History, 5(2), 187-214 doi:http://dx.doi.org/10.1017/S1740022810000057. Mugavin, D. (1999) A philosophical base for urban morphology. Urban Morphology, 3, 95-99. Sauer, C. O. (1925) The morphology of landscape (Vol. 2).(University of California Press, Berkeley, Los Angeles). Scott, J. C. (2009) The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia (Yale University Press). Smailes, A. E. (1955) Some Reflections on the Geographical Description and Analysis of Townscapes. Transactions and

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Papers (Institute of British Geographers)(21), 99-115. doi:10.2307/621275. Sopher, D. E. (1964) The Swidden/Wet-rice transition zone in the Chittagong Hills. Annals of the Association of American Geographers, 54(1), 107-126. doi:10.1111/j.1467-8306.1964.tb00477.x. Van Schendel, W. (2002) Geographies of knowing, geographies of ignorance: jumping scale in Southeast Asia. ENVIRONMENT AND PLANNING D-SOCIETY & SPACE, 20(6), 647-668. doi:10.1068/d16s Whitehand, J. W. R. (2012) Issues in urban morphology. Urban Morphology, 16(1), 55-65. Retrieved from http://qut.summon.serialssolutions.com/link/0/eLvHCXMwY2BQSDFKTklKSktMSUoEln0GyZYmZqmJaWZGacag w-_A1yYgBtuQSnM3UQZtN9cQZw9d5CogvgBy7EI8pPkOSkMG4JaHGANvImgVeF4JeLdYCpMiuzS9v3e8qsDWcWXmEy_CACaWiZC Wilkinson, E. M., & Willoughby, L. A. (1962) Goethe. Poet and Thinker: Essays.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Abstract. To defend the sovereignty of the ‘Conquered Territory’ along the eastern frontier of the Boer Republic of the Orange Free State (now Free State Province of South Africa) and Basutoland (now Lesotho), the former government passed Occupation Act, 1866, and established three rows of farms and as many towns. As these towns were government founded and of strategic importance, unlike the Boer tradition of church-founded towns with parishioners settling around the place of worship and sustaining themselves from urban agriculture, residents were offered an inducement to settle conditional upon the implementation of prescribed security measures and house building, but the town plans made no spatial provision for religious or market practices. This article aims to show that even under these circumstances the towns came to feature the usual diagnostic characteristics of Boer-founded towns and concomitant repertoire of inherited townscape traditions. To bed the argument, the morphology of Boer-founded towns as developed in history is briefly investigated with the implication that the adjustments made to merge the planned towns with the family of Boer towns were culturally driven. While a number of authors have studied various aspects of South African town planning history, these planned towns and urban entities have received scant attention within the family of Boer-founded towns. Keywords: Boer town planning, planned urban form, quartered blocks, urban agriculture

The Conquered Territory and the Occupation Act (Occupatiewet) of 1866 To bring the Boers, who had left the Cape and trekked northward from 1835 onward, under the authority of Queen Victoria, British commander Sir Harry Smith proclaimed the Orange River Sovereignty (ORS) on 3 February 1848. This meant that the eastern, central region of South Africa between the Orange River and its largest tributary, Vaal, would henceforth be British (Eloff, 1979, p. 8). However, the Basuto nation, indigenous to the mountainous region east of the ORS did not accept the Caledon River, the major eastern tributary of the Orange as the frontier. In consequence, Smith instructed the appointed British Resident of the ORS, Major Warden, to seek a redefinition in consultation with Moshesh, king of Basutoland (Lesotho since 1966). This lead to the proclamation of the Warden Line of 1849, but did not bring peace (Eloff, 1979, p. 9). After suffering humiliating defeats to Moshesh in 1851, 1852 and 1853, Britain abandoned the ORS, which, in terms of the Bloemfontein Convention of February 1854, then became known as the Orange Free State (OFS). In consequence the fledgling independent Boer republic inherited the border problem.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

A protracted dispute resulted in three further wars (1858, 1865-66 and 1867-68) with the resignation of as many OFS president. Inaugural president of the new Republic, Hoffman, resigned 1855 (Eloff, 1979, p. 10), as too did successor Boshof in 1859, and third, Pretorius, in 1863. Under the fourth, President JH Brand, (appointed February 1864) the delimitation was settled. However, the foundations had been laid during the 2nd Basuto War (May 1865 - April 1866) when the northern commando under Boer Commandant-General JIJ Fick and the southern under Commandant Louw Wepener concluded successful military operations, and in the process acquired considerable territory. This shifted the boundary eastward of the Warden Line to the Caledon River with a tangential extension south-eastward to the Orange, the bend beginning where the town of Wepener was later established. This arable stretch of land, referred to as the Conquered Territory, was ceded by Moshesh in terms of the treaty of Thaba Bosiu on 3 April 1866 and annexed by the OFS (Eloff, 1979, pp. 17-19) (Fig. 1). The 3rd Basuto War resulted from the refusal of the Basuto to vacate the area.

Figure 1. The three planned towns of Wepener, Ladybrand and Ficksburg within the Conquered Territory (shaded), which eastern boundary is marked for the most part by the Caledon River, and the Orange River in the south (Kobus du Preez).

The “wall of flesh” A prerequisite for peace and order was the colonisation of the Conquered Territory with the establishment of a large number of settlers to serve as a buffer against renewed border violations. The proposals of a commission of the OFS Volksraad (legislative assembly) were adopted and the Occupatiewet (Occupation Act) was written into law in 1866. In terms of this ordinance three rows of farms, each of 1500 morgen (approx. 1300ha), stretching the entire border were to be surveyed. The first farms would be granted to citizens who had seen active commando service during the 2nd Basuto War (1865-66) and the remainder would be sold by public auction. The object of the occupation scheme was to form a bulwark of farmers to safeguard the OFS against Basuto raids, “the wall of flesh”, as founder OFS President JP Hoffman termed the strategy (Eloff, 1979, p. 21). Each owner would be required to erect a house of at least 20x10ft (6.1x3.05m) within six months of occupation; live on it; at all times be in possession of “one horse, saddle, bridle, rifle, 200 bullets, 5lbs (2.27kg) of gunpowder and 500 percussion caps or flints” and be prepared to perform whatever civil or commando duty was considered necessary by a field cornet (local government official or military officer) (Eloff, 1979, p. 20).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

The OFS government delayed implementation and the Basutos continued where they had left off. Surveying was hazardous and no troops were assigned to enable progress, under which circumstances farmers were unwilling to settle. Nevertheless, surveying of 605 farms was complete by 1869 (Eloff, 1979, p. 35) and house-building could begin, all employing sandstone, the material particular to the area (see Du Preez, 2012).

Town establishment The government realised that occupation would not, on its own, ensure an effective buffer against Basuto incursions and a decision was therefore taken to lay out towns in order to achieve a greater concentration of settlers. In terms of section 9 of the Occupation Act, the Volksraad session of 1867 approved the establishment of three ‘military’ towns (Eloff, 1979, p. 26) in the conquered territory: Wepener in the south, some 6km east of the Caledon River; Ladybrand in the centre of the Conquered Territory, some 16km west of the Caledon; and Ficksburg in the north on the western bank of the Caledon. Sites for these towns were identified at the feet of mountains, naturally defensive and well watered. Due to the haste, land surveyors were invited to tender for the survey and layout of the towns by way of an advertisement placed in the Gouvernements Courant (government gazette) of 27 June 1867, and the same issue also contained an advertisement for applicants to occupy the towns. Concessionaires would be required to keep a rifle and adequate ammunition at all times, were obliged to enclose the assigned property with sods, stones or other appropriate fencing material within three months, and build a house thereon within six. The conditions prevailing at mid-1867 were not conducive to the founding and development of towns (Eloff, 1979, p. 27); their strategic positions scared off prospective residents, and the outbreak of the 3rd Basuto war in July 1867 postponed both survey and lot allocation. However, this war ended when the tide turned against the Basuto, and Moshesh reiterated his earlier requests to have Basutoland declared a British protectorate, which was so proclaimed in March 1968, and this finally established the present boundaries. The delimitation arrangements of February 1869 between the British and the Boers entailed that the OFS was to cede part of the Conquered Territory east of the Caledon to the Basuto, which resulted in the Caledon marking the boundary southward to Wepener with a south-eastern field boundary to the Orange. The survey of stands in the three towns could now begin with land surveyors G van der Bijl Aling, GAGP van den Bosch and KJ de Kok assigned to Wepener Ladybrand and Ficksburg respectively (Eloff, 1979, p. 26). A common brief was given the surveyors and published in the Notulen (resolutions) of the Volksraad, namely lots of 300x150 ft (91.44m x 45.72m) in blocks of four with a total of 200. Ladybrand was completed in March 1868, although the first residents arrived much later; Wepener by mid-1868, but only 35 of the 400 stands were occupied at the beginning of 1869 when with the relative peace colonisation of the Conquered Territory could begin. During the Volksraad session of May 1870, President Brand announced that 60 houses had been erected in Wepener, about 15 in Ladybrand, and in Ficksburg ten stands had been fenced off, six houses complete with four under construction (Eloff, 1979, p. 35).

The character of Boer-founded towns According to Haswell (1979, p. 687), many of the Boer-founded towns grew around the places of worship established by the Nederduitse Gereformeerde Kerk (Dutch Reformed Church, NGK) because members widely dispersed on farms in the interior of South Africa, needed to converge quarterly to attend the nagmaal (communion/eucharist) service over several days. The church building was often located on a bult (knoll) and placed on a large square for farmers to encamp and trade during the nagmaal weekend. In due course, those who could, would acquire properties in close proximity to the church and build tuishuisies (‘homes in town’, cottages actually) for accommodation over the nagmaal weekend. The embryonic church towns would consist of a single street, lined on two sides by cottages, built like their Dutch precedents on the street boundaries in one wall of development, leaving maximum space for a back

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

garden. When the grid extensions were added to the plan, church and square retained their focal positions within the street system while cemeteries were placed on the edges of towns with graves aligned east-west. Typically, Boer founded towns were gridded and laid out on a spur site that the long streets could take best advantage of the slope and facilitate the leading of water for irrigating the elongated lots, which usually stretched from street to street. Water was obtained from a river or spring, and was lead by means of leivore (water leads) through the settlement in such a manner as to provide for the irrigation of the erven (lots), which were identified as ‘water’ or ‘wet’, as opposed to ‘dry’, unconnected to the system. Dry erven could be suitable for tuishuise or commercial use, but as pointed out by Floyd (1960, p. 11), shops and businesses did not play an important part as the distant farmers were served by travelling pedlars. Despite the aforegoing, Haswell concludes that the raison d’être of Boer-founded towns was neither ecclesiastical nor commercial, but water for irrigation to accommodate agricultural communities, for which reason such settlements should more correctly be termed neither village nor town but dorp. Consistently, Boer-founded towns were devoid of social appurtenances such as recreation facilities or sports fields. Pietermaritzburg, the capital of the Boer republic of Natalia (now KwaZulu-Natal), was founded by Voortrekkers in 1838. It too had a furrow system which led water from a tributary to the cultivated erven, each sized 50x150 paces (150’x450’ or 45.7m x137.2m). Development was controlled by six regulations passed by the Volksraad of Natalia, of which articles 4 and 5 are the most important from the dorpsgesig (townscape) point of view: that erven be planted within two months of purchase, and enclosed with a side wall or wooden palisade; and that houses be built on the street boundary and in one continuous line. As a result, the fledgling dorp represented the picture of regularity and orderliness, a prime characteristic of Boer towns (Haswell, 1984, p. 19).

The establishment of towns in the Orange Free State Almost all settlements of the first decade of the OFS republic were established by either the church or mission, without the approval of the Volksraad, and located demographically in the centre of larger farming communities, with water availability a prime consideration (Moll, 1977, pp. 23-28). This resulted in a satisfactory distribution to suit the needs of the time, and varied by the carrying capacity of the land. Freehold ownership was introduced from inception (Floyd, 1960, p. 37). The town plan was in most cases a grid-iron pattern with the blocks sub-divided into lots backing onto each other, not as long narrow lots stretching from street to street. Common features were the large church and market squares. Whenever possible, town plans were designed that the church became the dominant building and feature in the town. In fact, everything deferred to the towering NGK church buildings, often of a later generation, that the townscape of a dorp can be termed medieval. In keeping with the small population of English settlers, the Anglican church, though typically Gothic and often of stone, is diminutive in size. A school building would follow and state funds would cover teachers, the administration of the magistrate’s court, justices of the peace, public works etc, the commitment of which brought with it the residences of the officials.

Planned towns in the Conquered Territory As Wepener, Ladybrand and Ficksburg were founded by the government i.e. a secular organisation, and not by the church, the obvious distinction on their plans is the omission of any dedicated positioning for a church building let alone a church square. The plans of Wepener and Ladybrand are uniform without any variation in block size or lot designation; only Ficksburg made provision for a square each for the church and market. However, like many Boer-founded towns, these sites were irrigated by streams crossing the sloping or spur sites as well as by a system of water leads. The layouts are gridded, but unusually of blocks neither

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

sub-divided in rectilinear lots in series stretching between streets nor by backing lots, but by blocks each of four lots quartered as corner lots, which according to the town plans, were sized not in English feet but in 300x150 Cape Feet (94.46m x 47.23m). Quartering could only be found in one other town in the OFS, Reddersburg, established in 1859 by the Gereformeerde church, the smaller reformed faith community, as a “Kerkplaats...ten behoeve van de gemeente” (as a church town for the settlement of the congregation) (Moll, 1977:25). The lots in Reddersburg are square, which may have been a more efficient layout for the purposes of urban agriculture, a priority shared by a contemporary newspaper, The Friend, which, when writing about the lots in Ladybrand concluded that “... this will furnish a good opportunity to poor people for getting a permanent residence, with the prospect of their being able to cultivate the greater part of their provender for themselves” (13 March 1868, cited in Bosch, 1967, p. 45). While every stand-owner was obliged to enclose his property with sods, stones or other appropriate fencing material within three months of allocation, and to build a house on it within six (Gouvernements Courant, 27 June 1867), unlike Pietermaritzburg, there was no prescription of the positioning of the house. In fact, houses would probably be at their securest placed away from the street boundaries, in the centre of the lot. While this freedom could change the townscape, interestingly, historical photographs show many buildings on the street boundaries. Wepener This town was named after Commandant Louw Wepener (1802-1965), killed while storming Thabo Bosiu. Due to a reconsideration of location, Wepener in the south, was the last to be surveyed and its establishment is dated 1869, two years after the others. The rectangular plan was laid out south of the Jammerberg mountain, on a sloping site dissected by the Sandspruit (stream), a tributary of the Caledon (Fig 2).

Figure 2. [LARGE] Wepener; original town plan of 1867. Church Street is the fourth long street from top (Free State Archives).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

As could be expected, Aling imposed a gridded plan with long streets running roughly north-south, streets 60 Cape feet wide (18.89m), to which names were assigned much later, as Oberholster concluded (1969, p. 40). Besides Church and Brug (bridge) streets, which reflect their designations, all others carry surnames of pioneering local residents. Due to the marshy ground, the first houses were built on the higher-lying portion of town, on the south-east. The NGK congregation was established in 1870 and consecrated its first church building a year later on a square outside of the town plan, in such a way that the church terminates the vista of the street which came to be named Church Street. To this were added another 88 dry lots tangential to the general alignment, by government surveyor C. Vos in March 1875 (Fig. 3). From their position relative to the church, the small sizes, and the fact that these were dry lots, one can only conclude that this addition was designated for tuishuisies. This adds a familiarity with the way Boers built towns replete with the towering church erected in 1884 (Fig. 4).

Figure. 3. Wepener; extension of 1875, which includes â&#x20AC;&#x2DC;Church Squareâ&#x20AC;&#x2122; and 88 smaller dry lots, most likely for tuishuisies, which came to be fitted on the eastern end of the original gridded plan (Free State Archives).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 4. Wepener today; photograph showing the towering church of 1884 terminating church Street (towards left) and the deferring tuishuisies. Photo from the higher-lying south-east, by author.

Though no square of any kind was designated on the map, much like a colonial Roman town a block was given over in the principal east-west street for purposes of a market, at right angles to the main northsouth (Church) street. This probably sprouted a market hall like neighbouring towns had, before being subsumed for the town hall built in 1928. The diminutive Anglican church occupies no landmark site but is situated along Church Street. While cemeteries were generally positioned on the edges of towns, and usually on lower lying land, in Wepener the cemetery is on the northern end, above town, probably to protect the site from flooding. Ladybrand Established in 1867, Ladybrand is located at the foot of Platberg and named after Catharina Brand, mother of the 4th OFS-President who resolved the border issue. The tender submitted by land surveyor KJ de Kok was accepted, and he carried out the task of laying out the initial 200 blocks each of four quartered erven during February and March of 1868, under the protection of 150 men. An apt description of the development was carried in The Friend: “The town of Lady-Brand will be situated on a little rise running from S.S.W. to N.N.E. The main street will go along the top of the rise, and on both sides of it the erven will be laid out. Two strong fountains, coming out of the Platberg … will supply the town with water. These fountains issue from both sides of the rise, and will supply the erven on either side … The town and its commonage are sheltered by the Platberg on the S.E. through W. to N. and the country is open from N. through to E. and S.W.” (13 March 1868; cited in Bosch, 1967, p. 44) (Fig. 5). This is the description of an almost ideal spur site, perfectly orientated, protected from the elements and generously watered. The nine inclined long streets bear the names of pioneering families except for Church Street which marks the rise (ridge/watershed), while the nine cross-streets are simply numbered 1-9. Unlike Wepener where the stream subsequently ravaged and divided the town, the flow of the streams of Ladybrand could easily be contained in furrows as can clearly be seen on an historical photograph of c. 1870. This photograph also shows the enclosures to the properties and the positioning of many cottages on or near the street boundaries, as is the legacy of the Dutch town-making tradition.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 5. Ladybrand, plan of 1868. A gridded plan of quartered blocks along long streets running in a north-easterly direction down the gentle slope at the foot of the Platberg mountain from whence the two fountains originate. The central long street runs along the watershed, and two crossed blocks, right of centre, indicate those taken by the NGK church (top) and market (centre) respectively (Free State Archives).

While the original church building was a cottage, probably occupying one lot of a block, in 1890 the permanent sandstone church was consecrated in the position it still occupies today, on the rise and in the centre of a block, effectively commanding what is referred to colloquially as Church Square, the characteristic feature of Boer town planning (Fig. 6). Lower down along Church Street a block became assigned to the market, and like many other towns this public space became marked with the town hall in 1931. The police station, a fine work of OFS Republican sandstone architecture, is located on the block south-west of Church Square where it adds architectural definition to the Square.

Fig. 6. Ladybrand today. The church of 1890 on Church Square. Source: Photograph by author

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Ficksburg The Friend labeled Ficksburg the “best situated in Conquered Territory” (10 March 1870, cited in Bosch, 1967, p. 30). Named after Comdt-General JIJ Fick, the roughly east-west aligned grid-iron plan is sited at the foot of the majestic Imperani mountain and on the western banks of the Caledon, ie on the frontier. Although surveyor Bosch commenced in December 1867 progress was slow and he had to be protected by a large commando as “pegs were pulled out by the Basuto and used as firewood” (Bosch, 1967, pp. 29-30), but the laying out of the 260 lots of both dry and wet erven along eleven long and nine cross-streets was completed in March 1868 (Fig. 7). Consistent with Wepener and Ladybrand, the object was to settle poor people who would sustain themselves by their garden lots and were required to surround these with walls of sod or stone to serve as a barricade against attack. Remains of such walls were reported still on existence in 1970 (SESA, Vol 4, p. 485).

Figure 7. Ficksburg, plan of 1867. The gridded plan runs downhill from the foot of the Imperani mountain (at left) to the Caledon River (at right). All blocks are quartered but for those truncated to create squares for the market (top left) and NGK church (top left of centre) respectively, aligned with the watershed. Three fountains at the foot of the Imperani Mountain provided for the irrigation of the allotments (Free State Archives).

As already stated, Bosch made an exception by including in the gridded plan dedicated squares each for the church and market, and most unusual, Market Square occupies the elevated position with Church Square lower-lying. These squares lie in tandem on the watershed and, interestingly, this street is not named Church but Hoog (High) Street, which is the name given the main street of English-founded towns in South Africa e.g. Grahamstown. Nevertheless, both squares on the watershed were shaped by truncating one quarter from each of the four surrounding blocks, a precedent set at Graaff-Reinet in the Eastern Cape, founded 1830 already. As a result, the market and church buildings interrupt and terminate both the main long street as well as two cross-streets named Hout and Church St respectively. But, interestingly, Market Square accommodates the town hall (1897) and magistrate’s court (1892) and, as these two public buildings are in alignment with the church (1905), the most important buildings of Ficksburg are experienced in series along the watershed, the sole piece of urban design in the three towns (Fig. 8).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 8. Ficksburg today. The main public buildings on the axis: town hall in the foreground and the magistrateâ&#x20AC;&#x2122;s court, of which the turret is just visible (towards left) on the former Market Square, and the spire of the church (left) on Church Square. Source: Photograph by author.

Conclusions Because a common brief was issued each of the land surveyors, the layouts of the towns Wepener, Ladybrand and Ficksburg bear a distinct familial similarity but for the two squares of the last. With the gridiron layout as the determining morphological rule of towns primarily for agricultural communities, they also bear an uncanny resemblance to the bastide towns of 13c south-west France, implemented as a key factor in establishing English authority, compare Monpazier, founded 1285. No OFS town had a central arcaded market square and all streets are equally wide, but the rigidity of the grid and the church and market each eventually occupying a square, follow the prototype of a bastide, which agricultural community is an appropriate precedent as it was designed as security from small-scale localised attack. While not church-founded, the NGK church buildings got their usual position of visual dominance and concomitant squares. Wepener has the church on a knoll-like position; in Ladybrand the church occupies a block on the watershed in the centre of town, and while lower down in Ficksburg, the verticality of the twin spires accords it the traditional dominance. Like all Boer dorpe, these â&#x20AC;&#x2DC;militaryâ&#x20AC;&#x2122; towns too are agricultural by nature. The quartered blocks are a different configuration from most but were designed for the settlement and sustainment of people who would use the lots for agricultural purposes. While this layout and the free positioning of houses, brought about a break with the inherited paradigm of recreating Dutch townscapes, the regulation of enclosing lots and the distinctiveness of ecclesiastical architecture marking or terminating the axis of the main street did happen, as did a square each for the church and the market. These planned towns, though designed to accommodate issues of defence and urban agriculture, came to acknowledge inherited cultural and religious traditions and townscapes, which in their regularity and morphological features, fit in well with the family of Boer-founded dorpe.

Acknowledgements The author gratefully acknowledges the assistance of Mr Felix Sourour of Ficksburg. This work is based on research supported by the National Research Foundation of South Africa. Any opinions, findings and

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

conclusions or recommendations expressed in this material are those of the author and therefore the NRF does not accept any liability in regard thereto.

References Bosch, J. (1967) Ladybrand 1867-1967 (Ladybrand Town Council). Du Preez, J. (2012) The historical development of farmstead architecture in the Brandwater basin of the eastern Free State until Union in 1910. Unpublished M. Arch thesis (University of the Free State). Eloff, C. (1979) The so-called Conquered Territory disputed border area between the Orange Free State (Republic of South Africa) and Lesotho (Basutoland). (Human Sciences Research Council, Pretoria). Eloff, C. (1980) Oos-Vrystaatse grensgordel: Ĺ&#x2030; streekhistoriese voorstudie en bronneverkenning. (Human Sciences Research Council, Pretoria). Floyd, T. (1960) Town Planning in South Africa (Shuter & Shooter, Pietermaritzburg). Gouvernement Courant (1867) (Free State Archives). Haswell, R. (1979) South African towns on European plans. Geographical Magazine, July, 1-8. Haswell, R. (1980) Early Afrikaner dorpe: functional and aesthetic compositions. Planning & Building Developments, July/August, 9-13. Haswell, R & Brann, R. (1984) Voortrekker Pieter Mauritzburg. Contree, 19-22. Haswell, R. (1990) The making and remaking of Pietermaritzburg: the past, present and future morphology of a South African town. In Slater, T (Ed) The built form of western cities (University Press, Leicester) 171-185. Lewcock, R. (1963) Early nineteenth century architecture in South Africa (Balkema, Cape Town). Moll, J. (1977) Dorpstigting in die Oranje-Vrystaat: 1854-1864. Contree, No 2, July, 23-28. Notulen der verrigtingen van den Hoog-Edelen Volksaad. Bloemfontein, 1868 (Free State Archives). Oberholster, J. (1969) Wepener 1869-1969 (Town Council of Wepener). Standard Encyclopaedia of Southern Africa (SESA) (1970) (NASOU, Cape Town). Van Rhijn, P & Klopper, A. (1967). Die Geskiedenis van Ficksburg (Ficksburg Town Council) Van Zyl, W. (1993) The origins of country towns in the Free State. In Proceedings of the Symposium on South African Planning History. Planning History Study Group, 73-86.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

OFF CENTRE Urban form in emerging economies and postcolonial countries

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Urban plot characteristics study: Casing Centre District in Nanjing, China Lina Zhang School of Architecture and Urban Planning, Nanjing University, China. Email: zln.nju@gmail.com

Wowo Ding School of Architecture and Urban Planning, Nanjing University, China. Email: dww@nju.edu.cn

Abstract. Public building on commercial plots plays an important role for urban development, urban tissues and downtown character. However, relatively weak control by plot indicators, with commercial plots as a land-use type in China being more than 20 years old, leads every building form to play its own role in the city. This results in the loss of the cityâ&#x20AC;&#x2122;s common spaces. In trying to ameliorate this situation and improve landscape management, urban managers and urban designers are seeking more efficient land use through indicators for the common regularities of architectural design. Research on building pattern with plot setting is needed. Since preliminary studies were mostly limited in plot size or property, the plot transformation, plot geometry and boundary characteristics have been investigated. In researching the morphological transformation process of the commercial centre district in Nanjing in this paper, the plot characteristics and its related street network has been investigated historically. The research carefully analyses the discrepancy of plot boundary characteristics and its affect on the building arrangements. This tests the role of plot in shaping the contemporary urban fabric. Keywords: morphologic element, plot character, plot boundary, urban transformation

Chinese cities, especially coastal cities, have undergone tremendous changes and development in the process of urbanisation over the last 30 years. The rapidly emerging physical form in urban areas is the direct product of modern urban planning, urban design, architecture design and the subsequent construction. Compared with current practices, however, the corresponding theoretical framework and methodologies which support urban landscape formation lag behind. In particular, the morphological knowledge about the formation mechanism of the urban fabric is still too vague (Whitehand, 2009), and this has directly led to poorly controlled practice and chaotic urban environments. Therefore, urban design theory is needed to enhance understanding and articulation of the formation mechanism of the urban fabric in order to offer greater guidance for practice. As a main city in the economically advanced area of China, Nanjing has experienced three decades of urbanisation, which has resulted in significant changes to this 1800 year old historic city. The scale of the city has increased by 6-7 times, and the city has undergone dramatic changes in physical form. In general, as with other Chinese cities, the urban fabric of Nanjing can be interpreted as the composition of four main morphological categories: Historical blocks, Scenic area, Residential blocks, and Commercial blocks (Fig. 1). Since the functions of sites tend to be differentiated by physical form, each morphological category tends to have distinctive form and can therefore be studied individually (Whitehand, 2009; Abramson, 2008). In Nanjing, the first two morphological categories were protected by urban conservation plans and are less impacted by urban development. The last two morphological categories are located in the major construction area of the city and form the main body of the contemporary urban fabric. In the case of the residential area, morphological regularities have been gradually appearing due to

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

living culture, economic and health considerations. Its plot pattern has been studied using quantitative description (Zhang, Ding, 2012; Ding, Liu, 2007). Due to the diversified design conditions, such as the various sizes and geometrical features of urban plots, the urban fabric in commercial districts does not show notable morphological regularities. Instead, it displays complexity, even chaos. In this paper, taking the XinJieKou centre district of Nanjing as an example (Fig. 2), the physical form in the commercial district is interpreted so as to search for a method for better understanding its formation mechanism.

Theoretical review and approaches Study the formation mechanism by describing the physical form There are three main ways of understanding and describing the formation mechanism of the urban fabric. Only the perspective of the morphological study directly concentrates on physical form, the form complexes and its historic-geographical structure. The other perspectives, being focused on social, economic, religious and political aspects (Munford, 1968) and a static historical view (Kropt, 1991), are concerned with spatial phenomena which affect and are affected by urban form but do not describe it. In the field of urban morphological study, the morphological analysis of towns was born out of questions addressing architectural types on the one hand and the analysis of plans on the other. These were led by the Italian School and Conzenian School respectively. Those two kinds of investigation established the theoretical framework of morphology. It has been acknowledged that M.R.G. Conzen made a major contribution by identifying plan element complexes (street system, plot pattern and building pattern) and providing a widely applicable method. However, research by urban morphologists undertakes urban fabric analyses based on training in geography, and this results in a weak relationship between research and practice. Therefore quite a few scholars have tried to bring the two methods together to provide a powerful tool for identifying and describing the physical characteristics of the urban fabric which can communicate with planners and urban designers (Kropf, 1996, 1997, 2011).

Historical blocks

Scenic area Residential blocks Commercial blocks Figure 1. The four morphological categories of urban fabric in city of Nanjing. Source: Google Earth in 2010.

As a widely accepted contribution, Conzenâ&#x20AC;&#x2122;s morphological approaches are effective for the traditional texture, notable between the late 1950s and the late 1980s. However, the approach meets a major challenge facing urban morphology today, especially for the modern cities outside of Europe. It has been emphasised by many scholars that it is meaningful to understand postmodern landscapes and apply morphological knowledge to urban design practice (Whitehand, 1992, 2009; Levy, 1999). The current morphological studies on the physical form of the modern city fall into three principal groups. The first is the application of Conzenian approaches in modern cities. Siksnaâ&#x20AC;&#x2122;s comparative study shows that Conzenâ&#x20AC;&#x2122;s approach can be employed to study the characteristics of the morphological elements of modern cities (Siksna, 1997). The second is the identification of relations between the diverse elements of urban form for an integrated framework of urban analysis (Osmond, 2010). The third is the clarification that physical form is a common reference aspect to coordinate different descriptions of urban form (Kropf, 2009, 2011). It also asserted that the form of urban fabric can be described at a number of levels of resolution (Moudon, 1979; Kropf, 1993, 1996; Gauthiez, 2004). Therefore the study of morphological elements in different levels of resolution and their interrelation should be emphasised especially for the

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

research on modern urban fabric (Levy, 1999). However, building arrangement is still not a prominent theme in urban morphology study and need to be further researched (Whitehand, 2009; Oliveira, 2013).

Figure 2. XinJieKou centre district in 2010. Source: left, Google Earth in 2010; right, the drawing of the site map is the result of a site survey based on the site map of 2007 from the planning bureau of Nanjing.

Morphological research on the plot level Efforts in searching for the method of describing the physical form mainly concentrate on the plot level. Theoretically, the physical framework of street, plot and buildings are identified as three basic morphological elements. Practically, land use parcel (plot) is consistent with the minimum unit of urban planning and is also the starting point for architectural design. This means the morphological element of plot is the key link between the two dimensional planning process and the three dimensional forming process of design. Therefore it is most likely to be used for interpreting the formation mechanism of the urban fabric. In the morphological study of plot, plot characteristics, such as plot size and the layout pattern of plots, came under particular scrutiny (Siksna, 1997). However, the results of much morphological research cannot be applied in China, as the current research only responds to square blocks or rectangular blocks while plot sizes in China are much more complicated. The transformation of blocks and plots was also important to survey, especially in the contribution by French scholars (Panerai et al., 2003). Accordingly, the transformation of block, plot and building fabric in China and its particular features has been inspected (Abramson, 2008; Fei et al., 2009; Yang, 2009; Whitehand and Gu, 2007). However, this research on plot characteristics is seldom related to the physical form. In recent years, further research on plot characteristics was undertaken for linkage to the description of physical form (Oliveira, 2013). As the comparative study by Siksna already testified, small plots produce more predictable building forms, indicating that building forms are affected by plot size. Greater rigor is needed in describing the relation between plot characteristic and the form of building pattern. In China, as typology has become devalued (Chen, 2009), the potential description of the formation mechanism of urban fabric lies in examining plot characteristics of Chinese cities and investiating the relation between plot characteristics and building pattern. Parallel research on this issue was conducted in the traditional Chinese city and the types of morphological elements are interpreted (Chen et al., 2009). However, the relative research on Chinese modern fabric is largely absent in the current literature.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Previous studies inspire the research focus as follows. First, study of the historic-geographical transformation of XinJieKou centre district is undertaken in order to understand the diversity of plots and identify plot characteristics. Second, the plot characteristics are studied to understand its influence on building arrangement. Within this section plot typological study are executed. Third, by mapping plot characteristics and the related underlying information, the relation between plot type and the form of building pattern is examined.

Morphological transformation The Nanjing XinJieKou Centre was firstly formed in 1929. The urban texture of XinJieKou district before 1929 was mostly impacted by natural factors, and in 1929 the proposed street system in The Capital Plan profoundly influenced the development of this district, which led to the complexity of the overlapped two street systems today (Fig. 3). Issued in 1929, The Capital Plan was the first modern urban plan of Nanjing (also China). After 1929, the partial implementation of the main street system in The Capital Plan triggered the XinJieKou district changing from an ordinary urban residential area to the most important shopping and transportation centre of Nanjing. This stimulated the intensive development in the district from 1929 to 1937. Later on, the construction activity was suspended for eight years due to the Anti-Japanese War.

Figure 3. Three phases of the street structure in XinJieKou district. Source: The Magic Square, 1996, pp.300, 301; Nanjing Local Chronicle on Urban Planning, 2008, pp. 15,17, 37; The capital road system in 1931 based on The Capital Plan from Nanjing Land Bureau.

During 1945 to 1976, Chinese political change and instability in the economic environment slowed down urban development. In the 15 years after 1976, although China’s economy slowly recovered due to Reform and Opening Policy, conservative urban development policies resulted in the retention of the original urban fabric. After the 1990s, when state-owned allocated land, such as danweis (work units) was brought to the land marketisation system, plot development took another step in the era of commercialisation. During this period, the ambition of constructing ‘international metropoli’ by the local government accelerated regional development. This construction boom lasted until the early 21st century. Almost until 2010 the urban street structure and plot pattern of XinJieKou district were relatively stable.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Since the previous studies about urban transformation of XinJieKou district showed that updating the urban street system was undertaken simultaneously in the plot development process (Xu, 2004), it is necessary to exam two morphological elements of street-block and plot. This paper investigates the historic-geographical transformation of XinJieKou district in relation to two aspects – street structure and plot pattern. Morphological transformation on street structure The current street system in XinJieKou district is the result of overlapping and integration of two street systems. The historical road network was closely related to the ancient water system. During the Six Dynasties (229-589AD), Canalising the YunDu (240AD) and ChaoGou (241AD), which were used for goods and passenger transportation, directly influenced the urban texture of surrounding areas (the current XinJieKou district). Most of the urban roads, built near the ancient water system and connecting the north and the south parts of the city, ran parallel to this water system and finally formed the main road of the city running southwest to northeast (about 24 degrees to the east of north). Although the ancient water system gradually declined during subsequent dynasties, it is still visible in the map of the late Ming Dynasty (1368-1644). The historical road network running south-west to north-east had already formed in XinJieKou district (front image in Fig. 3). The later street system, especially its main road structure, was primarily determined by The Capital Plan. In this modern planning, the design principle of the main road is established to be straightness, and the “ZhongShan avenue system” which is the heart of the main street system passes through the centre of the current XinJieKou district (middle image in Fig. 3). With the construction of “ZhongShan avenue system” completed by 1937, the two urban street systems with certain angles overlapped. The integration of the two road networks formed the current street structure in XinJieKou district (last image in Fig. 3). The two main roads running through the XinJieKou district are both in the direction of due north-south, which profoundly changed the direction of the historical roads system in northeast-southwest direction, while the other two main roads in west-east direction are almost consistent with the historical road network. Since The Capital Plan only implemented part of the main road structure, the branch roads within the street blocks were less influenced by The Capital Plan and obviously consistent with the historical road system (Fig. 4). Until 2000, as the branch roads were incorporated into urban street system, the branch roads of XinJieKou district have been broadened and developed.

Figure 4. The transformation of branch roads within 4 street blocks. Source: Based on Xu, 2004, pp. 25, 57, 84, 112, 138.

The plot development bears a close relationship with the formation process of the modern street system, especially in the three periods of 1929-1945, 1976-1991 and 2000-2010, with the crossing main road of “ZhongShan avenue system” completed, the “Loop Roads” around XinJieKou broadened, and the branch roads within the street blocks widened. The plots at both sides of the urban roads were always the areas with highest development priority (Fig. 5). The renovation of the urban street system also influences the plot pattern within the blocks. The current shape structures of the plots in XinJieKou district are influenced by the two direction system of the streets. The shape structures of the plots might be controlled by one of the two road systems or be influenced by both two road systems at the same time (Fig. 6). Consequently, the shape structures of plots in XinJieKou are varied, and this influences the forming process of the street structures.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 5. The plot development always has close relation with the improvement of the street system. Source: Based on Xu, 2004, pp. 22, 56, 83, 111, 137.

Fitting historical street system

Fitting modern street system

Fitting both two street systems Figure 6. With the influence of the overlapped street systems, the shape structures of plots are diversified.

Morphological transformation on plot pattern With unceasing reform in the paid land-use system in China, the manner of plot development has been changed, from government led in a planned economy to shared government and free market economy led, and eventually to market led in a market-oriented economy. However, concurrently with plot development, the planning and administration process has fallen behind. For lack of the appropriate planning policies in plot division and integration, the new emerging plots have inherited a lot morphological information from the former land owners, and this increases the complexity of plot morphology. In this process it can be read that both land use and plot size have become the important factors of variation at the same time (Fig. 7). The first mode of plot transformation means that the plot size is expanded or reduced with the same land use. The project of Workersâ&#x20AC;&#x2122; Cultural Palace (WCP) during 1945-2000 could be taken as a sample (model 1 in Fig. 7). The predecessor of WCP was the former JieShou Hall (1929-1945), an auditorium with capacity for 400 people. Due to the demand of proletariat on activity places as well as the influence of social mentality of â&#x20AC;&#x153;going and going all outâ&#x20AC;? from 1945 to 1976, the WCP project not only occupied the JieShou Hall but rapidly expanded its site. From 1976 to 1991, since the introduction of new government restrictions on commercial development, cultural and educational projects were encouraged. The WCP project continuously expanded its land acquisition and in this period a bookstore and science and education centre were built. During 1991-2000, with the establishment of a market economic system and the recovery of free market economy, lots of the land on the site of WCP was transformed. The original buildings were either demolished or transformed for other uses, such as office buildings, commercial and entertainment complexes, and the original outdoor public places were changed to car parks. Only the science and education building of WCP was retained. After 2000, the site of the science and education building was incorporated into the surrounding plots and formed a new plot for commercial use. Until then, the project of WCP eventually disappeared from XinJieKou district. The expansion and contraction process of the plot in the WCP project over 50 years represents a dramatic change to plot size and plot shape.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 7. Three plot transition models show plot characteristics change in two aspects: land use and plot size.

In the second mode of plot transformation the land use is changed while the plot size and shape remain almost the same. The project of Nanjing Cultural and Art Centre (NCAC) during 1965-2000 is an example (model 2 in Fig. 7). In 1965, a four storey factory of Nanjing Machinery Plant was built on the site designated for industrial land use. In the 1990s, the land use of such plots was transformed from industrial to public and exhibition. Subsequently, the six storey NCAC project was built in 1999 and has been used up to now. In this mode of plot transformation, there is no change on the plot shape, but the change of land use may cause a series of changes to plot access location and building arrangement. In the third mode of plot transformation both the land use and plot size are changed. The project of International Financial Centre of China Merchants Group (IFC) from 1996 could be taken as an example (model 3 in Fig. 7). Before 1991, a series of land use units occupied the site of IFC, such as a fuel company, traditional folk houses and multi-storey residence. From 1996 to 2008, the project of IFC acquired its site by land auction, and seven storey podium buildings, 28 storey auxiliary tower and 56 storey main tower were constructed successively. The emergence of new plot lead to a new access position for the plot, and also caused boundary changes with adjacent plots. The different plot transformation modes resulted in differences in the aspects of plot size and plot shape as well as the distinct morphological characteristic of plot. Statistics study on the morphological transformation in XinJieKou district The transformation of street structure and plot pattern in XinJieKou district eventually produced 53 plots with significant morphological differences (Fig. 8). The previous historic-geographical analysis of plot usage, plot size, plot shape and plot boundary condition are the important aspects of discrepancies in plot characteristics. The statistics on branch road numbers and plot numbers in the transformation process in XinJieKou district show that the process of plot development accompanied the maturation of the urban street system (Fig. 9). From 1940s to 1990s, with the concentration of urban infrastructure, the plots experienced the period of fiercest change. Up to 2010, with the stabilisation of the urban street system, the plots have gradually stabilised. This indicates that the morphological characteristics of plot were closely influenced by the morphological characteristics of street.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 8. The discrepancies of current plots exist in the aspects of land use, size, shape, and boundary condition.

Figure 9. The statistics of morphological transformation in XinJieKou district: number of branch roads, number of plots.

Plot characteristic study Drawing on the above analysis, the main morphological distinctions of the plots exist in four aspects: usage, size, shape and boundary condition. Consequently, in order to observe the role of different plot characteristics on the formation mechanism, further investigation of the relationship between those plot characteristics and the form of its building fabric is needed. In this paper, the land use of commercial buildings, office buildings and hotels (referred to as â&#x20AC;&#x2DC;commercial plotâ&#x20AC;&#x2122;) are taken as the objects of morphological analysis. In XinJieKou district, 24 plots with built fabric were selected from 39 commercial plots, excluding plots designed for a commercial land use which were either under construction or without construction development yet. To obtain a more accurate analysis result, 45 commercial plots were selected from the built physical of HuNanLu, another important commercial district in Nanjing which experienced a similar plot forming process as XinJieKou district. This means the study is comprised of 69 commercial plots. The relationship between plot characteristics and building fabric Usage: the relationship between land use and physical form should acknowledge that physical form is closely related with functional organisation, such as the different land use for commercial building and hotel. The interpretation of the physical form of the commercial district depends on sorting land usage. It is one possible path for research, but it counts against seeking the general morphological characteristics and the formation mechanism of the urban fabric in the whole commercial district.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Size: plot size directly relates to the urban planning indicators, such as Floor Area Ratio (FAR), coverage and building height limitation. Those indicators defined the building form quantitatively not physically, and building layouts could be very different in the same size with same indicators. Shape: in general understanding, plot shape impacts on building shape. However, the former study in this area has shown that there is no direct relation between shape ratio of plot and the shape ratio of building layout in Nanjing (Ding et al., 2007). The result reveals that building shaping cannot be effectively interpreted by plot shape. Boundary condition: on the morphological level of plot, since the block-plan of commercial plots relate to pedestrian flow and car flow around the site, it is reasonable to hypothesise that the plot boundary could be treated as a factor for better understanding the building arrangement within the plot. Additionally, at the morphological level of street, the boundary conditions of plots are closely worked with the hierarchy street system of Nanjing. In this city, as well as all over the country, the road network is composed of the classified roads by its width, namely the main road above 40m width, the secondary main road of 25m to 40m width, and the branch road of 12m to 2m width. Each plot boundary corresponds to its adjoining road of a certain grade (Fig. 10). If there is certain relationship between the hierarchy of the street system and car/pedestrian flow within the plot, the plot characteristic of boundary condition could be employed for interpreting the general morphological characteristics and the formation mechanism of the urban fabric in the whole commercial district.

Figure 10. The hierarchy of the street system and the related plot location

The statistics show that pedestrian activities in plot tend to adjoin the higher ranking road than the lower ranking road (Fig. 11a). The open square for public activities tends to be located close to the crossing of main roads and secondary main roads (Fig. 11b). Furthermore, by investigating the block plan within the plot (Fig. 11c) the adjacency between building layout and plot boundary tends to be located next to the boundary of branch road. The adjacency declines when the road ranking is higher. These statistical studies clearly identified that there is reliable relativity among the hierarchy street system, car/pedestrian flow within plot, and block plan, which could support morphological analysis for the universal rules of the whole commercial district. Unfortunately, the earlier research did not put a high value on boundary condition, probably because morphological study in those countries does not involve this issue. By contact, in the case of Nanjing, the characteristics of the plot warrant further investigation. Notably, not all plot boundaries next to the urban street system could be used for car access location. According to the Code for Design of Civil Buildings in Nanjing, from the security perspective, car accesses are not permitted within 70m from the crossing of the main road (Fig. 12). Therefore, some planning policies on urban landscape management should be taken into account during the following analyses.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

(a)

(b)

(c) Figure 11. (a) The pedestrian activities in plot tend to be located adjoining the higher ranking road rather than the lower ranking road. (b) The open square for public activities tends to be located close to the crossing of main roads also secondary main roads. (c) The adjacency between building layout and plot boundary always occurs next to the boundary of branch road, and the adjacency declines when the ranking of road gets higher.

Figure 12. The positions of car access are limited by planning codes in Nanjing

Plot boundary type In order to understanding the relevance between plot boundary characteristic and the form of building fabric, the concept of â&#x20AC;&#x2DC;plot boundary typeâ&#x20AC;&#x2122; was developed as a tool for the study. Two steps are needed to define a plot boundary type. In the case of Nanjing, the plot in XinJieKou district has diverse shapes, and the boundary conditions of the plots can be very complicated, particularly length and orientation. The first step focuses on how to understand the plot shape. Since the quadrangle especially the rectangle is the most common shape of the

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

plot in cities, for this research the various plot shapes were simplified as a quadrilateral. For example, the plot with triangular head could be understood as the length of a side equal to zero, and other irregularly shaped plots could be understand as a quadrilateral with one or two missing parts within its corner (Fig. 13). In this way, the quadrilateral plot model could serve as the basic plot type in this research.

Figure 13. The various plot shapes can be simplified.

The second step concentrated on assigning the significance of the boundary of the quadrilateral plot model to the same meaning as the plot boundary condition (Fig. 14). There are four basic kinds of boundary conditions in the cases: next to main road, secondary main road, branch road, and adjoining plot. The principle of translating the plot boundary conditions to the quadrilateral plot model is as follows: ď&#x201A;ˇ ď&#x201A;ˇ ď&#x201A;ˇ

if the jointed plot boundaries have different characteristics, they will be counted as two boundaries if the angle between the boundaries with same characteristics is less than 90 degrees, those two arms must be counted as two boundaries if the angle between the boundaries with same characteristics is equal or more than 90 degrees, those two arms could be counted as one or two boundaries recognising a maximum of four different boundary characteristics in the quadrilateral plot model.

By such two steps, it was possible to express the plot boundary conditions of 69 cases using the quadrilateral plot model. At the theoretical level, by the mathematic method of permutation and combination, it clearly shows that there are a total of 55 types of the quadrilateral plot models (Fig. 15). After translating the plot boundary conditions of 69 cases to the quadrilateral plot model, the statistics of the quadrilateral plot model, shows that, in Nnajing, main roads or secondary main roads never exist in the opposite position of the plot simultaneously (Fig. 16). This is very different with the situation in major western cities. With the help of the quadrilateral plot model, further analyses of building arrangement are executed. At first, the position of car access on plot edges was studied, and the data shows that car access in the corner of the boundary holds absolute superiority than in the middle (Fig. 17a). Then, the relationship between the position of corner car access and plot boundary conditions was studied, and the statistics show the corner car access is prevalent on the plot boundary of the lowest-ranking road (Fig. 17b). Furthermore, the findings in relation to corner car access position and the corner of the plot shows that corner car access is prevalent on the plot corner next another plot (Fig. 17c). These findings are further illuminated in Fig. 18. The similar findings from the study of pedestrian flow highlights that open squares for public activities tend to be located next to the corner of highest ranking road (Fig. 19).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 14. Translate the characteristic of plot boundary to the quadrilateral plot model.

Figure 15. All types of the quadrilateral plot model.

Two key findings emerged from the study of plot boundary type. First, not all plot boundary types exist in Nanjing and those plot boundary types in Nanjing have specific features. Second, plot boundary types imply certain information about building arrangement, such as pedestrian flow and car flow which influence the location of the building within the plot. For the commercial plot in Nanjing, this means that one plot boundary type probably has a preferred building layout. Experiment In the experimental stage, all the underlying plot information was loaded back into the original site map of XinJieKou centre district: the hierarchy of the urban streets, the car accesses limitation by the planning code, as well as the preferred car accesses locations and most likely pedestrian activities. This was called the plot boundary condition map (Fig. 20), which exhibits the complexity of the plot boundary situation. The research process in this paper indicates that all the site maps of the commercial district could be translated into the plot boundary condition map, which offers the reasonable design suggestions for architects.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 16. The statistics of the quadrilateral plot model in Nanjing shows that main roads or secondary main roads never exist in the opposite position of the plot.

(a)

(b)

(c)

Figure 17. (a) Car access in the corner of the plot edges holds the absolute superiority than in the middle. (b) Corner car access prefers the plot boundary of the lowest-ranking road. (c) Corner car access prefers the plot corner next to another plot.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 18. Car accesses prefer to be located on the corner of the lowest-ranking boundary.

Figure 19. The open squares for public activities prefer to be located next to the corner of highest-ranking road.

Figure 20. Plot boundary condition map.

By translating the site map of commercial plots back to the quadrilateral plot model, the complexity of site conditions in Xinjiekou district for architectural design could be further clarified. There are 13 plot boundary types from 24 commercial plots (Fig. 22), which means 13 kinds of building layout appeared. Once again, it illustrates the reason for the chaotic urban fabric in XinJieKou centre district of Nanjing.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 21. Proving the relevance between the plot boundary characteristics and building arrangement.

Figure 22. The multiple type of the quadrilateral model exhibits the complexity of the plot in its aspect of boundary.

Conclusion and Discussion This study has focused on the morphological level of plot by studying a sample from the commercial district in Nanjing. The plot development and its related street network changing has been investigated historically. In order to test the role of plot in the formation of urban form, the plot characteristics in the contemporary city of Nanjing has been examined.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

The articulated study of the historic-geographical transformation of Xinnjiekou centre district has shown the process of plot development was influenced by two main factors. The first influence is the development of the urban street system, including two main aspects of modern planning, primarily the master plan in 1929 and the changes in various branch streets. The second influence is the process of socio-economic development. The constant interacting of those two factors eventually led to many differences in plot characteristics, especially in irregular shape and multiplex boundary conditions, which represents the urban morphological characteristics of this district. This study validates the morphological methodology developed by both Conzen and Caniggia and its methodological relevance, as well as demonstrates that the morphological approach is still meaningful in contemporary urban fabric studies. In this paper, the plot is a very important morphological element not only in interpreting the physical form but in the discussion of its formation mechanism. Since plot acts, on one hand, as the land use unit in urban planning, and on the other hand as the site conditions closely related with architecture design, plot was proven to be the vital carrier for a better understanding of the formation mechanism of urban form. In contemporary China, given the huge amount of construction during a tremendous process of Chinese urbanisation, enhanced understanding of the morphological element of plot and its characteristics is even more significant than ever before. The case study in Nanjing demonstrated that plot characteristics deserve to be carefully investigated. With the help of the quadrilateral plot model, the plot boundary types were identified and research on plot boundary condition was undertaken. As an effective control on building arrangement, plot boundary type developed perceptions of the aspects of plot setting especially in Nanjing. This method could be applied in other districts or cities for their morphological studies. The site map could be easily translated into the plot boundary condition map to assist in studying the morphological characteristics of the whole district. The case study of XinJieKou district has shown that plot boundary is the crucial issue for interpreting the urban form of this district, and more cases in the city need to be analysed to attain clearer categories of plot boundary type. This study suggests that the meaning of the plot should be further excavated and the method of Plot-Typology deserves to be further development. Through more discussions about the method of Plot-Typology, the interaction between plot usage, geometry (size and shape), boundary and other related factors as well as their influence on physical form, could be further investigated.

Acknowledgements This research is jointly funded by The Doctoral Fund of the Chinese Ministry of Education (Program No. 20120091110055), the Natural Science Foundation of Jiangsu (Program No. BK2009244) and Doctoral Creative Project of Jiangsu (Program No. CXZZ12_0059). The authors are indebted to the helpful academic exchange with J. W. R. Whitehand, Susan M. Whitehand and Kai Gu in Nanjing, and to the anonymous referees for their valuable comments.

References Abramson, D. B. (2008) ‘Haussmann and Le Corbusier in China: land control and the design of streets in urban redevelopment’, Journal of Urban Design13 (2),231-256. Schinz, A. (1996) The Magic Square: Cities in Ancient China, (Edition Axel Menges, Stuttgart/London). Chen, F. (2009) ‘Typomorphology and the crisis of Chinese cities’, Urban Morphology 12(2), 131-33. Chen, F. and Romice, O. (2009) ‘Preserving the cultural identity of Chinese cities in urban design through a typomorphological approach’, Urban Design International 14, 36-54. Code for Design of Civil Buildings 2005 (GB 50352-2005), (Architecture & Building Press, Beijing). (In Chinese) Conzen, M. R. G. (1969) Alnwick, Northumberland: a study in town-plan analysis, Institute of British Geographers Publication 27, 2nd edn (Institute of British Geographers, London). Ding, W. W. (2007) ‘Nanjing urban morphological study and planning control policy’, unpublished research document, Nanjing University, China. (In Chinese) Ding, W. W. and Liu, Q. (2007) ‘The Resolution at Cognitive Scale to Urban Physical Spatial Form’, Modern Urban Research (8). (In Chinese)

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Gauthiez, B. (2004) ‘The history of urban morphology’, Urban Morphology 8(2),71-89. Kostof, S. (1991) The City Shaped: Urban Patterns and Meaning though History, (Thames & Hudson, London). Kropt, K. S. (1993) ‘An enquiry into the definition of built form in urban morphology’, unpublished PhD thesis, University of Birmingham, UK. Kropf, K. S. (1996) ‘Urban tissue and the character of towns’, Urban Design International 1(3),247-63. Kropf, K. S. (1997) ‘Typological zoning’, in Petruccioli, A., (ed.) Typological process and design theory, 127-40, (Aga Khan Program for Islamic Architecture, Cambridge, MA). Kropf, K. S. (2009) ‘Aspects of urban form’, Urban Morphology 12(2), 105-20. Kropf, K.S. (2011)‘Urbanism, politics and language: the role of urban morphology’, Urban Morphology 15(2), 157-61. Levy, A. (1999) ‘Urban morphology and the problem of the modern urban fabric: some questions for research’, Urban Morphology3, 79-85. Liu, Q. and Ding, W. W. (2014) ‘Morphological study on the unit of urban fabric of contemporary residential plots in Yangtza River Delta, China’, (ed.) New Urban Configurations, 689-694. (2012 International Conference on New urban Configuration, IOS Press, Amsterdam). Moudon, A. V. (1997) ‘Urban morphology as an emerging interdisciplinary field’, Urban Morphology 1, 3-10. Munform, L. (1968) The city in history: its transformations and its prospects, (Marine Books,). Nanjing Local Chronicles Compilation Committee, (2008) Nanjing Local Chronicle on Urban Planning, (Jiangsu People’s Publishing House, Nanjing) Oliveira, V. (2013) ‘Morpho: a methodology for assessing urban form’, Urban Morphology 17, 21-33. Osmond, P. (2010) ‘The urban structural unit: toward a descriptive frame work to support urban analysis and planning’, Urban Morphology 14(1), 5-20. Panerai, p., Castex, J., Depaule, J. C., and Samuels, I. (2003) Urban Forms: The Death and Life of the Urban Block, (Architecture Press, Oxford, UK). Siksna, A., (1997) ‘The effects of block size and form in North American and Australian city centers’, Urban Morphology 1, 19-33. Whitehand, J. W. R. (1992) ‘Recent advance in urban morphology’, Urban Studies 29, 617-34. Whitehand, J. W. R. and Gu, K. (2007) ‘Extending the compass of plan analysis: a Chinese exploration’, Urban Morphology 11, 91-109. Whitehand, J. W. R. (2009) ‘The structure of urban landscapes: strengthening research and practice’, Urban Morphology 13(1), 5-27. Yang, Z. and Xu, M. (2009) ‘Evolution, public use and design of Center Pedestrian Districts in large Chinese cities: A case study of Nanjing Road, Shanghai’, Urban Design International 14(2), 89-98. Zhang, L. N. and Ding, W. W. (2014) ‘Density, height limitation, and plot pattern: quantitative description of the residential plot in Nanjing China’, (ed.) New Urban Configurations, 679-688. (2012 International Conference on New urban Configuration, IOS Press, Amsterdam). Xu, N. F. (2004) ‘Study on urban transformation of XinJieKou district in Nanjing’, unpublished Master’s thesis, Nanjing University, China. (In Chinese)

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

The Denied City: how the crisis is leading cities to the edge Nicola Marzot Faculty of Architecture and the Built Environment, Department of Architecture, TU Delft, The Netherlands. Email: N.Marzot@tudelft.nl

Abstract. Urban form has always been the historical evidence of any process of civilisation, recording its driving forces – at political, social, economic and cultural levels – in relation to the inertia to modification of the existing built conditions. If the Territory defines the dynamics internal to this process, the Landscape identifies its immediate output and aesthetic dimension of where it is experienced. Since the second half of the 1980s, the increased widening of the market at a global scale abruptly changed the traditional relation between the Territory and the Landscape, based on spatial continuity. The overwhelming mobility of people, information, goods and finances interfered with local conditions and affected the related urban form. If not necessarily designed, the new international strategy was clear in its own ambitions. It multiplied differences and discontinuities within the consolidated territorial framework, by favoring locations characterised by a high level of intermodal accessibility. Since 2007, the global financial crisis resulted in an unpredictable direction to the aforementioned ongoing process. The immediate real estate shrinkage saw multiplied building vacancies and waiting lands. This phenomenon has continued to escalate and offers the opportunity to retroactively think about a completely new urban configuration, where the unconventional use of edge areas can trigger new kind of urbanities and subjectivities. This paper aims to describe the phenomenon, tracing its history and showing its potential, with broad reference to Europe and specific focus on Rotterdam, the Netherlands and Bologna, Italy as case studies. Keywords: vacant buildings, waiting lands, temporary use, urban practice, building type

A political definition of the urban form The Territory is defined by the perennial dynamics internal to any process of civilisation among its driving forces at social, economic and cultural levels (Conzen 1960). The Landscape identifies the output and aesthetic dimension of where these dynamics are experienced. Its historical relation to Territory has always accounted for the inertia to modification of existing natural and built conditions, reciprocally arranged according to different levels of complexity, or “scale”, as an implicit limitation to any strategy of change. This implies that the definitions of Territory and Landscape inevitably tend to Idealism to the extent that their actor intentionality is forced to face the original conditions they aim to overcome as an undeniable objective basis for any attempt of corresponding reform (Muratori, 1959). Nonetheless, they are indispensable parameters to understand the transformation of the built environment. Furthermore, the same drivers of change need time to reach a fully developed self-awareness and to start claiming a subsequent leading role within a new phase of the civilisation process. Under these conditions, urban form therefore implies a collective will, considered as a clear expression of a political intention performed by the abovementioned leading forces, to be translated through the Project of the City as the utmost result of a new territorial vision. This explains how, since its inception, the project was meant to act as an instrumentum regni through which it was possible to conceive, project, realise and control a social reality coherent to the power relations performed by all those claiming an active role in the transformation and management of the existing territory. Consequently, any kind of set

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

of rules related to land subdivision, settlement location and definition of building, just to mention few among all aspects of the anthropic process, was more than a mere technical act, and identified itself with the foundation of the same political order prompted by any phenomenon of civilisation. Vitruvius Pollio (80 B.C-15 BC), commonly considered the Father of western building civilisation, was fully aware of it, and symbolically addressed its treatise on Architecture as the Art of Construction to the young Emperor Augusto, to make the related framework clear and transmissible over the centuries up to the contemporary era (Vitruvius Pollio, 1999 ed.). His statements were particularly effective and incisive in the field of Monuments, but are applicable to any artificial output. All Public buildings always required a high level of complexity to set a building site, to select, move and prepare on site materials, to coordinate the different implied labour force skills, to prepare one-to-one partial samples and even to build appropriate machineries. Every architect involved in institutional duties was, and still is, first of all, a job manager, without a clear distinction in term of responsibilities from an engineer. The more challenging the task, the more impressive the required ability to control all aspects of it through space and time. This inevitably leads to saying that any Architectural Order embodied within the project reflects a civil order through its physical realisation, according to a sort of ante litteram conscious panoptical relation between society and the related material culture. In different terms, one can say that for every public building, for its own implicit complexity, its inception establishes a power relation: through its project principles, through its set of rules for realisation and through its shared use behaviour, it becomes an instumentum regni. This capacity increases and consolidates whenever the character of the building is endlessly repeated. The widespread diffusion of the law implicit in any architectural performance, which institutes a sort of social ritual, therefore defines a historical norm that echoes far beyond the specificity of architectural discipline boundaries. If the phenomenon is evident within the production of public buildings, it also manifests whenever the body of knowledge implied by any private building, like a simple house, is repeated so many times that it translates into a shared set of values, contributing to the process of civilisation development. Consequently, the type concept expresses the collective will embodied within the repetition of individual building acts. These share the same order, identifying at each level of complexity of the built environment, i.e. the “scale”, the enduring persistence of a political project, being the Project of the City within its Territory. To fully exploit its potential, that required will must run through the subsequent stages of intentionality expression (the Project), physical evidence (the Realisation) and shared use (the Behaviour). The technique itself cannot ever overwhelm the Project because it is instrumental in its deployment and embodiment, and it has to be rooted into the general framework of the social reality.

Urban form change and politics When the conditions under which the urban form acts as a political device are assumed, together with the crucial role played by the concept of the building type within the related framework, it becomes very interesting to focus on the deep meaning of “urban change”. Regardless of the fact that it occurs at the level of the Project, the Realisation and/or the Behaviour, urban changes always reveal a transformation process within the existing social reality and related drivers. The social reality defines an unstable equilibrium among all those forces that claim a specific and distinct role in its inception, establishment and persistence, and one may assume it as conventionally stated within specific historical conditions. However, the power relations established and prompted by any process of civilisation do not have the same effectiveness even within a supposed period of internal societal stability. They manifest differently, according to their proximity to institutional centres, where the effects mostly reverberate. The dialectical relation between central and peripheral urban locations therefore assumes an unexpected value once highlighted by a political perspective: it emphasises the limitless conflict between the leading agents of transformation and the supposed emergence of new ones. This is why the edge condition always plays a crucial role within any subversion of the existing urban and territorial status quo. The term “edge”, better than any other, expresses the existence of a situation which manifests a weaker degree of control with respect to a centre, independent of any kind of topographical relation.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Within a well established social reality, conventionally assumed as homogeneously distributed, the edge/centre relation expresses a different quality of corresponding social, economic, cultural and political constraints as well as obligations to respect. As a consequence, an edge condition always offers fertile ground to cultivate unexpected forms of demand at all levels. The weaker institutional coercion, the stronger the capacity for experimenting about new possibilities. This basic distinction becomes fundamental to understand the important changes that occurred within the western civilisation process over the last decades. Almost all European countries witnessed, since the second half of the 1970s onward, a major transformation of their own territories due to an overall redistribution of former manufacturing locations, usually within the body of the consolidated urban tissue. The process of change tended to favour new locations, very close to intermodal hubs, as an immediate response by the leading production forces to the increasing widening of the demand/offer market. As a consequence of this relocation phenomenon, major urban centres quickly and increasingly have become porous, offering a wide spectrum of waiting lands and vacant buildings in strategic positions to the real estate market. Impressive refurbishment programs were launched by Paris, London, Barcelona, Amsterdam and Berlin, among the most highly rated urban centres, that systematically made use of master plans to fix the vision of the new growth perspective and called international competitions to develop it on a broader cultural framework. The corresponding diffusion of capital investments by local and state institutions pulled the transformation process, making use of the available territorial labor force. The newly accessible emptiness was almost immediately filled, and assimilated to the existing residential city, coherently to the 19th century bourgeois model, even if differently interpreted according to local constraints and traditions. A decade later, the overall situation had already changed. Originally fostered by new international manufacturing chains, the industrial relocation process in close proximity to highly congested intermodal junctions was soon followed by retail and service markets. Logistics proved its importance, while systematic use of territorial marketing strategies succeeded to attract private capital to the new urban locations, which aimed at competing with the existing ones in order to multiply business opportunities. Lille soon become the leading example of a new territorial vision, witnessing the progressive rise of urban networks, a new unstable constellation of polarities claiming increasing importance with respect to traditional metropolitan entities (Castells, 1996). These configurations triggered extreme consequences through a silent but pervasive process that started at the very edge of the former stable situation. The multiplication of territorial entities – each of which expressed different power relations and interests – occurred within the same civil framework originally assumed as internally coherent and homogeneous. The related landscape fragmentation, once recognised and mapped, becomes clear evidence of the socalled globalisation phenomenon with its deep political implications (Koolhaas, Boeri, Sanford, Tazi, and Obrist, 2000). From this moment onward, the edge/centre relation was overwhelmed by the local/global one, and traditional politics were threatened by new international driving forces at an unknown level (Bauman, 2000). As a side effect of this phenomenon, urban networks progressively attracted, within their impressive gravitational system, the most charming international metropolis, already characterised by a high level of intermodal accessibility. Private equity investments carefully selecting locations, coherently positioning themselves to guarantee the most profitable conditions. This repeated strategy of “impatient capitalism” (Merotra, 2011) introduced evident distortions within the edge/centre dialectical relation as it occurred before. This new process of civilisation was not negative and often broke through local power relations established in Europe since the consolidation of the Welfare State, after the Second World War, which was progressively substituted and impoverished by the rise of Bureaucracy (namely the Power of the Office). This was systematically prompted and controlled by political parties, without any care of existing State democratic institutions which were expected to safeguard and interpret, especially in the Mediterranean area. However, the main consequence of the globalisation process was an overall redistribution of forces and related competencies at different scales, where the original reciprocal coherence was lost, although carrying a different inertia to modification. This dilemma, which is well represented by the contemporary strife between European and national institutions – at political, social, economical an even cultural level – is still acting at a deep level and represents a crucial topic to face.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

The Denied City as new edge evidence Within the framework described in the previous section, the 2007 global financial crisis has impacted worldwide and enhanced the consequences of the global/local tension to the extreme. From that moment, any kind of vacancy prompted by the multiplication of new territorial orders within a supposedly well established former social reality becomes a dramatic unresolved vulnus in the existing Project of the City and its Territory – its legacy and eventually its right to existence is questioned. In fact, neither private nor public interest could afford the investments required to process any kind of possible transformation, both on a local and/or on a global scale. The Brownfield was not anymore a temporary condition within the ongoing process of urban transformation, regardless from the related driving forces. In fact, the latter, which are of course responsible for having conceived and constructed the “world” we have inhabited and inherited up to now, are not any longer in the condition to perform it. The almost embarrassing endurance of the vacancy phenomenon simply witnesses that the leading force which acted over the last decade have completely exhausted their commitment to civilisation and cannot make step ahead. Notwithstanding the dramatic evidence of facts, they are simply postponing solutions, or taking time, threatened by the implicit menace of a prolonged inactive interval without proposals. None is taking seriously the possibility that they should “leave space” to new forces to drive the change and to experiment new policies. This radical solution of course would imply two conditions to come into existence: first, the Project of the City “moratorium” and, second, the search for new edge conditions. The former aspect is implicit in the same definition of the Project as the rational result of a collective will. This will implies a self-consciousness, which is expressed by the new driving forces, and is still very far from being self-aware. Without a new idea of the collective, the Project should be suspended, and not founding social actors to be performed, paradoxically confirming the brilliant provocative solution offered by the Nobel Prize winning Italian dramaturge Luigi Pirandello in his masterpiece, Six characters in search of an author (Pirandello, 1921). The latter aspect should cater to increasing vacancies, spread throughout European cities, as the uncontestable evidence of a lost world without any possibility of resurgence. The critical mass of vacancies ultimately defines a forgetful “city within the city”. They are not filled as if according to a nonexisting project, and can be left available for the unexpected and the unpredictable. They comprise an unrequested urban capacity to make space for experimental practices, involving individuals to free their creativity from social constraints and obligations, to pour it into those “condemned to void” areas, and to simply observe reactions without any specific expectation of durable traces heralding a new beginning. Or, maybe, it would be wise to take advantage of this new condition and survey a new possible blossoming of social reality. This would entail working in the informality which spontaneously spans in between the “not yet” and the “not any more”; the exceeded codification and the potential one. Loosing the challenge offered by the overwhelming urban vacancy implies condemnation to a “Denied city” with its potential inhabitants behaving as if they were its unconscious prisoners. Some Intellectuals, sharing a compelling adhesion toward a romantic vision of life, applaud this landscape of ruins as the ultimate evidence of Capitalist decay. They plead to preserve it untouched as the working Monument to its incapacity to face contemporary problems and drive the change, and enthusiastically proclaim it to be engulfed by an unplanned gorgeous nature (Clément, 2003). Many planning Technicians, not accepting the idea of the Project as a political instrumentum regni and irresolutely claiming for its neutrality and independence from any historical expression of collective will, simply avoid facing the problem. They adduce the impossibility of founding a solution based on the existing set of rules and their legality, as if it was neutral too, to implicitly save the Plan as a sort of imperfectible Monument to a supposed Absolute rationality (Aureli, 2008). But within the social reality nothing occurs without reason or, in other terms, there is no space for any form of Negative Thinking (Cacciari, 1976). The Denied City is not accessible to anyone, not observable by anyone, and not attemptable by anyone. It is already a working reality, the ultimate counter project of the extinct drivers of civilisation which deliberately, by erasing the vacancy existence and memory and softening the debate on it, declare it as “out of our time reach”. The implication is no to surrender to the vacancy threat. The acquired privileges are deliberately

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

defended and not left to others for chance claims. To completely understand the implication of the abovementioned scenario it is helpful to compare two different European case studies which both face the new real appearance.

From paternalism to conservatism: Rotterdam, the Netherlands vs Bologna, Italy The Dutch policy on vacancy has endured since the 1970s when, in order to counteract illegal occupation of unused or abandoned buildings, the State officially introduced the anti-kraak (i.e. anti-squatting) procedure. To find a solution between the apparently diverging expectations of the owner and the potential occupier, it legalised the temporary use of vacant buildings, waiting to be refurbished without a definite schedule, after a 12 month vacancy. Such occupancy is under obligation to leave it as soon as officially required by the owner for their own unquestionable purpose, simply affording expenses for ordinary maintenance. This compromise, in political terms, fitted in the long social-democratic tradition of Dutch Paternalism, which is a subtle blend of Protestantism, mercantile Pragmatism and humanistic cooperation, which avoids any kind of conflict for the sake and maintenance of social stability. By accepting the deal which guarantees security from undesired illegal occupation and limits any process of additional building decay, the user could temporarily have an interesting space for their own needs at an incredibly favorable rate. The owner could temporarily profit from implicit control. The decision was so successfully applied that it was Constitutional Law until 2010 and it is still used as a standard procedure to fill vacancy gaps. This means that the Dutch were not unprepared in facing the new worldwide crisis when it dramatically affected the local building market. Previous to the crisis, this procedure was intended as a temporary stage within the ordinary building value production cycle, with the implicit role of informally triggering it. Now it is systematically used to face vacancy as a specific way of experimenting with new urban practices, and improving participatory approaches with respect to more exclusive and standardised protocols. Making use of not yet codified drivers – at social, economic and cultural levels – seems non-threatening to both public institutions and private partners. It is considered as an en plein air and very successful laboratory to develop new tendencies that cannot spontaneously spring from well-established procedures, and used during periods of stability, with their own implicit obligations. A politically “guided radicalism” is therefore the Dutch recipe for facing the crisis, where Mauss’s social science distinction between the Individual and the Person become instrumental to overcome the economic shortage, once deprived of any political implication (Mauss, 1938). Rotterdam, among the most important urban centres in the Netherlands, shows an incredible dynamism that counteracts the vacancy phenomenon. The local Municipality, immediately after the crisis outbreak, started mapping the new reality, recognising the impossibility of proceeding further with the great re-densification program. This program was officially launched at the beginning of 2000 within the city centre to improve tax income, and granted by astonishing master plans negotiated during times of abundance between the private partners and the public authorities. As an additional result, it accepted the advisorship by a local well known architect, Henk Hartzema, to record vacancies on a systematic basis (Fig.1) and integrated the municipality dense database with deep on-site analysis, to size and investigate the phenomenon entity (Hartzena, 2012). Afterward, it officially launched a website called Plug Rotterdam (http://www.plugrotterdam.nl), under the auspice of AIR (Architecture International Rotterdam), where the use proposals for any specific site and the ongoing process could be judged. The website also publically marked their popularity and shared a bottom-up vision of the city within an enduring period of standstill. The Italian situation seems to be completely different. Occupation of unused building started at the very end of the 1960s as a radical declaration, not necessarily political, against building speculation, which resulted in empty spaces within the most crowded central locations, to artificially increase real estate pressures on new peripheral areas in order to transform them. Due to the general lack of independence of Parties from local political conveniences, it was impossible to guarantee a limit in the legality of the occupation in favor of the owners, who were forced to address endless and often unsuccessful protests to the National Court. The situation matured as a widespread distrust about the process. Under these conditions, it has never been possible to face the matter at an institutional level, except in very specific

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

situations. The Milan Metropolitan area is one of these. In recent times, the Municipality of Sesto San Giovanni, which owns the ex Officine Falk steelwork area, an important brownfield, which is waiting for a major transformation after the financial crisis blocked Renzo Piano Building Workshopâ&#x20AC;&#x2122;s very ambitious high rise proposal. A public competition was launched to find temporary occupiers involved in the broad field of creativity. Under a contract of three years at no rent, they are now obliged to offer monthly public facilities to the local community. These facilities deal with their own business body of knowledge through the legal form of a supposedly convenient barter for both the involved parties. However, nothing has ever been officially offered to private owners facing increasing vacancies through a systematic legalised framework.

Figure 1. Klein and Fijn Rotterdam. Map of the city vacancies within the bombardment fracture line Source: Studio HARTZEMA copyright

The City of Bologna, over the last decade, has been confirmed as an extraordinarily potent Laboratory of social participation and related practices. It appeared for the first time during the 1960s under a unique combination of well-prepared local technicians, a diffused sense of political participation via the

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

administrative reform of Quarters, and a subtle equilibrium between social welfare and economic solidarism. Under the auspices of both the Communist and Christian Democrat Parties, the well known Program for the Historical Centre Restoration was launched to re-inhabit it after years of abandonment and decay according to a shared political vision of architecture (Cervellati, Scannavini, and De Angelis, 1977). The so called “Golden Age” of Bologna’s urbanism, internationally well known, was unsuccessful because it ultimately failed to guarantee the old occupants movement back to their homes once transformed coherently and to the quality standard fixed parameters. However, it had indisputably reestablished the civil capacity of architecture to fully activate society after a period of episodic uncertainties. More recently, the city had a new opportunity to reaffirm its legacy, but local authorities are not reacting as they should. The city hosts an important real estate compendium belonging to the State, whose components are mostly scattered in a profitable position within and around the edge of the historical centre (Fig. 2).

Figure 2. Bologna North West Sector vacancy map Source: Studio Performa A+U copyright

It is strategic and highly qualified for its intrinsic monumental value that the city was selected in 2007 by the Italian Ministry of Economics to launch the first national Piano Unitario di Valorizzazione (Unitary Valorisation Plan), to activate public-private partnership for its improvements and use. The overlaps with the financial crisis almost immediately revealed the plan was based on the old principles, set of rules and expected behaviours. Notwithstanding the crude evidence of reality, local authorities insisted on following its protocol. Meanwhile, the overall amount of urban vacancies was incremented because all the brownfield introduced within Bologna’s Structural Plan – mainly old manufacturing areas and freight

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

terminals – where not in any condition to be transformed to address the changing conditions already discussed (Fig. 3). Other vacancies arrived, and still improve, due to the current building private market’s dramatic asphyxia.

Figure 3. Ravone Area. Dismissed freight terminal. Survey of the existing vacancies and waiting lands Source: Studio Performa A+U copyright

As an immediate consequence of this increasing phenomenon, the local authority is today facing a building recession without precedents, where edge conditions multiply to trigger a unique possibility for the near future. But this high potential cannot be tackled within the existing procedure and protocol framework because the latter has been built according to a society which does not exist anymore, and the unstoppable diffusion of vacancies constitutes its doubtful continuation. To exit this paradoxical situation it is necessary to suspend the Plan – or its principles, its set of rules and its behaviours – at least within the brand new Vacant City. This would imply leaving a space and making room for a season of experimentation within a building heritage which often presents a surprising intrinsic quality (Fig. 4 and Fig. 5). But this is not going to happen. To confirm suspicious interpretations, the matter of vacancy has been addressed by the Department of Culture and its Alderman, depriving the Department of Urbanism its institutional authority and its capacity to adopt it as a new vehicle of civilisation. The aim is even more evident: to soften the public discussion on it and weaken it. Making the issue of vacancy a mere cultural aspect of the existing reality in fact equals not recognising that it is still hidden evidence of a new reality which is being recognised at an institutional level for its intrinsic political, social, economic and cultural potential. This inevitably leads to an additional aspect: by implicitly supporting the policy of not acting in the direction of the Vacant City, the local municipality is explicitly fostering a new tabula rasa attitude, which is consolidating the Denied City phenomenon. This favours those who are interested in developing greenfield areas, because they have been heavily granted by the local credit system. Business and bank companies, coming into crisis too, are claiming back their financial resource loan and not observing any investment revenue from them. It is a surrealistic chess play between a non-existent leader, the local authority, and a non-existent troop, the local traditional developers and their supporting forces.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 4. (L) Ravone Area. Vacant building available for temporary uses Source: Studio Performa A+U copyright Figure 5. (R) Ravone Area. Dismissed freight terminal. Vacant building available for temporary uses Source: Studio Performa A+U copyright

Conclusion The Denied City is the ultimate result of the edge condition when the latter is not strategically recognised as the most powerful chance any society has in order to subvert a previously existing and well established civic order. When it is no longer capable of driving change, room is left for the other. This statement confirms that the urban form is not simply a representation of a civilisation process, but its preliminary and necessary condition, i.e. its legitimising infrastructure. The principles, set of rules and rituals, upon which its inception, realisation and use are based, are instrumental to the development of its political, social, economic and cultural contents. Referring to the original Roman terms, which still convey their timeless significance, the Urbs, i.e. the physical structure of the city, brings the Civitas, i.e. its civic institutions, to come into existence, and not vice versa. The building type expresses the collective project â&#x20AC;&#x201C; or the system of values and expectations shared by all the forces claiming a role within a process of civilisation â&#x20AC;&#x201C; embodied by any built aspect of society through the different plan tool degree application. To define this project a high level of stability is required. This is why the contemporary lacks it and it fails to achieve its basic requirement condition, i.e. the self-recognition of its drivers of change. However, stability always nurtures itself through that informality and spontaneity to express a state of critical awareness, which solely flourishes through the unexpected. Consequently, to focus on the survival of the edge condition, where the new hides itself, brings contemporary focus on the Denied City (so called because it is not accessible to anyone), where the struggle between the old world, which is desperately aiming at its survival, and the new one, which is claiming its coming into existence, urgently flourishes. Ultimately, within the dialectic between established and informal, codified and latent, critical and unconscious, urban form always performs its political role, and prompts the crucial function of Urban Morphology and Building Typology, i.e. the study of it and its project, to be recognised as a fundamental interdisciplinary field to understand the change.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

References Aureli, P.V. (2008) The project of Autonomy. Politics and architecture within and against Capitalism (Princeton Architectural Press, New York). Bauman, Z. (2000) Liquid Modernity (Cambridge Polity Press, Cambridge). Cacciari, M. (1976) Krisis. Saggio sulla crisi del pensiero negative da Nietzsche a Wittgenstein (Feltrinelli, Milano). Castells, M. (1996) The Rise of the network society (Blackwell Publishing Ltd, Oxford). Cervellati, P.L Scannavini, R. De Angelis, C. (1977) La nuova cultura delle città ( Edizioni Scientifiche e Tecniche Mondadori, Milano). Clément, G. (2003) Le Tiers-Paysage (Editions Sujet/Objet, Paris). Conzen, M.R.G. (1960) Alnwick, Northumberland: a study in town-plan analysis (George Philip, London). Hartzema, H. (2012) Klein and Fijn Rotterdam. Stedenbow per kavel (Gemeente Rotterdam Stadsontwikkeling/Rotterdam). Koolhaas, R. Boeri, S. Sanford, K. Tazi, N. Obrist, H. U. (2000) Mutations (Actar, Barcelona). Mauss, M. (1938) “Une catégorie de l'esprit humain: la notion de personne celle de moi”, Journal of the Royal Anthropological Institute, vol. LXVIII, 263-281. Merotra, R. (2011) Architecture in India since 1990 (Pictor, Mumbai). Muratori, S. (1959) Studi per una operante storia urbana di Venezia (Istituto Poligrafico dello Stato. Libreria dello Stato, Roma). Pirandello, L. (1921) Six Characters in Search of an Author and Other Plays (Penguin Group, USA). Vitruvius Pollio, M. (1999 ed.) Ten Books on Architecture (Cambridge University Press, Cambridge).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

ON THE EDGE OF THE CITY peripheral areas and urban form in suburbia

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Looking for the mall: public life in the city of dispersal Luisa Bravo School of Design, Queensland University of Technology, Brisbane, Australia Email: luisa.bravo@qut.edu.au

Abstract. When people are asked to describe the city where they live, especially in Europe, they will probably answer referring to the old centre or to the downtown area. Often they are able to describe the network of public spaces and to give information on how to reach on foot shops or public facilities, searching buildings, churches, squares, paths and facades in their own mental map, flipping through the urban-ness of the old city (Sieverts, 2003). However if they are asked where they spend their leisure time on a typical day, most of them will probably answer referring to shopping malls in the suburbs, which are related to major roads and infrastructures. In the Italian fragmented landscape of outskirts, urban form is marked by big box retail stores as masses and automobile connections as voids. During past decades, malls have progressively changed to best accommodate consumers’ needs: they strongly enlarged their structure introducing a variety of services and entertainments, such as movies and restaurants, fast-food arcades, libraries and other recreational activities, together with office developments, thus becoming attractive community centres. This paper will investigate how Italian suburbs have deeply changed due to large luxury shopping malls, in terms of urban typology and patterns but mainly in terms of mobility and public life. Keywords: dispersal city, public life, shopping malls, suburban typology

In Europe, public spaces, intended as a network of structured sequences in a dense and compact urban scenario, are mostly linked to timeless and beautiful historic cities. Monumental buildings, churches, market places, squares, pedestrian promenades and facades appear to be the main characteristics of the built environment, intended as urban elements that are open to members of “the public”, in a topographical approach (Iveson, 2012). In opposition to a deep knowledge and an endless interest in the myth of the old, compact, mixed-use city, suburbs, namely that very large area outside the historical perimeter, have been considered for a long time by scholars as a negative consequence of the application of the modern principles expressed by Congrès International d’Architecture Moderne (CIAM) (Hall, 1974; Fishman, 1990; Duany, 2000) referring to the four functions of 20th century life, such as dwelling, work, recreation and circulation (Charte d’Athènes, 1933). Suburban design served the basic need to accommodate new residents with little consideration of the implications of creating an extension of the existing core or a new, independent urban centre. Mostly inhabited by bedroom communities, such as large housing estates and suburban satellite housing, dependent on the old city for employment, shopping and entertainment (Fishman, 1987), suburbs were marked with a fragmented and dispersed urban design, often empty of morphologically defined public spaces. The American critic and urbanist Lewis Mumford believed CIAM’s four functions missed the city’s most necessary ‘fifth function’: space for civic life, like the agora, forum or piazza (Mumford, 2000; Costanzo, 2016). These spaces were mostly conceived as a void between buildings filled with plenty of green space. Especially in Europe and Italy, city managers and politicians were accused by scholars of rejecting the successful model of the old city: suburbs, in their morphological settings, based on the separation of functions on the basis of zoning were completely detached from the historical language.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

This paper will discuss how the debate about suburbs moved from the negative scholarly judgement started in the 1990s to the current understanding of the potential of a new, complex, even rich city. Earlier critiques addressed the suburbs as uninteresting and lacking of personality (Barnett, 1996) because of their monotonous, functional districts, or even ugly places due to the poor quality of architecture. More contemporary critiques address the suburbs as a site for different forms of social engagement that is starting to take advantage of the limits of suburban design With references to academic debate in USA and Europe, this paper will focus on Italian suburbs to outline the genesis of shopping malls as an opportunity for public life, and redefine concepts and analysis to be used for the interpretation of the morphological environment. The research is mostly based on empirical analysis, observations and practical experience, as a starting point of more structured research, still to be developed. The main aim is to define a different approach in suburban studies, based on context-based analysis, overcoming preconceptions and stereotypes that are still affecting the general discourse on the built suburban world, even questioning the name ‘suburb’, which implies dependence on the old city, and the concept of ‘public space’ in its morphological sense. The historical background of the Italian situation is the result of a cross-disciplinary research, using different sources, mainly from architecture theory and sociological and anthropological studies, with references also to local newspapers, interviews, archival documents and websites.

The suburbanisation process: a new city shaped by morphological fragmentation It is evident that suburbs are different in scale, proportion and density when compared to consolidated environments, with building types and urban patterns completely different from historic morphologies. Suburbs developed over time when, at the end of 19th century, early industrial activities settled far away from the city centre, as new functions and work places, and were supported by railway infrastructures for long-distance connections as a result of modern needs required by a modern society. Outside the historical perimeter, outside the main core where all functions and consumptions were located, a large amount of cheap undeveloped land was available for specialised purposes and activities. That was the starting point of an epochal urban transformation in terms of mobility and public life. The city expanded after World War II in order to accommodate a large number of new residents attracted by the possibility of employment. It started with minor settlements in the rural landscape through to large housing complexes: in Italy, over a period of 14 years (1949-1967), the INA Casa program (National Social Housing Program) delivered more than 300,000 new dwellings of good quality in terms of design and architecture, due to the involvement of talented and well-known architects. In 1930 the Italian Institute of Urban Planners was founded with the aim of promoting studies and to disseminate the principles of planning (INU, n.d.). From the 1960s, the urban growth rate increased, due to favourable economic conditions, and cities suddenly enlarged their form, occupying new territories with sprawling environments. For three decades this growth continued uncontrolled, without any homogenous action in planning, development, typological or pattern design for residential communities, infrastructures and public facilities: planners just implemented laws and codes to ensure a general standard requirement, with a little attention to urban design and quality in neighbourhoods and communities. A sense of exclusion, in both geographical and social senses, from main urban activities and public functions started to configure peripheries as a place of poverty to be escaped from as soon as possible (Ciorra, 2010). A large debate during the 1990s promoted a general reflection on city and suburbs, all over Italy and Europe. One criticism, based on morphology and figure-ground diagram studies, intended to demonstrate the lack of density and the absence of a coherent urban fabric, non-conventional forms of public spaces, the prevalent use of automobile due to a larger scale, a difficult walkability through dispersal functions qualified suburbs as non-places, compared to the beautiful, dense structure of the old city. At the beginning of the new millennium, this debate, intended to deny any existing value in the vast land of suburbia, has started to demonstrate its weakness in front of the demographic evidence that for the

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

first time in history half of the people worldwide are living in town or cities (UN Habitat, 2007), and mainly in suburban settings. According to CRESME (Italian Economic and Social Research Centre for the Building Environment and Territory), addressing ‘suburbs’ as the environment built after 1946, in 2001, 15% of the Italian population were living in the centre, 10% were living around the centre, 75% were living in suburbs, meaning that 13 million inhabitants were living in the centre and 43 million were living in the suburb. The 2011 census, promoted by ISTAT (Italian National Institute of Statistics), revealed that during the past ten years the Italian population has increased by four million inhabitants, due to increasing immigration, mainly located in suburban areas (Bellicini and Ingersoll, 2001). While the old city still stands as a consolidated, functional and geographic centrality, many kinds of varied and complicated suburbs developed in different historical periods. This resulted in many different centres, mainly related to functions, coexisting in the same city of dispersal as a superimposition of different layers over one century of growth.

Moving through the suburban environment Currently cities host two different spatial and urban conditions, two different ways of inhabiting and living, as a series of contradictions: compact versus modern, historic versus suburban, too full versus too empty, figure versus ground, over-defined versus undefined space, pedestrian use versus private and public transportation (Bravo, 2012). Suburbs are a new kind of city, with major highways intersections and powerful railway infrastructures, where connections between different parts and functions are no longer provided by pedestrian paths. The use of automobile, related to ‘circulation’, one of the main functions according to CIAM principles, has completely changed the balance between time and space at the urban scale. Inhabitants of sprawled environments turned themselves into drivers, apparently without claiming for the absence of pedestrian streets. They are accustomed to moving on large corridors through high speed devices, experiencing their world through a jump-cut urbanism (Ingersoll, 2005), changing the dominant mode of form perception from a single vanishing point to multiple perspectives, seeing forward and backward simultaneously, through windshields while driving like fragments in an accelerated montage at various speeds. This cinematic metaphor is able to explain a relevant change related to suburban space system: while the old city buildings involve constituent parts of city blocks defining spaces, in the modern urban vision buildings are objects-in-space, like freestanding elements in landscape settings (Carmona et al, 2010). So the perception of the morphological dimension is not similar to a fixed proscenium, but it is made of discontinuous images related to large-scale road networks surrounding superblocks. People move from one function/building to another for specific activities in a kind of space left over after planning (Huet, 1998; Hebbert, 2008), where cars are the main partners in a broad, satisfactory consumption. The spread of suburban malls has progressively accustomed large numbers of people to behaviour patterns: they drive for miles through dusty landscapes of houses, factory sheds, road intersections, old farmhouses and car washes. Following the signs they come to the car park, where the short walk to the mall itself is the only direct contact they have with external reality during the whole journey. Leaving behind the world of weather and time, they enter a large, climatically protected, artificial space and move among individuals who, in slightly different way, perform similar actions. When the subjective time of the visit has elapsed, they go through the same sequence of events in reverse: exit with bar-code shoplifting protection, full reality immersion in the car park, landscapes sliding past on the other side of the windscreen, front doorstep (Boeri, 1998).

Looking for places where public life can happen In the Italian landscape of suburbia, shopping malls stand as landmarks both in the morphological and social dimension. The former is expressed by a specific suburban typology, namely a big box windowless structure, usually made of cheap concrete blocks, typically sited next to an arterial road or near a freeway

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

interchange with high traffic volume (Hayden, 2004). Approached from a distance, big boxes are standing alone on a uniform asphalt terrain in an otherwise undisturbed field. This kind of typology derives from the American model designed by Victor Gruen for Southdale Center in Minneapolis, USA, which opened in 1956 and is considered the first indoor, climate-controlled, multi-store under-one-roof shopping centre in the United States. The latter is related to functions taking places inside them: chain shops, discount buyer clubs and department stores, together with cafés and restaurants, allow people to spend several hours in the same place, because many possibilities are available at the same time and place, and easily accessible by automobiles with large parking lots. During past decades, the urban typology of malls has progressively changed to best accommodate consumers’ needs: they strongly enlarged their structure introducing a variety of services and entertainments, such as movies and restaurants, fast-food arcades, libraries and other recreational activities, together with office developments, thus becoming multi-purpose town centres. At the same time shopping has shaped and refashioned different functions, such as airports, train stations, museums, schools with a dramatic voracity, so that places started to look undistinguishable. Expanded and enriched with the internet experience, shopping is spreading into virtually all areas of our lives. According to Gruen urbanism, the shopping mall was a vehicle towards the ambition to redefine the contemporary city. For Gruen, the mall, as developed in the very first shopping mall, Northland Center, which opened in Detroit in 1954, was a utopian communal space intended to bring people together (Gruen and Smith, 1960; Kaliski, 2008), it was the new city. In opposition to a largely abused concept that consider shopping malls as non-lieux (Augè, 1992), it is now quite common to consider them as the best form of contemporary public life, able to work as super-places (Agnoletto, Delpiano and Guerzoni, 2007), namely as new successful urban occasions at the highest level of attendance in the metropolitan territory. Research developed in Italy at the University of Bergamo (Lazzari and Jacono Quarantino, 2010) revealed that about 35% of young digital natives in the Bergamo area choose shopping malls as relevant landmarks in the suburban morphology: they represent for them places of social interaction, more than oratories, sport games and recreational centres, and many even find their first jobs in malls. Shopping malls, as privately owned spaces, are a new kind of attractive community centre, able to enhance different forms of social interaction. Even if not performing as the democratic public realm because the management have the liberty to exclude whoever they feel unwanted, these quasi-public “third places” signify the great variety of spaces that host the regular, voluntary, informal and happily anticipated gatherings of individuals beyond the realms of home and work (Oldenburg, 1999; Banerjee, 2001).

Shopping malls in Italy The first shopping mall in Italy was built in Bologna in 1971, the “Fossolo 1”, as the name of the suburban neighbourhood where it was placed. With a large supermarket and several shops at the ground floor of a residential high-rise building, it was fully integrated in the urban tissue and was intended to give services to a new development area, establishing a formula that was successful for several years in Europe (Paolucci, 2006). In 1993, 22 years later, a big box freestanding building in Bologna, situated in an industrial area, few kilometres away from the city centre, became the largest shopping mall in Europe, “Euromercato”, presently known as “Shopville Gran Reno”. Designed by a renowned Bolognese architect, Enzo Zacchiroli, this two-storey building was conceived with a retail area 45,000m2 wide and a parking lot able to host 2,400 vehicles. Starting from the 1980s and 1990s, larger and larger big box buildings flourished throughout the Italian territory, mainly to allocate commercial and primary services as an integration to bedroom communities sprawled during the economic growth of the 1960s and 1970s. Some companies, in order to mitigate the impact of a standardised typology, involved famous architects to work on the corporate identity. The better known developments include Aldo Rossi’s design of the “Cento Torri” mall in Parma (1985-88) and the “Terranova” complex in Olbia (1996-1997), Adolfo Natalini’s design of “I gigli” in Campi Bisenzio,

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Florence (1993-97), Ignazio Gardella’s design of the “Esselunga” superstore in Saronno, Varese (1989) and subsequently (1990-95) in other different locations in the Milan metropolitan area. These modern designed shopping malls located in the suburban setting tried to mimic successful models established in the city centre, such as “Le Bon Marchè” and “Galerie La Fayette” in Paris, or “Harrods” in London, as multi-storey commercial buildings. Also, “Galleria Vittorio Emanuele II” in Milan and “Galleria Umberto I” in Naples as luxury passages through the urban morphology. The attention was also related to modern mall buildings designed by acclaimed Italian architects during previous years, mainly in consolidated contexts, such as “La Rinascente” in Rome (designed by Franco Albini, 1957), as symbols of modern society progress and economic growth. More recently Italian architects have been involved in scenographic design of gigantic malls, as landscape icons: Massimiliano Fuksas for “Etnapolis” near Catania (2006) with over 100,000m2 of retail area and a parking area for 6,000 vehicles, and Renzo Piano for “Vulcano buono” in Nola near Naples (2007), with about 150,000 m2 of retail area and a parking area for 8,000 vehicles. In 2008 a huge shopping mall “Campania” has been erected in Marcianise, near Naples, by Corio, one of the largest listed retail property investors and managers in Europe, with shopping centres in The Netherlands, France, Italy, Spain, Germany and Turkey. “Campania” is considered as the largest mall in Italy and consists of four integrated covered parts with a lettable gross area of more than 100,000m2 and 7,000 parking spaces. The catchment area consists of 1.2 million people living within 30 minutes driving distance (excluding Naples). The shopping mall reaches 11,000,000 visitors annually. “Creating a destination” is the key buzz-phrase for the superstore builders: these are places not only meant to shop but also to visit, places to which tourists make ritualistic pilgrimages (Klein, 2000). They symbolise the illusion of a city scaled to human beings that could restore to citizens in virtual terms the social relations and sense of belonging that were denied them in the real city outside (Irace, 1998). They serve as spatial centrality, public focus and human density – all the elements lacking in sprawling environments (Crawford, 1992).

From the big box mall to the townscape mall: an Italian classic The McArthur Glen Group, an English owner, developer and manager of stylish outlets, opened his first Italian outlet village in Serravalle Scrivia, near Alessandria, in 2000 (at that time the largest outlet in Europe), then in Castel Romano, near Rome, in 2003, in Barberino del Mugello, near Florence, in 2006, in Noventa di Piave, near Venice, in 2008, and finally in Caserta, near Naples, in 2010. The philosophy behind the commercial activities established by McArthur Glenn is to create a sense of place: the shopping mall is no longer an introverted place to live inside a box but it is a not-covered street in a new large urban complex, with traditional buildings and public spaces, so that the shopping experience is mixed with a pleasant walk in a place that recalls urban features of a small village. This kind of mall was conceived as the “townscape mall”: instead of a single architectural style there are multiple styles so that the strip appears to be broken down into structures built at different times with varied character, creating stronger pedestrian scale and identity (Southworth, 2005). The street structure provides each shop with direct access and individual identity. The parking area is outside, all around the commercial settlement, which is accessible from different gates. The main focus of this kind of mall is a central private-public square. Serving as a social space, it is often animated with programmed activity such as art and craft shows, farmers’ markets, car and talent shows, with also movie theatres, restaurants, interactive game centres, play areas for children, spas and sports facilities, to keep people entertained while having a shopping downtown-feel walk, thus recreating a kind of urbanity and emotional engagement to the place. Various symbolic elements of the main street are integrated into the mall and there is more attention to pedestrian amenities. Townscape mall have the ambition to become places of urban renaissance through faithful copies of a small traditional city, with columns, porticoes, gables, stone sculptures and other symbolic references to monumental buildings at a smaller scale.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Even if the urban stage is more complex and intriguing, compared to that offered by big box malls, the mobility behaviour, related to function, is always influenced by the use of the automobile. To perform like a city centre, townscape malls should be embedded in a real existing urban environment, with a potential to growth, instead of being set in the midst of a suburban development, close to rural land or productive districts. They seem to stand as the very first attempt of urbanisation, but actually they are bargainoriented urban complexes. Buildings, no more than two or three storeys tall, host chain shops at the ground floor and sometimes offices or warehouses at upper levels, so that windows on the main façade are just used to create a theatrical performance of the space because they are lighting empty rooms, with no activity inside. Designers claim they worked on the organic inclusion in the suburban context and on the figurative revival of the genius loci, which is actually the required premise of every urban and territorial intervention in the Italian context. That is why in Castel Romano the architecture recalls the Aurelian walls in Rome and in Barberino del Mugello the urban morphology of a medieval village in the Tuscan countryside. They are proposing a reassuring model, related to tradition, that people can feel as part of their own identity. However, they are offering a fake copy of a city centre, in terms of scale, proportion of space and architectural language, that is totally decontextualised. Nevertheless, this streetscape strategy has been applied by another European outlet company, Neinver, based in Italy, Poland, Portugal, Germany and France. In Italy two “Style Outlets” have been opened: Vicolungo outlet in 2004, located near Novara, and Castel Guelfo outlet in 2007, near Bologna. The appreciation of common people and the large number of daily visitors have acclaimed shopping malls as offering urban regeneration: public administrators, who are in charge of approving these new commercial settlements, believe that territories can find enrichment and that communities can share new values through these large complexes, as new centres of civic attraction. By not being integrated in a existing urbanity, they stand as punctual events on a map in the dispersed land of suburbia, always located ‘near’ something that can be considered as a real place or a recognisable urban entity. Thus mobility is always affected by that kind of jump-cut experience with no continuity in the narrative plot of the urban realm.

Shopping as key element for urban regeneration In Italy, shopping is usually considered by city managers and public administrators the best activity to regenerate abandoned industrial sites, close to or part of central urban areas, to give services in newly designed neighbourhoods (like in the Meridiana complex – a gated community – at Casalecchio di Reno, in the province of Bologna) or to create new job opportunities: using a fashionable design and a mix of trendy shops, entertainment and food activities, providing also a large parking area, these places are a daily opportunity of services for residents and users, going to the cinema or to the gym, and a week-end destination for dinner. In opposition with suburban outlet villages, these urban shopping malls can perform as an effective opportunity for public life throughout everyday activities, modifying or influencing established rhythms and human behaviours, generating new neighbourhood centres. Nevertheless, they are not always successful: the ability to attract public life is the result of several factors, depending on the quality of the space and its architecture and the variety of shops and events, but also dependent on the urban dynamics of the whole city that influence accessibility including proximity to other relevant urban features. In Bologna, at the beginning of 2000, an industrial building hosted the Officine Minganti since the beginning of the 20th century. It was a famous machine tool factory that was transformed into a shopping mall, which opened in 2006, with a supermarket, a gym, several luxury shops and neighbourhood services. The complex received the ‘Plaza Retail Future Project’ prize at MAPIC 2005 (an international shopping centre developers and owners event taking place in Italy), in the category of ‘urban regeneration’. But after only a few years of activities, Officine Minganti was facing serious decay and failed to accomplish the enthusiastic premises that inspired its transformation and attracted huge investments. Today only a few activities are still operating and the complex has become part of a city debate on how to establish an effective strategic vision related to the refurbishment of industrial buildings.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

A recent book, as a research project developed at Politecnico di Milano, entiled Demalling (Cavoto, 2014 – which is also a website demalling.com), explores the unusual phenomenon of the failure and demise of big box stores, intended as enclosed spaces, in the Italian context with several case studies. This process started about two decades ago in the USA: named in 1998 as one of the top 50 most revolutionary consumer innovations, shopping malls flourished during the 1950s but now, particularly in the American Midwest where economic decline has sped up the ‘going out of business’ process, even once-vibrant shopping malls are decaying after being abandoned. Recently, several major journals (like The Guardian in 2014) published surreal pictures of dead American shopping malls, leaving ghostly buildings in their wake. The process of retrofitting suburbia, namely a process of entirely revamping, and in many cases completely replacing, the conventional zoning that has dominated land use decision making and development for decades (Dunham-Jones and Williamson, 2009), is currently working on dead or dying malls to establish a new model of adaptive reuse, taking into account a morphological analysis, of big boxes and strip malls for community-service activities.

It’s a mall world Worldwide competition is underway for the largest shopping mall: according to the Guinness Book of Records the largest shopping mall is the Dubai Mall, located within Downtown Dubai, United Arab Emirates. Opened in 2008, it has an internal floor area of about 550,000m2 with 1,200 retail outlets and over 160 food and beverage outlets. It includes also an aquarium, Olympic size ice rink, and a 22 screen cineplex. In 2013 the New Century Global Centre opened in Chengdu, south west of China, with a gigantic structure, as a single body, 500m long, 400m width, 100m tall, and a floor area of about 1,700,000m2, able to accommodate three times the Pentagon. It is considered the biggest building in the world: developed by billionaire Deng Hong's Entertainment and Travel Group (ETG) it hosts various commercial opportunities and an aquatic park with an artificial beach, illuminated the whole day by a simulated sun. Some of the world’s largest shopping malls are in Asia, according to Forbes, and the United States also has some of the biggest malls in the world, such as Macy’s in Herald Square, New York. In Australia many shopping malls are branded by the Westfield Group: in Sydney the “shop till you drop” experience is upscale and luxury. Brisbane city centre contains a pleasant, well designed pedestrian townscape mall, the “Queen Street mall”, as a result of a project intended to revitalise the centre and attract people. In Moscow, the large building facing Red Square, which features a combination of elements of Russian medieval architecture and a steel framework and glass roof in a similar style to the great 19th-century railway stations of London, hosts the famous Gum, from the name of the main department store in many cities of the former Soviet Union. Hong Kong is probably the most impressive shopping destination in which shopping is a fulfilling experience: from the “shopping and dining” formula in “iSquare”, a 31-storey high shopping mall on Nathan Road, designed by the architecture firm Rocco Design Ltd, to “The Landmark”, the home of numerous prestigious international brands, on the Hong Kong Island. In comparison, the malls in Europe might seem much smaller, but not necessarily tiny. At the time of publication, the largest malls in Europe are located in England (Westfield Stratford City near London, opened in 2011, contains over 330 shops, 50 restaurants and 11 screen cinemas, located at the gateway to London’s 2012 Olympic Park), Turkey (Cevahir in Istanbul, opened in 2005, contains, shops and a parking area capable of holding 2,500 vehicles, 48 restaurants, including many fast food options but also some high-class, exclusive dining spots, as well as a puppet theatre, a mini-roller coaster, a haunted ride, a carrousel, a laser arena and a 4-D theatre), and Portugal (Colombo in Lisbon, opened in 1997, is the

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

largest mall in the Iberian Peninsula, houses 420 stores and more than 60 restaurants, as well as plenty of entertainment options, such as an indoor amusement park with roller coaster, and 10-screen movie theatre, health club and bowling alley). It is difficult to work on a classification on shopping malls though, based on square metre surface, because developers regularly open building sites for the construction of a new and larger shopping mall elsewhere in the suburb as well as in central districts. The architecture of consumption, and in particular the architecture of shopping, follows the logic of the Next Big Thing, namely a recurring promise of a new typology that will deliver greater profits and higher consumer satisfaction (Herman, 2001). The principle of newness is the engine of the market system, so that enduring shopping buildings, department stores, big boxes and malls need to be constantly updated through renovation and expansion while reflecting increasingly refined consumer profiles. Now malls are conceived to accommodate not only an impressive number of shops and restaurants but also plenty of entertainment options to hold people on site long after shopping.

Historical settings: looking for the mall? While shopping malls - accommodated in a safe, clean and controlled typology, more or less complex – served for many years as the hub of public life in the dispersal city of suburbia, historic centers, all over Europe and Italy, throughout the 1960s and the 1970s started to change their accessibility: the drive for pedestrianisation banned automobiles from significant portions of central areas with the intention to enhance public life and allowed a safe and more pleasant browsing pace for shoppers. Public administrators realised that cars in the city centre were like cars inside a painting of the Medieval or Renaissance period, thus restricting urban potential and general consumption. This action allowed commercial activities to flourish. In Italy many city centres, especially in Rome, Florence and Venice, as final destinations of worldwide tourism, progressively no longer supported everyday residents’ activities, but en-plein-air shopping malls where monuments are standing as historical anchors. Milan, as Italy’s fashion capital, is the country’s biggest shopping centre. Many outlets are located in Milan and the surrounding area where Italian labels and famous brands discount their goods by as much as 50%. This process generated a progressive loss of urbanity, seriously impacting on residents’ activities and daily uses. The social significance of public space as a meeting place for people started a still ongoing transition into a place of commerce and consumption, where locals and tourists mixed while replacing neighbourhood services with attractive and luxury chain shops. As a result, many historic residents have moved to suburban neighbourhoods or far away in the metropolitan area, looking for better housing conditions including facilities and daily costs, freely accessible parking areas and green amenities. At the same time, paradoxically, remaining historic residents accustomed to the move to suburban malls, where different choices of public life, vibrant hypermarket services and flagship stores are available. While urban retailers and governments, following a commercial strategy, started to convert historic streets into shopping centres, acquiring characteristics of suburban malls, suburbs have begun to resemble cities with townscape malls. In our contemporary society, shopping has become the basic glue of every urban experience in search of a sense of community.

Further development This paper aims to establish a theoretical background of the Italian situation related to the fragmentation of the suburban territory and the importance of shopping malls as places for services and public life. It is intended to be a general overview of broader research based on the investigation of suburban everyday life in order to unveil a complex urbanity, based on ever-changing and unpredictable factors. The research assumption is that suburban public life can be discovered in different functions, places and practices, not necessarily in a morphologically designed space, facilitated by accessibility provided by

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

mobility – mainly cars – and urban and territorial infrastructures. If it is true that public life in the old city is intertwined with the design and morphology of public spaces, it is not true to say that there is no public life in the suburb because of the lack of designed – and well-maintained – public spaces (Segal and Verbakel, 2008). So we should start working on amorphous so-called suburban ‘publics’ trying to define and map their spaces, exploring human itineraries and marking facilities able to host them (Bravo and Crawford, 2014). As this paper has started to explore, shopping malls are a worthy reference to develop the research assumption because they offer multiple opportunities of investigations.

References Agnoletto, M., Delpiano, A., Guerzoni, N. (2007) La civiltà dei superluoghi. Notizie dalla metropoli quotidiana (Damiani, Bologna) Augè, M. (1992) Non-lieux (Seuil, Paris) Banerjee, T. (2001) “The future of public space. Beyond invented streets and reinvented places”, APA Journal, 67(1), 924 Barnett J. (1996), The Fractured Metropolis: Improving The New City, Restoring The Old City, Reshaping The Region, Westview Press, Boulder Boeri, S. (1998) “Il centro commerciale nel territorio periurbano”, Abitare, 373, 164-167 Bellicini, L and Ingersoll, I (2001) Periferia italiana (Meltemi, Roma) Bravo, L. (2012) Public spaces and urban beauty. The pursuit of happiness in the contemporary European city, in Pinto da Silva, M. (ed.), EURAU12 Porto | Espaço Público e Cidade Contemporânea: Actas do 6° European symposium on Research in Architecture and Urban design, Porto, Faculdade de Arquitectura da Universidade do Porto (FAUP), 12-15 September Bravo, L. and Crawford, M. (2014) Publics and their spaces: renewing urbanity in city and suburb, in Cavallo, R. et al. (eds), New urban configurations, EAEE/ISUF International conference proceedings, pp. 784-789 (IOS Press & TU Delft, The Netherlands) Costanzo, D. (2016), What architecture means. Connecting ideas and design (Routledge, New York) Crawford, M. (1992), “The world in a shopping mall”, in Sorkin, M. (ed.), Variations on a theme park. The new American city and the end of public space, Hill and Wang, New York Carmona, M., Tiesdell, S., Heath, T., Oc, T. (2010) Public places, urban spaces. The dimension of urban design (Routledge, London and New York) Cavoto, G. (2014), Demalling (Maggioli, Rimini) Ciorra, P. (2010) La fine delle periferie in XXI secolo, in Enciclopedia Treccani (Allemandi, Torino) Duany, A. (2000), Suburban Nation: The Rise of Sprawl and the Decline of the American Dream (North Point Press, New York) Dunham-Jones, E. and Williamson, J. (2009) Retrofitting Suburbia. Urban design solutions for redesigning suburbs (Wiley, Hoboken) Hall. P (1974) Urban and regional planning (Penguin, Harmondsworth) Hayden, D. (2004) A field guide to sprawl (W. W. Norton, New York) Hebbert, M (2008) “Re-enclosure of the urban picturesque. Green space transformation in post-modern urbanism”, Town Planning Review, 79(1), 31-59 Herman, D. (2001) The Next Big Thing, in Chung, C. J., Inaba, J., Koolhass, R., Leong S. T. (eds), Harvard Design School Guide to shopping, Taschen, Koln Fishman R. (1990) “Megalopolis unbound”, Wilson Quarterly, vol. 14, n. 1/1990, pp. 24-45 Fishman, R. (1987) Bourgeois Utopias. The rise and fall of suburbia (Basic Books, New York) Huet, B. (1998) Il sistema e il modello, in P. Di Biagi (ed), La Carta d’Atene. Manifesto e frammento dell’urbanistica moderna (Officina, Roma) Gruen, V. and Smith, L. (1960) Shopping Towns USA. The Planning of Shopping Centers (Reinhold Publishing Corporation, New York). Full version available at: https://archive.org/details/shoppingtownsusa00grue Ingersoll, R. (2005) “Jump-cut Urbanism. L’estetica dell’ambiente motorizzato”, Parametro 256, 34-39 Ingersoll, R. (2006) Sprawltown. Looking for the city on its edges (Princeton Architectural Press, New Yok) Irace, F. (1998) “Il centro commerciale”, Abitare, 373, 162-164 Istituto Nazionale di Urbanistica (INU). Website: http://www.inu.it/italian-national-institute-of-planning-istitutonazionale-di-urbanistica-inu/ Iveson, K. (2012) Publics and the city (Blackwell publishing, Oxford) Kaliski, J. (2008) “The present city and the practice of city design”, in Chase, J. L., Crawford M., Kaliski J. (eds.), Everyday urbanism (The Monacelli Press, New York) Klein, N. (2000) No logo. Taking Aim at the Brand Bullies (Random House of Canada Limited, Toronto)

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Lazzari, M. and Jacono Quarantino, M. (2010) Adolescenti tra piazze reali e piazze virtuali (Bergamo University Press) Mumford, E. (2000) The CIAM Discourse on Urbanism, 1928-1960 (The MIT Press, Cambridge) Oldenburg, R. (1999) The great good place: cafes, coffee shops, bookstores, bars, hair salons and other hangouts at the heart of a community (third edition) (Marlowe & Company, New York) Paolucci, G (2006) La seduzione dell’entertainment. Consumo e leisure nello shopping contemporaneo in Amendola, G. (ed) La città vetrina (Liguori, Napoli) Segal R. and Verbakel, E., (2008) (eds), Cities of Dispersal, AD (Architectural Design) issue, (Wiley and Sons, London) Sieverts, M. (2003) Cities without cities. An interpretation of the Zwischednstadt (Spon Press, Oxon – English translation of the German edition published in 2000 by Birkhauser) Southworth, M. (2005) “Reinventing Main Street: From Mall to Townscape Mall”, Journal of Urban Design, 10(2), 151– 170 UN Habitat (2007) State of the World’s Cities 2006/2007 (UN Habitat, Nairobi)

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

The impact of social housing on the urban form in the suburbs of Shanghai Dong Yijia Tongji University, China Email: dyj_217@hotmail.com

Li Zhenyu Tongji University, China Email: zhenyuli@tongji.edu.cn

Abstract: This study reviews the effects of political strategies on the planning and design of new social housing in the suburbs of Shanghai, China. In 2009 and 2010, the Shanghai municipal government approved plans for 31 suburban â&#x20AC;&#x2DC;Large Communitiesâ&#x20AC;&#x2122; with an aim of constructing over 1.2 million apartments. These were a response to central government decision to accelerate the construction of new social housing. In this government led urbanisation, the speed of construction and the efficiency of housing delivery was prioritised over place-making and environmental considerations. As a result, the typical neighbourhood block is characterised by generic concrete slab highrises that are repeated in rows and have few unique features to differentiate each community. In determining the relationships between the policies and the resultant urban form, 38 suburban social housing cases were analysed. Comparison is made between the large communities in Shanghai and the early practice of the functionalists in Europe. The analysis found that overall control by the political level on urban-making is leading to conflicts in new urban living and communities. The study concludes by discussing alternatives for improving the planning and design of the new social housing suburban communities. Keywords: social housing, political strategy, urban form, suburbs of Shanghai

Background: A big target Like many other major Chinese cities, Shanghai is challenged by the conflict between high demands for housing and rapidly rising housing prices. In 2007, the central government of China issued a policy to reestablish the social housing system and committed to the construction of new social housing. The construction of new social housing in China had ceased for 10 years from 1997 to 2007. However, the decision was made suddenly and the target was set unexpectedly high. The national annual new construction of social housing rose from 5.8 million apartments in 2010 to10 million in 2011 and 7 million in 2012. This task has been delegated to every major city by the central government. In response, the municipal government of Shanghai decided to construct over 950,000 social housing apartments in the five years between 2011 and 2015. In 2009 and 2010, the city approved plans to build 31 large communities in the suburbs of Shanghai (Fig. 1). These communities cover an area more than 105km2 and are aiming to construct over 1.2 million apartments that will accommodate approximately three million people. Based on the concept of mixed income communities, two thirds of the construction area of each community is planned for social housing. At an average distance of 30km from the city centre, these communities are planned at town scale. The average size of the community is over 450 hectares accommodating a population more than 95,000 people (Fig. 2). Neighbourhoods of social housing are proposed by the government as stimulating the development of larger communities. With an average floor area ratio (FAR) of 2.0, which is two times that of a normal private suburban housing compound, the government expects that the new population will provide sufficient urban vibrancy for the future development of the communities.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 1. The layout of the 31 large communities in Shanghai Source: Yijia Dong

Figure 2. The size of the 'Large Communities' in Shanghai Source: Yijia Dong

In 2012, the Authority of Social Housing in Shanghai held an open competition for â&#x20AC;&#x153;My favorite social housingâ&#x20AC;? to select good samples of social housing projects. For this study, 38 cases from the winner list were chosen and all of these were designed or built between 2008 and 2011 in the suburbs of Shanghai. The case studies addressed three fields to determine the relationship between the policies and the urban form: the analysis of the political strategies applied in the cases, the character of the new urban form, and the interviews with the residents.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Strategies and policies oriented to the target Strategy 1: Whole deal in organisation From the observation of the scale and management of the cases, some common characters in the organisation of new social housing projects were found. These included: a) b)

Super block. The average size of the block is over 6.4 hectares. The dimension of the blocks ranges from 250m to 500m. Huge size of project. The average size of a project exceeds 17 hectares, which is almost three times that of a normal private housing project. More than half of the cases occupy more than two blocks. ‘Big developer for big site’. As a slogan from the municipal government, this illustrates the reliance on a sole state-owned developer with design by one architecture firm which was assigned by the government. The purpose of this is to ensure efficiency through a single decision (Wang, 2011).

The current organisation of social housing projects shows a high level of centralisation of authorities, which is referred to as the ‘whole deal’ strategy. This approach minimizes the chance of competition or disputation, so that the projects can be accomplished with minimal hesitation or delay. Strategy 2: Simplified controlling frame From the observation on the controlling frame behind these cases, we found that it has a simplified and straight forward structure with the following characters: a)

Single function plots in the planning level. All the cases are single function plots. The corresponding community service facilities are often separated from the housing parcels and located in an isolated corner. Similar density but under same coverage and height control. Most of the cases are set at a similar FAR at 2.0, while they are given the same limitation on building coverage (<30%) and height control (60m). These parameters almost result in the same highrise model for all the projects. From the case study, we found that the most popular prototype for building was 18 levels and three to four units per floor. This is used by 65% of the cases. Focus on orientation and sunlight control. In the current building regulation for housing the main rooms for every apartment should be facing south, and the distance between buildings is determined by one-hour sunlight on winter solstice. These control methods ensure all the apartments offer the same standard of living condition. However, at the same time, it makes slab highrise the only solution for high density neighborhoods (Fig. 3).

The strategy for the controlling system aims for simplified standardisation to facilitate rapid construction and massive distribution of work. However, the planning does not address requirements for either mixed land use or the quality of public space. Specific considerations on open space are neglected to avoid exception and alternatives in design work. Strategy 3: Uniformed production In the study of the building design of the cases, it was found that ‘uniformity’ prevails. Most of the cases share similarity in the following aspects: a)

Few alternatives in building types. Only two types of social housing are built in the suburbs of Shanghai. One type is affordable housing for low income groups and the other is resettlement housing for residents removed from areas redeveloped by the city. Of the 38

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

cases examined, 35 feature only one kind of housing. However, all the blocks in the cases provide one type of housing. Few alternatives in units. The guideline for social housing is limited to three kinds of unit setting for both housing types. As a result, a single case usually includes only three or four alternatives of units. In the case of Yuan Xiang Fang Residence, the designer emphasised on the similarity of units to ensure justice in the delivery of housing (Yao, 2012). Justice in social housing is often interpreted to mean homogenisation. Repetition in building prototype. As soon as a building plan proves to fit well with the planning code, housing design regulation and area control, it becomes an example for the other projects. Copying and repetition is commonplace to save time and cost. A preferred prototype in the study cases features three or four units with a vertical transportation core in the middle, and two thirds of the cases use this kind of floor plan for housing (Fig. 4).

Figure 3. The constraints on housing design in Shanghai Source: Yijia Dong

Figure 4. Typical floor plans of new social housing in Shanghai Source: Yijia Dong

The impacts on the urban form The efficiency oriented policy and regulation helps the government to achieve the annual targets. In 2009, new construction rose from four million to 10 million units. In 2011, this number increased further to 15 million units. From 2008 to 2012, the control plans of 21 large communities were approved. About 45 million km2 of new social housing were constructed, which means more than a third of the plan for suburban large communities is being built. Compared to the rapid and vast construction of social housing in Europe in 1920s and 1930s, rapid urbanisation in the suburbs of Shanghai goes further than the concept of functionalism.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Homogeneity of housing in larger scale Emerging rationality within an urban order usually appears with the necessity or opportunity to rapidly construct a large number of buildings and an agency that can assume this responsibility (Panerai et al., 2004). This rationality converges in the suburbs of Shanghai as a national political mission but it is realised on a much larger scale than the early practice of the Modern Movement. In the case of Yuan Xiang Fang, a resettlement project for the local farmers, all the living rooms and bedrooms are designed with the same width and same orientation. All the buildings are in the same prototype, with the same height and façade. On a site of over 19 hectares, this neighbourhood consists of 43 buildings that look the same. It features 3600 apartments, more than 10 times the Marseilles Unité d’Habitation, in which there are 337 apartments (Fig. 5).

Figure 5. The façade of Yuan Xiang Fang neighbourhood Source: Yijia Dong

Dissolution in highrises and discontinuity In the figure-ground analysis, 36 cases out of 38 are in the same parallel row structure (Fig. 6). Compared to the plan of Siedlung Westhausen built in 1927 (Fig. 7), both show a strong dominance of objects over the space and impacts of functional rules in organisation. However, given the requirements for high density communities, the new urban form of social housing in Shanghai results in dissolution of highrises and less consideration of scale and integrity.

Figure 6. Figure ground study on the new social housing projects in Shanghai Source: Yijia Dong

Figure 7. Ernst May: organisation of row houses in the Siedlung Westhausen (1929) Source: Panerai et al., 2004

Separated urban function If the Marseilles Unité d’Habitation presents Le Corbusier’s ambition of total control by the architect of a city (Panerai et al., 2004), the current practice of suburban social housing in Shanghai is a demonstration of the overall control by policy of a city. The policy is more focused on on-time completion of construction than the construction of livable places. The indifference towards the site and city tissue overwhelms the

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

large communities. Different functions are separated not only in space but also in construction phases. In many cases, the commercial and service facilities are isolated in blocks outside the gated community to facilitate centralised management by the district government.

Conclusions: conflicts and suggestions Conflicts If the purpose of social housing construction is to improve the living conditions of low-income groups and if the meaning of urban planning and building regulation is to provide a sustainable controlling frame, these concepts are missing from the current construction of large communities in Shanghai. Following the rules of the modernist city, which focus on building efficiency and average distribution of light and openness, the repetition of slab highrises in the suburbs of Shanghai reproduce the disconnection between dwellings and urban space, which has been investigated in the form of the Unité d’Habitation and the Siedlung Westhausen (Panerai et al., 2004). Through the case studies, a series of conflicts in the use of the space were observed: 1.

High living density but lack of urban vibrancy. In many cases, the housing was occupied soon after construction. The average living density of these neighbourhoods is over 800 people per hectare, based on an average FAR at 2.0, average unit area at 65m2, and family population of 2.7 people. This is similar to the density in the downtown Shanghai. However, instead of a busy urban life, the streets in the suburban communities are often empty. In the interviews with residents, many complained about the lack of community facilities, long distances to the public transport and inconvenience in grocery shopping. Interviewees commented that ‘Houses [are] only for sleep’, which indicates the poverty of urban vibrancy. High green area ratio but no public space. Although many of the cases provide a high ratio of green space, these places are often isolated pieces of grass measuring over 40m, and located between 60-metre-highrises. Often situated in the shadow of the highrises and with no adjacent facilities, these green spaces are often empty. A public plaza or street for civic activities was rarely identified. In some cases, the edges of the neighbourhood are blocked by a mass of infrastructure or parking areas, with little access between the street and the community. However, there is no rule to control this kind of design mistakes. Large amount of new construction but rejects on innovative design. Instead of stimulating a boost in innovative housing design, the current regulation and large amount of construction seldom tolerates new ideas. In the case of Lingang New Town, the architect initially proposed a block structure using a lowrise and open community model (Lu and Li, 2012) (Fig. 8). However, the plan was rejected in the planning due to orientation. A fast start but slowdown in developing process. The vast copies of buildings in huge scale distinguish the new social housing in the surrounding context. In the interviews, positive feedback on the impression of the large communities was rarely provided. The general description is ‘cheap’ and ‘low quality in design’. The original idea to stimulate the development with social housing projects doesn’t work very well with a slowdown in the construction of other kind of housing projects.

The survey shows the negative influence of the policies on the urban development, ranging from impacts of urban form to the conflicts. To avoid more mistakes in the future, it is recommended that the municipal government consider the sustainability ahead of the construction efficiency and consider the following alternatives of policies and regulations.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 8. Tongji Architecture Design and Research Institute: The proposal for the master plan of social housing neighbourhoods in Lingang New Town (2012) Source: Tongji Architecture Design and Research Institute

Suggestion 1: Break up the whole into parts Rather that the whole deal strategy in organisation, it is suggested that the large projects be broken into parts to achieve a controllable scale for diversity and cooperation. This would involve three corresponding contents: a) b) c)

Limit the size of the block to avoid super blocks, while providing human scale grids in the community centre for walkable streets Limit the size of the project within blocks to encourage a mix of different projects on the same block Encourage collaboration between multiple developers and architects to avoid single decisions.

Suggestion 2: Introducing an urban design leaded controlling system Urban design plays an important role in the urban frame in many large cities. Instead of control for a common living standard, urban design can provide coherent consideration of open space systems, the civic dimension and public realm. Some issues missing from the current practice need to be regained in the future: a) b) c) d)

Adopt mixed land use as the basic strategy to build a centre for a large community Emphasise a coherent open space system, creating public green space instead of separated green areas in gated communities Emphasise detailed design of streets and public spaces, while providing human scale, good accessibility and functional support for these areas Encourage innovative housing design, especially new ideas for mixed-income or mixedfunctional housing type, and ideas for new forms in communities, involving tolerance and compromise in regulation.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Suggestion 3: Provoke diversity in both planning and housing design levels Mono-form and function providest many lessons in the history of social housing. Instead of uniform production, diversity in residents group, functions and building form is beneficial for sustainable living environments. The achievement of diversity requires efforts that: a) b) c) d)

Introduce more types of housing into the large communities, giving alternatives to a wider range of residents Encourage the mix of multiple housing types and promote mixed-group living at the planning level in block scale Differentiate the density and height control by following an open space structure Limit repetition in housing design.

Above all these suggestions, it is necessary to reduce the speed of construction and to allow more space and time to create better urban spaces in the suburbs of Shanghai.

Acknowledgement This paper is supported by the National Natural Science Foundation of China, Research Project Number: 51278337.

References Lu, B. and Li, Z. (2012) ‘Planning and architectural design of affordable housing project WNW-C1 Block 6A in Shanghai Lingang New City: Typology innovation of affordable housing in China’, Urbanism and Architecture (105), 34-37. Panerai, P., Castex, J. and Depaule, J.C. (2004) ‘Urban forms: the death and life of the urban block’ (Architectural Press, Oxford). Shanghai Municipal Government (2011) The 12th five-years-plan on housing development in Shanghai (2011-2015). http://www.shanghai.gov.cn/shanghai/node2314/node25307/node25455/node25459/u21ai584440.html (2013-06) Shanghai Planning and Land Resource Administration Bureau and Shanghai Urban Planning and Design Research Institute. (2013) ‘The transformation of Shanghai’ (Tongji University Press, Shanghai). The management office of large communities (2012) ‘Portfolio: The selection of my favorite social housing’ (unpublished report, Shanghai). The State Council of China (undated). Suggestions on the solution for low-income urban families’ accommodation. National policy code:[2007]24. http://www.gov.cn/zwgk/2007-08/13/content_714481.htm (2013-06) Wang, Y. (2011) ‘Thinking on Large-scale residential community planning from the perspective of affordable housing construction’, Shanghai Urban Planning Review 2011 (3), 41-44 Yao, H. (2012) ‘Farmers relocation neighborhood design strategies: Yuanxiangfang Case’, Planners 2012 (28), 5-9

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

The urban village code in Guangzhou: Morphological self-evolution on the edge of the metropolis Quanle Huang Modern Architecture Design and Research Center, South China University of Technology, Guangzhou, China Email : archi_yellow@hotmail.com

Tao Li Urban Regeneration Research Center, Zhubo Design Group, Guangzhou, China Email: li.tao@zhubo.com

Abstract. The edge of the city does not necessarily mean featureless space where poor communities struggle to live in humble conditions. In some circumstances, to remain on the edge is a reasonable option for development in some communities. This paper explores the controversial case of the ongoing morphological transition of ex-rural villages through urbanisation. After losing farmland to new urban construction during the extensive urbanisation process of Guangzhou in the past 30 years, ex-rural communities which are now called “urban villages” have maintained a surprisingly dynamic self-development within and outside the village territory. Shipai Village is taken as a typical case study so as to examine the “code” that has led to its successful spatial and social self-evolution, and to reveal how a weak community can survive the pressure of the metropolis that encompasses it. Keywords: urbanisation, edged community, self-organisation, spatial transition, identity maintenance

The extremely rapid urbanisation in China during the past 30 years has been realised via a top-down model. During this period, as a most humble element in the heart of many new urbanised areas of the city, the urban village as a de facto urban community has seldom received attention. Urban village here refers to the original rural village which had all its farmland expropriated and urbanised by the city. At the same time, these villages retained their housing areas where the villagers live. Many of these urban villages can be found on the near fringes of city centres in China. These urban villages have not been studied from the perspective of its relationship with the urbanisation process or its mobility and mechanisms of selfevolution under pressure from the rapid urban expansion, which has overtaken territory once belonging to those villages on the edge of the metropolis. Shipai Village, as a typical example, is the main case study. It is located in the urban extension of Guangzhou city, 1.5km from the new city axis in Tianhe District, and in the heart of the urbanised area where four main universities and more than 20 grand IT markets are situated. All of these facilities are built on the territory that formerly belonged to Shipai village and was used for rice fields prior to urbanisation. Most of the urban villages in Guangzhou, such as Shipai village, have experienced rapid urban expropriation and lost their farmlands, which were transformed into multifunctional urban quarters. These villages have also participated actively in the ongoing urban development to meet the needs of

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

space by the urban market. In less than 20 years, between the 1980s and the 2000s, the village environment has been transformed, both functionally and socially, mostly by its collective leaders and villagers, so as to find the new ways of survival and development as urban habitats. Concurrently, the village continued its tradition originating from their agricultural roots, including conservation of religion and family ceremony sequences and strengthening the correlation between public and private spaces, so as to retain its own identity during the struggle for existence resulting from urbanisation. Urban villages in Guangzhou today The transformation of the urban villages in Guangzhou is generally considered to be caused by three aspects: the urban extension, the dual land system, and migration from outside the city. The urban villages transformed during the urbanisation As a capital city in one of the most developed regions in China, Guangzhou has doubled its urbanised territory in less than 20 years, from 136km² in 1980 to 276km² in 1998 (Fig. 1). The extended area enclosed no doubt a large number of rural villages, where vegetables were once produced for the city. At present, there are 139 administrative villages that are actively present in the highly urbanised city area (Fig. 2).

Figure 1. The extension of the urbanised area of Guangzhou city, and the changing relationship between the city and Shipai village (red area in the plan) Source: Figures edited by author on the base of plans provided by Guangzhou Achieves Bureau

These villages were formerly the habitat of the farmers who worked in the surrounding farmland, providing fresh produce to the city prior to the city’s large-scale extension. Now, they are scattered over several branches of the new urbanised area and present spatial homogeneity in relation to the distance between them and their interior spatial morphology. The city’s developers have overtaken all the farmland surrounding a village, although the village housing area continues to survive in the current urban territory. The expropriation of the periphery land has been an incomplete process. This is the result of the dual land system. Dual land system and the system of reserved land for urban villages Since 1956, when the “socialist reform” of the “private ownership of the means of production” was done, the land has been divided into a dual ownership system: the city land is owned by the nation and the rural land by the collective organisation. These two land ownerships do not share either the same disciplines or the same decision-making of land treatment. Even though the relative clauses were not written into the Constitutions of the People’s Republic of China until 1984, this division of land ownership has always been part of basic law which has profound influences on urban and rural development, especially when the land transitions from rural to urban. The extension of the city was realised by the expropriation of the farmland owned by the collective organisations. The expropriated land was then owned by the nation and used by different units that have the permit and have paid for the land use right.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 2. The urban villages that are found amid the extended urban constructed area in Guangzhou, 2005. A. San-yuan-li village fragments to the north of the Central Railway Station, surrounded by the urban fabric, 2005. B. The village habitat fragments in the quarter of the ongoing project of CBD in the east of Guangzhou, 2005. C. Xian Village mass, in front of the highrise buildings in the CBD area, 2007 (photo by author). Source: “Research on the countermeasure for the planning, construction and management of the ‘urban villages’ in Guangzhou”, edited by Guangzhou Urban Planning Bureau and Zhongshan University, 2000

Additionally, in Guangzhou, there is a special local policy concerning the expropriation called the “Reserved Land Policy” (liuyongdi zhengce). This can be seen as a “creation” aiming to solve the problem of land conflicts from the angle of the government: in order to “get” the land smoothly from the collective, the policy provides that some of the expropriated land must be returned to the collective. The early clause of this policy was found in the “Regulations on the compensations for the land expropriation and removal house by the nation” (Guangzhou government, 1993), clause N° 11: In order to meet the villagers’ need of life and production after the land expropriation, 5-8% of the total expropriated land shall be reserved to the village collective for the development of second and third production, and for the resettlement of the surplus labor force. If the whole land of the village is expropriated at one time, 8-10% of the total expropriated land shall be reserved for the village. In 1995, the Standing Committee of the National People’s Congress (NPC) in Guangzhou adopted the new

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

“Regulations for the land management in Guangzhou” was adopted and clause N° 16 states that “the administrative departments in charge of the land who expropriate the land from the collective must keep reserved land according to the following quota: 1) in city jurisdictional area: 8-10% of the total expropriated land; 2) in the manage area: 10-12% of the total expropriated land.” This has been a temporary measure that has proven effective and expedient during the 1990s when structural economic reform commenced, and the unemployment in the city was increasing. It was impossible for the government to resettle all the farmers whose land was expropriated. At the same time, the government did not have sufficient economic capability to pay for the expropriation at a reasonable price. At the beginning of 1980s, the expropriate price was only a few thousand Yuan RMB per mu (1 mu=0.0667 hectares). By the end of the 1990s, this had reached tens of thousands Yuan per mu, then hundreds of thousands Yuan per mu in the 2000s. The expropriators had paid only for the attachment and young crops on the land but not for the land itself based on market values. Conflicts and resistances from the village collective against the expropriation have become stronger, and they have blocked urban expansion. During the 1990s, the government responded with this policy of concession which seemed necessary for social stability and particularly for urban construction and economic development (Fig. 3).

Figure 3. The changing territory of Shipai village during the urbanisation process. Source: Plans drawn by author according to site surveys and interviews.

The villages where the farmers lived were retained along with parts of the expropriated land which were returned to the village on separated sites. As the reserved lands were returned separately by the expropriators, they had added to the fragmentation of the land distribution. Before the Guangzhou government ceased the policy of “reserved land for the collective” in 2002, there were already hundreds of odd pieces of reserved land that were scattered around the urban villages. Over the past 20 years, urban extension continued with more preserved village areas were totally enclosed by the newly built urban infrastructure and towers. These became isolated fragments, like the islets on the vast expanse of urban fabrics. What happened then in these “islets” was out of the urban planner’s prevision. The villagers transformed their one-floor houses in the village into multi-floor apartments for rent, and they built towers on the reserved land for commercial and service use. They raised the building volumes both vertically and horizontally. The villages were filled up with building blocks, and their physical appearance was changed completely. There was no more natural landscape and one-floor housing in harmony with the field. The urbanisation and the pressure of making a living forced the villagers to transform their last land holdings into built entities to meet the needs of the market economy. This had been done on the collectively owned land which was not controlled by urban planning regulations (Fig. 4). The urban villages in favour of migrants The third aspect can explain the intention of these transformations in the urban villages: accommodating the great tide of migrants from rural regions to large cities, which has resulted in a “peasant invasion”

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

(Chan, 1999). In 1979, the restrictions on migration defined by the Hukou Policy were relaxed. The Hukou Policy is the national system which defines each Chinese citizen or peasant’s identity by their location and profession, tying people a place of residence. It regulates and controls the population’s geographical transfer with administrative process. The transfer confers legal residency rights and eligibility for many urban jobs, with accompanied subsidised welfare benefits. Peasants and citizens were consequently divided into two worlds with invisible walls just after the issue of the Policy of Reform and Opening, with tens of millions of surplus laborers, who had previously been tied to a place of residence in the countryside, pouring into medium and large cities in search of jobs and better living conditions. The number of migrants (the so-called “floating population” or “liudong renkou” in Chinese) has been rising except in the three years from 1988 to 1990, when the state issued the “Economic Austerity Program” to adjust for over-heated development.

Figure 4. The densification and augmentation of buildings in Shipai village’s last territory of residence. Source: Drawn by author and Gang Song according to plans provided by Shipai village collective.

Most of the migrants met the need for labor in cities during the economic boom, yet at the same time they triggered “an unprecedented social change and a more diverse, open and plural society” in urban populations (Chan, 1999, pp. 49-68). However, the cities were not ready to house them. There was no social housing system to help low-income people in the Chinese urban system, and the cost of buying or renting an apartment in the real estate market is far beyond the financial capacity of migrants. The severe problem of housing the newcomers was “automatically” addressed without the government’s intervening action. The urban villagers who were seeking redevelopment met the need of migrants who were in search of housing. During the end of 1980s and the beginning of 1990s, the period when migration boomed in Guangzhou, almost all the urban villages were in at the peak of reconstruction. Most of the villagers’ houses were rebuilt, from one storied farm-house to multi-storied block, so as to provide rentable rooms to the migrants (Fig. 5 and Fig. 6).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

100

Figure 5. Multi-reconstruction on the same piece of land by the villager owner of residential base. Phase one is the legal transformation from one floor to three and a half floors. Source: Photo and drawings by author.

In these villages, the proportion of the original habitants and the tenants is 1:4 or more. In Shipai village, which was studied, there are 9,317 villagers and 42,000 temporary habitants (tenants) who lived together in an area of 0.7kmÂ˛, resulting in a population density of more than 70,000 per square kilometre. According to the official statistical data, in 1999 there were more than 135,000 apartments or rooms to let in the urban villages where around 1.82 million of temporary habitants lived. The additional rooms in these villages have housed more than 60% of the newcomers for the city in this year (Fig. 7).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

101

Figure 6. Multi-reconstruction on the same piece of land by the villager owner of residential base. Phase two is the illegal reconstruction from three and a half floors to seven or eight floors. Source: Modeled by author

Figure 7. Shipai village’s metamorphosis from 1978 (left) to 2000 (right), as a result of both urban expansion and the village’s self-transformation. Source: Plan and photo provided by Shipai village collective.

The urban village code The urban villages entered the urban world in a way that embarrassed the metropolis, with the low-rise but extremely high density buildings and the massive “rural” housing image. However, at the same time, a very strong dynamic from inside the village exist that pushes the urbanisation movement forward, and this is seldom observed and studied.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

102

They are misunderstood, and seen as a mass of small but densely collaged houses while various reports identify social problems that they have brought to the city. Local government has realised that these villages have contributed much to address the problem of housing low-income migrants. However, continued research is required to find the way of merging these “village” fragments into the urban structure. It is necessary for us to understand the villages’ own process of structural transformation, which might help to develop a complete vision. Facing the pressure of rapid urban transformation, what has been happening in the urban villages? How have they maintained a way of existence in such a strange urban world?

The tradition keeps working in the inner power organisation In order to better understand the strong autonomous dynamic of the village, the migration history of the villages in the south of China should be emphasised, as this has exerted a continuous and profound influence in the maintenance of internal power structures, as well as in the ongoing self-transformation of the village. The predecessors of today’s villagers had also undergone a long voyage before they settled in the rural area where they built the village. This voyage, to escape from the wars during the dynasties of Tang, Song and Yuan, involved tramping over mountains and through ravines, from the Central Plains to the southern regions beyond the Nanling Mountains, towards different provinces like Guangdong and Fujian. They moved in entire family groups so as to be strong enough to protect themselves from the turbulence social situation and natural difficulties, to gather all their forces and resources to rebuild their new settlements, and to compete or coexist with the local habitants. The family clan played a decisive role in the village’s management and organisation (Freedman 1958, 1966). There are four main family clans in Shipai village: families of Dong, Chi, Pan and Xian. Over a long period in the history, they have established an autonomous system for coexistence and development. The highest folk organ for power was called the Neutral Hall (zhongli tang), which consisted of the respected clansmen elected from each grand family and dealt with the most important affairs in the village according to traditionally recognised principles and conventions. As the subordinate power institution, each grand family elected their own responsible clansmen to manage the family affairs. The expenditure of the Neutral Hall was raised from these four families as a proportion based on their population: 40% from Chi Family, 30% from Dong Family, 20% from Pan Family, and 10% from Xian Family. Following the same principle, to maintain the village’s security, the Neutral Hall founded a group of guardians which was comprised of six persons from Chi Family, five from Dong Family, four from Pan Family and one from Xian Family. The revenue raised by the Neutral Hall was used mainly for important religious ceremonies and cultural festivals, construction of infrastructure, and maintenance of security in the village. At the same time, the organisation dominated 44.8% of the total farmland of the village in the name of the ancestor, which has enhanced its power position by strengthening their economic foundation. This kind of farmland was called the “field of the great-grandfather (taigong tian)” and was let to farmers to cultivate.

The assimilation of the national and local power structure Both historically and today, the national power force has always intervened in the control of rural society. In those villages where genetic connection was the original and basic force in autonomy, the state made use of this force to realise its intervention. This has been particularly the case after 1949, when the government carried out a series of reforms in rural society to replace the “old” system. From 1949-1958, during the period of the land reform, the administrative unit of the township level officially entered the village system, and the responsible persons in the unit were still family members selected on the basis of their proportion of population in the village. In 1958, the People’s Commune System (renmin gongshe zhidu) was introduced in all rural villages in China, and this transformed the organisation of rural society. The villagers were divided into many labor

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

103

units called “productive teams (shengchan dui)”, and the means of production as well as the land were shared equally in accordance with the ideological system. This was the strongest intervention of national power into the rural system and replaced the family controlled system. However, following the disasters of famine due to crop failures and the low labor efficiency caused by the egalitarianism this form of organisation ceased. The power that was centralised in the People’s Commune was returned to the family clans system, even though this later kept changing its title. The Party branch secretary role in Shipai was held by the clansman who held highest status in the village. The post was renamed several times and designations included Secretary of the Party Committee, Chairman of the Company Group, and General Manager of the Joint-Stock Company. Regardless of the title, the person entitled to hold it was always the most respected family member elected by the villagers. An overview of cadre statistics since 1949 shows a consistently balanced distribution of cadre members in the proportion to the population composition in Shipai village (Table 1). As the largest sub-community in the village, the Chi family has held the most positions in the power institution. However, the balance of organisation was well maintained, as clansmen from the three largest families’ took turns to be the director of the village.

Family’s Name

Table 1. The composition of population and cadre members in Shipai village Chi Dong Pan

Population in 2005 Proportion of the total population Number of cadre members since 1949 (person-time) Number of successive Secretary of the Party Committee

Xian and others

3498 37.5% 52

2189 23.5% 14

1358 14.6% 25

2272 24.4% 9

In 1997, when Shipai village lost its last pieces of farmland, the collective decided to reform the organisational system. Since the village was already part of the Tianhe District in the city, the previous villagers’ committee was rescinded. The new founded joint-stock company took over the same leading group who continued dealing with the village’s affairs, even though the administrative function was nominally transferred to the street office (jiedao banshichu, which is the subordinate administrative unit in the city’s organisation). This phenomenon warrants further observation. The system reform in Shipai separated administrative function from economic function. The former was incorporated into the city’s government system, and the later was retained by the collective. However, in reality, no administrative decree could have been carried out without the intervention of the local leading group. That is why the government of Tianhe District adopted a realistic attitude when it approved the foundation of the joint-stock company. The commission of the Tianhe District government provided that considering that the local organizational system has functioned over a long period of time, and that some of the folk regulations and conventions still play a positive role in the maintenance of the socioeconomical stability, the government authorizes the provisional reservation of the function and official seal of the villagers’ committee for the company, and authorizes the temporary administrative department to coordinate the appropriation, village restorations, security, education, sanitation and all the other affairs concerning the system construction (Tianhe District government, 1997). From 1997 to 2007, the “temporary administrative function” that the joint-stock company managed is still in progress. The collective has not only realised the management of the community, but also invested a great deal in constructing urban infrastructure facilities, like roads outside and schools inside the villagers’ habitat.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

104

On one hand, the collective carried out the city’s function and succeeded in preserving its own power structure. On the other hand, the local government took advantage of this process through its absence from administrative management and during urban construction.

Culture, memory and self-protection For the past 700 years, four main families lived in compact communities and formed this Shipai village. The Family of Pan lived mainly in the north-west of the village, the Family of Xian lived mainly in the north-east, the Family of Chi lived mainly in the north and in the east, and the Family of Dong in the middle and the south of the village. Since the last two families have kept connections through marriages, their houses are quite mixed. This state of space distribution remains unchanged to today (Fig. 8). Each big family has built their ancestors’ temple and gods’ temple, and thus maintained a sense of “family center” in the place. What we would emphasis is that the elements of cult, including the ancestors’ temples and the varied gods’ temples, have always been sacred in family members’ beliefs.

Figure 8. The actual owners of the houses in Shipai village Source. Drawn by the author according to the proprietor register document of the collective in 2005

The strengthened collective identity in the urbanisation process Together with rapid urbanisation, the temples were restored and their original character as the spiritual centre of the family was reinforced at the beginning of 2000s. The village achieved continuous productivity even after having lost its farmland. It shows the external dynamic as an active participation to the urbanisation process, and the internal dynamic by the consolidation of the traditional power and the spatial organisation (Fig. 9).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

105

Figure 9. The clan tradition is well preserved and serves as a collective and cultural power against the strong pressure from the metropolis to delete this sort of community from the new urbanised territory.

Conclusion: the self-evolution mechanism of urban villages, which are spatially and socially on the edge of the metropolis How far is it from the village to the city? Ten minute’s walk, even less. How long will it take to make a village integrate into the city? More than 20 years, and is it not yet accomplished. The city’s spatial extension has provided these villages with new opportunities for development. First, the expropriation indemnity has helped the collective to realise funds accumulation. Second, the proximity of these villages to the city’s functional centres has attracted external investments. Finally, the low life cost in the village habitat has attracted a great number of tenants. The urban villages have benefited much from its external urbanisation process. However, there exists a double definition of the “space value” for the urban village. The metropolis has tended to give prominence to its centre and thus de-emphasised existing cultural and spatial elements of the territory that it has enclosed. It has assimilated its periphery and suburbs by overlooking their value in history, as a potential contribution to the diversity of the urban space. This might be a reflection of the impact of the globalisation. This article has sought to reveal the strong dynamic and abundant characteristics of the urban villages that are still active in the contemporary urbanisation process. While there is spatial chaos in the merging of areas between city and its periphery, the “urban village” has become a de facto urban community that regroups certain types of population in considerable number and scale.

References Armstrong Warwick, Mcgee T.G., (1985) Theatres of Accumulation-Studies in Asian and Latin American urbanisation, (Methuen, London & New York).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

106

Beauchard Jacques (dir.) (1996) The city-rural: towards an alternative for the metropolization (La ville-pays : vers une alternative à la metropolization), Aube (Coll. « Territoires. Série Cités et territoires ») (France). Berlage Institute (2005) Village in the City: Unknown Urbanity in China, Research Report (PRO1-07.05) (Rotterdam). Chan Kam Wing (1994). Cities with Invisible Walls: Reinterpreting urbanisation in Post-1949 China (Oxford University Press, Hong Kong). Chan, K.W. (1999) Internal migration in China: a dualistic approach, in Internal and International Migration: Chinese Perspectives by F.N.Pieke & H. Mallee (Curzon, Surney). Clément Pierre, Goldblum Charles (1994) Asian cities (Cités d’Asie), (« Cahiers de la Recherche Architecturale », n° 35/36) (Parenthèses, France). Gransow Bettina, LI Han-lin (2001) Villagers in the city: Rural migrants in Chinese metropolises (dushi li de cunmin : zhongguo dachengshi de liudong renkou) (China Central Translation Press, Beijing). Guangzhou government (1990) Annals of Tian-He District (Tianhe Quzhi), Guangdong People’s Press, Guangzhou. Guangzhou Tianhe District Shipai Committee (Eds.), 2003. Annals of Shipai Village (Shipai Cunzhi) (Guangdong People’s Press, Guangzhou). Guangzhou Urban Planning Institut (2001) Urban design project of the new city axe. Guldin Gregory Eliyu ( 2001) What’s a Peasant to Do? – Village becoming town in Southern China (Westview Press, U.S. / U.K.). Hoa Léon (1981) Reconstruct China: thirty years of urban planning, 1949-1979 (Reconstruire la Chine—trente ans d’urbanisme, 1949~1979) (Le Moniteur, Paris). Huang Quanle, Li Tao (2014) In the shadow of the metropolis—Shipai village, in: De Meulder B., Lin Y.L. and Shannon K. (eds.), Village in the City (Park Books, Zürich). Huang Quanle (2015) Metropolis of ruralty: a spatial history of Shipai in Guangzhou from the perspectives of typomorphology (1978-2008) (China Architecture Industry Edition, Beijing). Li Pei-lin (2004) The end of the village: legend of the Yangcheng Village (cunluo de zhongjie : Yang-cheng cun de gushi), (Business Book Press, Beijing). Li Ping-ping, Lu Chuan-ting, etc. (2002) Studies on strategic planning for Guang-zhou city, China Architecture Press, Beijing. Ma, Laurence J.C., Wu, F.L. (2005) Restructuring Chinese City: changing society, economy and space, Routledge, London. Mirloup Joël (direction) (2002) Perimetropolitain Regions and Metropolization (Régions périmétropolitaines et métropolisation), Collection du CEDETE (Presses universitaires Orléans). Pieke Frank N., Mallee Hein (1999) Internal and International Migration: Chinese perspective, Curzon Press, Surrey. Shen Jianfa (Eds.: Diamond D. & Massam B.), 1996. Internal Migration and Regional Population Dynamics in China, (Coll. “Progress in Planning”) (Pergamon). Stanilov Kiril (2004) Suburban form: an international perspective (Andover, Routledge, London). Whitehand J.W.R., Carr C.M.H. (2001) Twentieth-century Suburbs: a morphological approach (Routledge, London). Xu Xueqiang, Li Lixun (2000) Research on the countermeasure for the planning, construction and management of the “urban villages” in Guangzhou (Guangzhou Urban Planning Bureau and Zhongshan University, Guangzhou). Zeng Zhaoxuan (1991) Historical Geography of Guangzhou (Guangzhou Lishi Dili) (Guangdong People’s Press, Guangzhou). Zhang Jian-ming (2003) Studies on Urban Villages in Guangzhou (Guangzhou chengzhongcun yanjiu) (Guangdong People’s Press, Guangzhou). Zheng Mengxuan, Huang Zhaowang, (Eds.) (2006) The urbanisation of Shipai Village (chengshi hua zhong de Shipai cun) (Social Sciences Academic Press (China), Beijing).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

107

The evolution of urban tissue and building types in the Guangfunan area of Guangzhou, from 1840 onward Ye Li School of Architecture and Planning, The University of Auckland, Auckland, New Zealand Email: yli602@aucklanduni.ac.nz

Pierre Gauthier Department of Geography, Planning and Environment, Concordia University, Montreal, Canada Email: pierre.gauthier@concordia.ca

Abstract. With reference to the typological theories and methods of the Italian school, this paper proposes a fresh perspective for morphological studies in the Chinese context. A case study was conducted in the Guangfunan area, a suburb located immediately west of the old city wall of Guangzhou, China, which used contemporary and historical cartographic documents, and data collected during an extensive field survey as well as secondary sources. The evolution of the residential forms and urban tissues from 1840 onwards is examined. The study addresses significant relationships between tissue configurations and inherited geo-morphological conditions as well as old settlement patterns and identifying architectural types and their variants. The authors also reveal a morphogenetic process according to which urbanisation and densification produce a series of residential forms that derive from one another. The study illustrates how spontaneous and purposeful building practices both respond to geographically bounded conditions and to constraints and potential for change ingrained into the morphological system. The research provides a better understanding of the morphogenesis of the Guangfunan area in Guangzhou. In doing so, it offers a new perspective on issues such as urban landscape management and historical conservation planning, at a time when China must cope with unprecedented pressures exerted on the built landscapes of cities. Keywords: urban tissue, building types, typological processes, morphogenesis, Guangzhou, China

Applying the Italian methods in a case study of Guangzhou Urban morphology is defined as the study of the spatial and physical characteristics of towns and cities as a result of the spontaneous and purposeful building practices governed by a habitus. Relying on theories and methods from cultural geography and architecture, studies of urban morphology focus on the built environment as well as architecture and the urban landscape. In the last few decades, the research of urban morphology and building typology has been widely explored and applied to diverse urban contexts in European and to a lesser extent to North America (Moudon, 1986; Baker and Slater, 1992; Cosini, 1997; Malfroy, 2001; Gauthier, 2003; Hofmeister, 2004). However, the approach of urban morphology of European schools has only been introduced to China in the last decade. Most of them are following the Conzenian ideas of the British School (Gu, 2001; Whitehand and Gu, 2006; Whitehand and Gu, 2007; Gu et al., 2008; Chen, 2010; Tian et al., 2010; Whitehand et al., 2011; Xu, 2012). China has an urban history extending over some 3500 years and the case of this research, Guangzhou has a history of some 2000 years. However, the typological theories were not discussed among Chinese architects until the late 1980s. The studies thus far have not developed a thorough theoretical model of the physical change process, or adapted such a theoretical approach for a case study in China. The absence of a solid theoretical foundation is a major problem in current practice of urban planning and design, especially for

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

108

conservation planning in China. Therefore, this research addresses this by conducting a case study of Cantonese dwellings and urban tissues in suburban Guangzhou that is informed by the typological theories and methods of Caniggian School from Italy. The Guangfunan area of Guangzhou is to the west of its most recent city wall and has a historical record dating back to the 6th century (Fig. 1). Yangren-li is the oldest street with a textual record to date (Zeng, 1991). In the core of the Guangfunan area, and parallel to the Yangren-li, a waterway named Daguan was canalised in 1472 during the Ming Dynasty. This canal was connected to the west city moat and flowed into the Pearl River. It provided the Guangfunan area with access to trade and transit by both land and seaway. In the 15th century, the Guangfunan area was rapidly shaped as a major commercial hub in the west part of Guangzhou. From the late 19th century, the Daguan canal started to dry out from its upstream and the deposited watercourse was replaced by residential dwellings or settling routes, which resulted in significant change in layout and street patterns of this area (Fig. 2). In 2006, in the proposal of the Masterplan of Guangzhou (2010-2020), the Guangfunan area was identified as a Historico-Cultural Conservation Area (HCCA) (lishi wenhua baohuqu). The research institute tried to write a specific regulation and public policy, to promote a conservation project in this community. To date, after the rapid development in the last few decades, Guangfunan is one of the few hybrid neighbourhoods comprised of traditional dwellings, early 20th century multi-storey buildings and modernist mid and high rise structures. The complexity of the building types and street system makes this area a valuable case study for conducting research on urban morphogenesis and typological process in southern China.

Figure 1. The location the Guangfunan area and the peripheral environment.

Figure 2. The navigable canals and 18-Fu commercial hub in Guangfunan over the Qing Dynasty.

In the morphological research tradition, the Italian School mainly derives from the idea of “studies for an operative history of cities” which was inaugurated by the Italian architect Savertio Muratori (Malfroy, 1998). The Italian School considers the urban themes as an architectural organism (Cataldi et al., 2002). Gainfranco Caniggia continued Muratori’s work by theorising the ‘typological process’ of the city of Como (Caniggia, 1963). The Italian School of Process Typology posits the possibility of conceptualising the built environment as a complex and dynamic system of interrelated elements, which functions at different scalar levels. According to Caniggia, the ‘type’ is the set of conventions and norms developed in the course of the building experience (Caniggia and Maffei, 2001). Therefore, one can obtain what he called the ‘typological process’ by examining types as they are progressively transformed in successive phases. Moreover, the typological method can be used not only when considering the building structure itself, but also, for the explanation of the evolutionary process of a larger area, like a neighbourhood or even a city. Based on this argument, Caniggia and Maffei brought out the concept of ‘urban tissue’, which is comprised of elements belonging to three different sub-systems that they defined as streets, allotment system, and the building coverage (Caniggia and Maffei, 2001). Therefore, the method of the Italian School relies on

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

109

two principal analytical procedures, a genetic reconstruction, and the construction of a typology. This research is a two-prong morphological analysis that presents the results on two different scales: first the morphogenesis of the urban tissues, then the typological analysis of buildings. This case study of the genesis of urban and architectural forms in Guangzhou’s Guangfunan area draws heavily on cartographic documentation, iconography and data generated during fieldwork in the form of a comprehensive photographic survey and by observing and mapping architectural attributes of buildings and streetscapes. The cartographic analysis is one the basic methods in morphological studies that aims to explore changes in the urban fabric over time. However, compared to most European and American cities, compiling the historical maps and data in Guangzhou or any other Chinese city is more difficult because of the lack of accurate historical maps showing building block-plans and the lifespan of the buildings. Some of these maps are diagrammatic drawings and only show the built-up area and streets (see, for example, Zhou and Xiao, 2003). The first systematic urban survey of Guangzhou was conducted by Guangzhou Land Bureau (Guangzhou Tudiju) from 1926 to 1935 (Whitehand et al., 2011). The large scale plans of the city have been prepared by Guangzhou Urban Planning and Design Survey Research Institute since 1949, and this city plan has had minor modifications several times during the last three decades to document changes due to the rapid urban construction. All of these maps are important sources for supporting a preliminary cartographic analysis in this research. After a preliminary cartographic assessment of the morphological conditions of the study area, fieldwork is another indispensable method for data acquisition in this research especially for those built–lots and building footprints that show unusual configurations and patterns. Following the fieldwork, building attributes were tabulated, mapped and triangulated with cartographic representations of lots, building footprints and street configurations and dimensions. This data was complemented by secondary source material which provided background information on the historical circumstances that have affected the study area.

The morphogenesis of the Guangfunan area Urban morphogenesis is the “creation and subsequent transformation of urban form” (Vance, 1977). More specifically, the making of urban tissue is the outcome of a dialectical interplay between purposeful planning practices, everyday “spontaneous” practices, and the structural resilience of the inherited built environment itself as well as the building culture of which it is the product (Gauthier, 2005). The reconstruction of the early settlement in the study area is based on the assessment of the historical maps and retrieved texture records from ancient books. The study area was located in the suburbs and was farmland back to the 6th century, when it first appears in historical records named as Yangren Fang. According to historical documentation, the shoreline of the Pearl River has been gradually moved to the southwest and the study area had been built on the reclaimed land from 6th century onwards (Fig. 3). As the substratum of other results, the reconstruction provides a general picture of the evolution of the Guangfunan area of its pre-urbanisation and early urbanisation periods. An archaeological excavation in 1911 revealed a tombstone with a record of 607 AD and indicated that the Yangren Fang, the core area of today’s Guangfunan area, was the private dwelling of Wang (Zeng, 1991). Hualin Temple was close to the waterfront and the built-up area along the Pearl River served the commodity trading between Chinese traders and merchants from west Asia. With the shoreline moving gradually towards the southwest direction, following the fluvial deposits, the farmland and vegetation area developed into a small town in the Song Dynasty. A sizeable commercial and residential settlement spread along the coastline and canals, while the sea lanes of merchant service connected the trades between China and more than fifty countries (as cited in Yang and Zhong, 1996, p. 132). This settlement on the fluvial area is the first built-up area to the west of the Song Guangzhou City and also the origin of the commercial hub in Ming and Qing dynasties. The canalisation of the Daguan canal provided a unique geographical advantage in the development of Guangfunan while the Eighteen-Fu commercial hub began to blossom along the canals in the Ming Dynasties. A catastrophic fire burned down the buildings in the southeast of Guangfunan area in 1822 necessitating the redevelopment of the commercial streets and the rebuilding of the dwellings. Most rebuilt lots changed the land use from commercial to residential use.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

110

These new structures closely followed the original models on the ground plan, but showed western neoclassical external ornament on the faรงade. After the lots were rebuilt, the street patterns and allotment system maintained a similar structure to that in the last century (Fig. 4).

Figure 3. The changes to the shoreline of the Pearl River over history. Source: Based on Zeng (1991) and Zhou and Xiao (2003).

Figure 4 The theoretical reconstruction of the early settlement patterns in the Guangfunan area..

After the plot-by-plot fieldwork, building attributes were tabulated, mapped and triangulated with cartographic representations of lots, building footprints and street configuration and dimensions. The characteristics of the urban landscape and urban tissues of the Guangfunan area were then delineated (Fig. 5-8), and the whole area was divided into different morphological zones based on the geomorphological conditions and settlement patterns (Fig. 9). The Traditional Residential Dwelling Zone (TRDZ) is the largest zone in the core study area. It consists of a sizeable standard traditional built lot that is based on a four metre width module. Most lots in TRDZ were occupied originally by zhutongwu, the most common building type in the Xiguan area. In most cases, dwellings were first built back to back and deployed in a linear shape. A single family occupied each dwelling. The most widely prevalent allotment in TRDZ is a rectangle lot of 1-jian wide, which could be consider as a basic unit in the study area. Therefore, the lots in TRDZ allow the most diverse transformation during the process among the whole study area. There are two shapes of allotment in this zone, the rectangle lot and L-shape lot (Fig. 10).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

111

Figure 5. The current morphological conditions in the Guangfunan area: Land use. Source: Based on authors’ field survey.

Figure 7. The current morphological conditions in the Guangfunan area: Periodisation of the buildings. Source: Based on authors’ field survey.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 6. The current morphological conditions in the Guangfunan area: Street hierarchy. Source: Based on authors’ field survey.

Figure 8. The current morphological conditions in the Guangfunan area: Building height. Source: Based on authors’ field survey.

112

Figure 9. Delineating morphological zones in the study area.

Figure 10. The types of built lots in the Traditional Residential Dwelling Zone. Source: Based on authorsâ&#x20AC;&#x2122; field survey.

The rectangular lots are easy to combine together as a double allotment in later transformations because of their regular shape. In some cases, the buildings along the traffic road have been converted into commercial use and subdivided to accommodate multiple shopfronts. In contrast, buildings in the Lshaped lot would have a small courtyard in the back, which allows natural light into the rooms. This is an effective way for getting more natural light in some particularly deep long lots. There is a street hierarchy in this zone. The smaller lots are located on a lower level street, which has a dwellings subdivision. In some cases, a cul-de-sac of less than one metre wide is the only street that connects a group of dwellings to outside in the core of units (Fig. 11).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

113

Figure 11. The street hierarchy and photographs of street views in the Traditional Residential Dwelling Zone. Source: Authorâ&#x20AC;&#x2122;s photographs.

The Former Watercourse Zone (FWZ) is located right above the watercourse of Daguan Canal. The development of the urban tissues in the FWZ has been strongly influenced by the watercourse of the former river, as well as the shape of built lots that were created on reclaimed river land. Following the natural conditions, the organisation of the allotment in the FWZ presents a natural shape and its boundary derives from the former watercourse. The dimensions of the lots fit the width of the river as described in historical records. In the first phase, buildings were built along the canal for the convenience of the waterborne trade. New buildings appear on the reclaimed land when the watercourses narrowed in the second phase. The pre-existing small buildings continued to extend horizontally by replacing the dried out area with backyards and patios in order to fit into the new lot sizes. After the canal dried up, the preexisting buildings are expanded or replaced by new constructions. However, the dimension of the built lots has been limited by the former canal, so the ratio of plot width to plot depth is about 1:1 to 1:2. Although this is not the ideal ratio for zhutongwu, these buildings still keep the one-room-and-one-aisle characteristic in the internal configuration to fit the four metre wide module. These buildings are considered to be a synchronic variant of the generic zhutongwu type, particularly an adaptation of the type to unusual circumstances (Fig. 12). The Former Riverfront Zone consists of a series of linear blocks along the canal in its initial phase. Most of these blocks were first developed as a waterfront commercial street after the canalisation of Daguan canal

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

114

in the 15th century. After the canal dried up in the early 19th century, the waterfront shopfronts lost their advantage and some of buildings underwent a land use change from commercial to residential in the later development. Meanwhile, new breakthrough routes positioned perpendicular to the canal increased the accessibility from the shopfront to the warehouses and dwellings in the backstreet. The pre-existing lots located at the corners were subdivided into small lots facing the new routes (Fig. 13). Some of the buildings in this zone still left the elements of the waterfront architecture, such as small footsteps outside the buildings that denote pre-existing private dock along the former canal. This has a high heritage value for the inherited built environments (Fig. 14).

Figure 12. Theoretical reconstruction of urban tissues formation and transformation phases of the Former Watercourse Zone.

Figure 13. The theoretical reconstruction of urban tissues formation and transformation phases of the Former Riverfront Zone.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

115

Figure 14. Photos of outdoor footsteps, streets and dwellings in the Former Riverfront Zone. Source: Author’s photographs. The Ancient Commercial Street is derived from the 18-Fu commercial hub. Characterised by the specific building type, qilou – a commercial building with an arcade cover the sidewalk at front – this unit represents the character of early 20th century commercial-residential neighbourhood in west Guangzhou (Fig. 15).

Figure 15. The distribution of Qilous and shopfronts along the Ancient Commercial Streets. Source: Based on authors’ field survey.

In the case of the pre-existing streets and buildings, the arcades were first built as an additional structure that attach to the buildings, but used different material and decorative styles on façades. The arcades added in front of the buildings not only successfully seperate pedestrians from vehicles, but also provide shade to the pedestrians. In the later development, the size of built lots became enlarged to accommodate these new structures, and gradually led to a new built qilou as a newer architectural type that fits the built lot area (Fig. 16). The Western style façade of qilou reflected the impacts of a variety of Western architectural styles on indigenous dwellings, and this Westernised style has influenced the building construction of Guangzhou in the last few decades.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

116

Figure 16. The subsequent transformation of built lots along the commercial street.

The modern unit shows the transformation of socialist construction. A set of pre-existing dwellings had been demolished and replaced by enclosed work units in the 1960s. Some built lots have been amalgamated and rebuilt as commercial and residential complexes after the work unit system declined in the late 1990s. This atypical circumstance does not fit the continuity of the morphological process as the previous zones, but it represents a manifestation of the centrally-controlled planning system of contemporary China from 1949 onward (Fig. 17).

Figure 17. Transformation of land use and lot size in the Modern Unit. Source: Based on Guangzhou Minguo Jingjie Tu, sheet 122-123, Guangfunan area in Guangzhou (Guangzhou Municipal Archives of Urban Development), Plans of the Guangfunan area in 1959 (revised in 1966) and 1998 (revised in 2004), prepared by Guangzhou Urban Planning and Design Survey Research Institute.

The typological process of residential buildings After examining the tissues, this research went a further step by zooming in the tissues and focusing on the building scale. This part of research focuses particularly on the traditional dwellings and their variants by distinguishing the characteristics and distribution of different building types and showing the diachronic development of the buildings from the period since late 19th century (Fig. 18). Zhutongwu is considered to be the foundation type in the Guangfunan area. Essentially the zhutongwu is comprised of a living room, bedrooms and a kitchen, in most cases, that are connected by a light-well. There are slightly differences in the floor plan of original zhutongwus based on the built lot sizes and location. Because of the pressure of land use, the most prevalent type of zhutongwus in the study area follows the Type-C in their floor plans. There are two synchronic variant type of zhutongwu which were found in the study area and each one played a role in the later transformation (Fig. 19).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

117

Figure 18. The classification of building types in the Guangfunan area. Source: Based on authorsâ&#x20AC;&#x2122; field survey.

Figure 19. The floor plan and faĂ§ade of original Zhutongwu and its synchronic variants.

The first variant is Zhutongwu-I, a small building that first emerged in the Former Watercourse Zone while the Daguan canal started to dry up. The original size of some built lots was only six to eight metres deep. The limitation of space triggered vertical extensions of the buildings in order to fit all basic

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

118

functions in small plots. By adding a small mezzanine floor, the Zhutongwu-I has a small interior stair that separates the bedroom and living room on different ground levels. Some of the Zhutongwu-I buildings are extended vertically in later transformations by adding another storey on top of the building. This original version of this sub-type barely exists in the Former Watercourse Zone nowadays, except in some minor back streets in the study area where a few small buildings remain in the field. The development of commercial activities triggered the need to acquire a more specialised space for the shopfront, and led to the appearance of the Zhutongwu-II. This variant retained a similar floor plan to that of zhutongwu but the interior space was reconfigured, then the commercial function was completely separated from the original building in the later transformation. After the Opium War, the western technique and architectural style became widespread within Guangzhou. The buildings started to syncretise western styles of decoration in the pediment and relief sculpture. In early 20th century, this variant type gradually turns into a new architecture form, qilou. In those deeper lots on the watercourses, buildings extended horizontally on the new gain land when the canal narrowed; the dried up area was replaced by courtyards, which later were replaced by a new building. These progressive transformations gradually led to a new iteration of zhutongwu, as a multifamily house. This transformation could also be found in those deep long lots in the Traditional Residential Dwelling Zone. One of the characteristics of most multi-family zhutongwus in the study area is a mixture of materials on the façades, since the internal configuration of the building had been adapted and redecorated several times during the process, leaving traces of different architectural styles in different periods (Fig. 20).

Figure 20. The floor plans and façades of multi-family Zhutongwu.

A sizeable number of apartment buildings that were constructed before 1949 still exist in the Guangfunan area. By comparing the floor plan with those single zhutongwus built in the 19th century, the successive process between zhutongwu and the apartment house is evident as they share considerable similarities in floor plan. The apartment house is regarded as a more economical land use version of zhutongwu as it extends the original house both horizontally and vertically to accommodate more inhabitants in a single apartment building (Fig. 21). In the later transformation, some of these apartment houses are “doubling” by combining two built lots together and constructing two apartment houses that share the interior stairway and light-wells (Fig. 22).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

119

Figure 21. The floor plans and faรงades of the Zhutongwu apartment.

Figure 22. The floor plans and faรงades of the Doubling Zhutongwu apartment. In previous research about traditional Cantonese dwellings, the origin of the Qilou is controversial, yet a consensus was reached that the Qilou is a colonial building type that mixes the characteristics of indigenous and Western architectural styles introduced into China from the 19th century onwards, especially after the Opium War in 1840 (Lu, 1990; Lin, 2000; Lu, 2004; Lin and Sun, 2004; Lin, 2006).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

120

However, despite the Western style faรงades, the floor plan and interior configuration of Qilou has considerable similarities to the sub-types Zhutongwu-II. This implies that the essential part of the Qilou is mostly inherited from the local commercial-residential building. Therefore, by looking beyond the architectural expression of the faรงade, there is evidence that Qilou is a newer architectural form derived from previous forms in a diachronic process and deeply ingrained in the local historical and geographical context (Fig. 23).

Figure 23. The floor plan and faรงades of local commercial-residential buildings and Qilou.

It is evident that there has been a typological process at play in the Guangfunan area from 19th century. The classification of the existing building types has provided a framework for understanding the diversity of the dwellings and settlements in the study area. The typology sheds light on the chronological development of residential forms by pointing to mostly spontaneous transformations that result in newer architectural forms derived from previous ones (Fig. 24). Doubling the built lots and conversion of the land use are two characteristics that are associated with transformation in the last century. The influence of exotic architectural style in the study area is another characteristic during the transformative process of buildings that denotes a cross-cultural interaction from the late 19th century onward.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

121

Figure 24. The interpretation of building typological process in the Guangfunan area.

Discussions and conclusion This study is one of the first applications of Italian process typology methods in China. It takes a step forward from previous typological study in Chinese traditional dwellings by applying the Italian approach of typology and urban tissue study in Chinese context. This research demonstrates that a predominantly European theory can be adapted to Chinese culture. It also demonstrates that the typological process of

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

122

Guangzhou can be analysed using Italian methods. It implies the possibility of expanding this type of research to other cities in China and demonstrates the possibility of adapting a predominately European research tradition to a Chinese context. The idea of the typological process sheds light on the continuity in the components of urban form during the historical development and unveils the significance of the intrinsic relations between the urban tissues and spatial configurations. Moreover, this analysis helps to better measure the extent of Western influence on the evolution of urban and architectural forms of Guangzhou in the 19th century. Such influences are more superficial than initially thought. Despite the stylistic expression of façades, the residential architecture system demonstrates that the evolution of architectural and urban tissue forms is deeply ingrained in the historical and geographical context from which they stem. By revealing the inherent connections between the building types, this analysis also highlights the structural permanence in the local residential building types, which is the essential element that confers a built environment with specific and recognisable identity. This is essential in each building type and remains the same despite inevitable changes of socio-economic context during the process. Although the continuity of the spontaneous building progress and urban tissues transformation in the study area have been interrupted by the socialist construction and reform movement after 1949, there are remaining elements in these traditional dwellings that retain some characteristics of their own identity. These show how the residential form has continuously engaged local conditions and inherited from the material culture. For instance, the small footsteps outside the dwellings in the Former Riverfront Zone, though it may only remain as a memorial structure now, denoted the existence of private docks that were one of the characteristics of waterfront commercial activity in the 15th century. These elements point to the heritage value of the inherited built environment in the study area, with pleas for their conservation in the practical planning management and policy-making. This research provides a new perspective on the issues of urban landscape management and historical conservation planning in China, especially in a context of rapid urban change in Chinese cities. First, at the street scale, maintaining a degree of uniformity in the aspect ratio during the process of widening the streets is important for the redevelopment; otherwise the street space will lose its identity even if the street line and building lots remain unchanged. Second, a successful building with an appropriate scale is one of the issues of crucial importance at the built-lot scale, as a large-scale redevelopment on an amalgamated block would create a dramatically different image and spatial pattern in urban tissues. Third, lessons can be learned from this urban experience as to how to conjugate building density and richness of the urban fabric in future. The conflict between the traditional and modern forms, especially in a historical district, is a complex challenge in current urban conservation and planning management. Consequently, in the formulation of planning policies that seek to balance inherited historical characteristics to be preserved or controlled during transformations and the new built environment emerging from the urban regeneration process, it is critical to formulate planning and design recommendations for the management of urban landscapes.

References Baker, N.J. and Slater, T. R. (1992) “Morphological regions in English medieval towns”. In Whitehand, J.W.R. and Larkham, P.J. (ed.). Urban landscape: international perspective (Routledge, London) 43-68. Caniggia, G. (1963) Lettura di una citta: Como (Centro Studi di Storia Urbanistica, Roma) (2nd edn, 1984, Edizioni New Press, Como). Caniggia, G. and Maffei, G.L. (2001 [1979]) Architectural composition and building typology: interpreting basic building. (S.J. Fraser, Trans.). (Alinea Editrice s.r. l, Firenze) Cataldi, G., Maffei, G. L. and Vaccaro, P. (2002) ‘Saverio Muratori and the Italian school of planning typology’, Urban Morphology, 6, 3-14. Chen, F. (2010) ‘Yige xinde yanjiu kuangjia: chengshi xingtai leixingxue zai zhongguo de yingyong’ (‘A research framework: applying typo-morphology in China’), Jianzhu Xuebao (Architectural Journal) No 4, 85-90. Compiling Committee for the Gazetteer of Liwan District, Guangzhou (1998) Guangzhoushi Liwan Quzhi (Liwan District Gazetteer, Guangzhou). Guangzhou: Guangdong Renmin Press.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

123

Corsini, M. G. (1997) Residential building types in Italy before 1930: the significance of local typological processes. Urban Morphology, 1, 34-48. First Historical Archives of China and Guangzhou Municipal Government (2002) Qinggong Guangzhou Shisanhang Dangan Jingxuan (Archives of Thirteen-Hong of Guangzhou in Qing Dynasty).Guangzhou: Guangdong Economic Press Gauthier, P. (2003) ‘Le tissu urbain comme forme culturelle: morphogenèse des faubourgs de Québec, pratiques de l’habiter, pratiques de mise en œuvre et représentations’, unpublished PhD dissertation, McGill University, Montreal. Gauthier, P. (2005) ‘Conceptualizing the social construction of urban and architectural forms through the typological process.’ Urban Morphology, 9, 83-93. Gu, K. (2001) ‘Urban morphology: an introduction and evaluation of theories and methods’, City Planning Review 25, 36-41. Gu, K., Tian Y., Whitehand, J.W.R. and Whitehand, S.M. (2008)‘Residential building types as an evolutionary process: the Guangzhou area, China’. Urban Morphology, 12, 97-115 Guangdong Research Institute of Urban and Rural Planning and Design and School of Architecture, Tsinghua University (2005) Guangzhou Lishi Wenhua Mingcheng Guihua (Conservation plan for the historico-cultural City of Guangzhou). Unpublished document Guangzhou Municipal Government (1996) Guangzhoushi Chengshi Zongti Guihua, 1991–2010 (The master plan for Guangzhou, 1991–2010). Unpublished document. Hofmeister, B. (2004) ‘The study of urban form in Germany’. Urban Morphology, 8, 3-12. Lin, C. (2000) ‘Qilouxing Jiewu de Fazhan yu Xingtai Yanjiu’ (‘Research on the development and morphology of qilou buildings’). Unpublished PhD dissertation, South China University of Technology, Guangzhou. Lin, L. (2006) Guangdong Qilou jianzhu de lishi yuanyuan tanxi (‘The historical origin of the Qilou buildings in Guangdong’). Jianzhu Kexue (Building Science), 22, 87-90. Lin, L. and Sun, Y. (2004) ‘Guangdong qilou de pingmian leixing ji kongjian fenbu tezheng’ (‘The plan types and the spatial characteristics of the qilou in Guangdong’). Nanfang Jianzhu (Southern Architecture), 27, 18-20. Lu, Y. (ed.) (2004) Zhongguo minju jianzhu (Chinese old dwellings). (South China University of Technology Press, Guangzhou) Lu, Y. (1990) Guangdong Minj (Dwellings of Canton). (Zhongguo Jianzhu Gongye Press, Beijng) Malfroy, S. (1998) ‘G. Caniggia: Ragionamenti di tipologia, (G.L. Maffei, ed.), Firenze 1997 [recension]’. Urban Morphology, 2, 116-117. Malfroy, S. (2001) Urban morphology and project consulting: a Berlin experience. Urban Morphology, 5(2), 63-80. Moudon, A.V. (1986) Built for change: neighborhood architecture in San Francisco. MA: The Massachusetts Institute of Technology Press. People’s Congress of Guangzhou (1999) Guangzhou Lishi Wenhua Mingcheng Baohu Tiaoli (Conservation regulations for the historico-cultural City of Guanzhou). Retrieved from: http://www.upo.gov.cn/pages/zwgk/fgzc/wwbh/2318.shtml (December 5th, 2012) Tian, Y., Gu, K. and Tao, W. (2010) ‘Urban morphology and conservation planning’, City Planning Review, 34, 21-26. Vance, J. E., (1977) This Scene of Man: Role and Structure of the City in the Geography of Western Civilization. (Harper Press, New York) Whitehand, J. W. R. and Gu, K. (2006) ‘Research on Chinese urban form: retrospect and prospect’, Progress in Human Geography, 30, 337-55. Whitehand, J. W. R. and Gu, K. (2007) ‘Extending the compass of plan analysis: a Chinese exploration’, Urban Morphology, 11, 91-109. Whitehand, J.W.R., Gu, K., Whitehand, S.M. and Zhang, J. ‘Urban morphology and conservation in China’, Cities, 28, 17185. Xu, Z. (2012) ‘From Alnwick to China, M. R. G. Conzen’s classic study in Chinese’, Urban Morphology, 12, 167-69. Yang, W., and Zhong, Z. (1996) The history of Guangzhou (Guangzhou jian shi). (Guangdong Renmin Press, Guangzhou) Zeng, Z. (1991) Guangzhou Lishi Dili (The history and geography of Guangzhou). (Guangdong Renmin Press, Guangzhou) Zhou, A., and Xiao, J. (ed.) (2003) Guangzhou Lishi Ditu Jingcui (A selection of historical maps of Guangzhou). (Zhongguo Dabaike Quanshu Press, Beijing)

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

124

Xi Guan: Edged community in the very heart of the city, the decline of Guangzhou’s traditional centre Tao Li Urban Regeneration Research Center, Zhubo Design Group, Guangzhou, China Email: li.tao@zhubo.com

Quanle Huang Modern Architecture Design and Research Center, South China University of Technology, Guangzhou, China Email: archi_yellow@hotmail.com

Abstract. This paper observes the impact of revolutions on the evolution of urban form. Through comparative analyses between cities in China and other countries, the research focuses mainly on the changes to property rights as one of the key elements that influences the transformation of built environment. Xi Guan area is the old centre of Guangzhou city and has experienced roleshifting from margin to centre in the past. It was re-marginalised after the Chinese revolutions in 1949, when private property rights were confiscated by the state. A series of subsequent changes led to the decline of the central position of Xi Guan area, and the decline of the built environment as well as the typical and traditional urban form. In Guangzhou, as well as big Chinese cities like Beijing, Shanghai and others, the decline of historical city centre was observed for similar reasons. Keywords: margin, centre, Xi Guan, revolution, property rights, social-spatial structure

“Margin” and “centre” are both relative notions that interact when they are used to describe areas in the city during its long evolution process. Most present central urban areas have historically been situated on the edge of earlier city territory. Yet, not all the marginal area can be turned into a future urban centre. Only those areas that are influenced or driven by particular social, economic or political forces will be able to play a more central role in the city’s development. Fujita et al. (2001, p. 384) proposes that once the concentration of population and/or production happens in a certain place, a self-organising force will also help to maintain this concentration so as to retain the position of being the centre. This explains how continuous urban development occurs. However, there are other cases that do not follow this process of urban evolution. When a formed urban centre is subjected to unexpected impact, such as social or political revolution, or, when a new government brings important changes to the property rights of land and real estate, the city centre might be transformed radically both in its social structure and in its spatial structure. It can lose its position of being the centre of the city. From the 18th to the 20th century, worldwide “revolutions” have brought important changes in urban form in cities, though the effects are different on traditional city centres in various countries and places. France began the revolutionary sequence, yet real estate property rights did not change much during the French Revolution. When Haussmann organised the grand transformation of Paris city from 1851, he showed respect to real estate property rights and did not modify the existing system. However, in Russia, when the October Revolution took place in 1917, the original private property rights of land and real estate was converted to the country thorough the reform of nationalisation (Tarkhanov et al., 1992). After the revolution, in the urban area of former Soviet Union and Eastern European socialist countries, many houses owned by the city’s former elite class were taken over and occupied by another elite class of the

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

125

new regimes. The owners of the real estate changed names, but they were from the social level similar with the ex-proprietors. In this way, the original urban centre continues to develop and its central position is relatively well-preserved by the new regimes. When the Turkish War of independence came to its end in 1922, the national capital was moved from Istanbul to Ankara. Beyoglu District, which used to be the heart of Istanbul, experienced a radical roleshifting from the dynamic centre of international trade and diplomatic body of the country to an empty district. The houses that used to belong to bourgeois households were gradually occupied by poor foreign immigrants. Although the house owners left, the house property rights did not change greatly. For this reason the urban form remains unchanged, although today’s Beyoglu no longer serves as the central part of the city. This paper observes the urban transformation of Xi Guan area in Guangzhou city, which was formerly in the heart of the city since the 18th century. It faced great changes in property rights resulting from a series of revolutions in the 1950s, and has experienced changes in its position during the process of centralisation and marginalisation in the city. This research is based on historical background of the Chinese Revolution, and attempts to explain – from the perspective of the Chinese revolution – important differences and imbalance in urban form among diverse Chinese cities. The abrupt changes of building functions and property rights in Xi Guan area have led to great changes of urban form in the city.

Xi Guan: from marginal suburb to city centre The name “Xi Guan” means the west gate of the city wall. Before the 18 th century, Xi Guan was a marginal suburb out of Guangzhou city. Compared with the other three gates of the city wall – the east, south and north gates – Xi Guan has more advantages due to its geographical condition. It is situated in the largest plain area of region, with a network of creeks and small canals well connected with the Pearl River, where a deepwater port allows sea craft to dock. With efficient accessibility via the water network and other advantages, Xi Guan area is a preferred location for further urban development. Guangzhou has been, for a very long period of time, the most convenient international port for China. Commercial activities have always been a key element in Guangzhou city’s urban history. Every year, during spring and autumn, international merchants came to Guangzhou for trade. They were not allowed to stay in the city and could only live in the west suburb, the Xi Guan area. With more newcomers – such as merchants, entrepreneurs and refined scholars etc – living and making business in the city, Xi Guan area obtained a more central position in the city (Table 1). At the end of the Qing Dynasty, the Xi Guan area has become a centre of the city, demonstrating strong accumulating capacity both for population and for urban functions. In the district of Thirteen Hang commercial and business has accumulated foreign merchants with special permission for management from the central government. The surrounding districts also demonstrated a great variety of urban dynamics: efficient water transport gathered along the Pearl River, and diverse commercial facilities, such as Chinese private banks, money shops, jewelry and gold shops, fabric shops, grain shops, herbalists, and general merchandise, developed along roads and streets (Fig. 1). The interaction between urban development and trade activities during this period reflects the trend of internationalisation. Following the two Opium Wars in 1840 and 185, between China and the United Kingdom, foreign countries have obtained the right to inhabit Xi Guan area and have then established their concession by the Pearl River. Since then, foreign cultures and international commerce have rapidly accelerated Xi Guan area’s urban development. Merchants from all over China relocated to open their stores, purchase houses and settle down in Xi Guan. A new class of urban elite came to live or work in Xi Guan in which they brought a new living style. New-type department stores, modern commercial modes, cinemas and other cultural and amusement places, tennis courts and other sports facilities gradually appeared in this area. A modern city form gradually came into being (Fig. 2).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

126

Ming Dynasty (1368-1644)

Table 1. A brief history of Xi Guan’s urbanisation process New comer Main Built element characteristic function Arabs merchants, Maritime commerce “Fan Fang” for Max. to 120,000 inhabit the Arabs population Local farmers Agriculture; Gardens Fields, gardens, temples Workers and plant Manufacturing Processing plants of managers, industry porcelain, kiln, and Max. to 40,000 embroidery workers garments. Tribute missions International trades “Huai Yuan Yi” for foreigners

Qing Dynasty (1644-1912)

Foreign and national merchants

Period Tang Dynasty (618-907) Song Dynasty (960-1279) Yuan Dynasty (1271-1368)

All kinds of commercial activities

“Thirteen Hang” for foreign trades; silk mills and all kinds of small commodities, services and houses

Spatial characteristic Single-functional street in suburb Mixture of suburb and town landscape Rudimentary stage for industry and commerce developments Particular quarter in town City form came into being

Figure 1. Xi Guan area, from suburban area to high density urban area.

For Xi Guan area, the process of becoming the city centre was paralleled to the history of Guangzhou city’s becoming China’s most important city of international trade. During the regime of the Republic of China (1912-1949), Guangzhou’s city wall was pulled down for urban expansion. This included the construction of new roads were constructed with commercial streets with arcades built along these roads. Residential houses followed the commercial growth and were divided into two categories. The first is called the “bamboo house” which is of traditional Cantonese style. It is built on a small piece of residential land with short width and long depth, with a narrow façade directly facing to the street (Fig. 3). This kind of plan with a long “bamboo” shape is typical in southern China. However, it is rather different from that of the northern Chinese house which features the form of courtyard with enclosed fences. The second residential category is an occidental residential style which was introduced by foreign merchants as a kind of colonial house with veranda.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

127

Figure 2. Urban life and culture in Xi Guan before 1949.

Figure 3. “Bamboo House” in Xi Guan area, the space organisation adapts to Guangzhou’s climate condition.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

128

Xi Guan under Revolution After the civil war, the People’s Republic of China (PRC) replaced the Republic of China in 1949, and the socialist system became the new regime of the country. Many of the middle class elite in Xi Guan, who preferred the previous regime, escaped the city and left for Hong Kong, Taiwan, Southeast Asia, as well as Australia and the United States. Their houses were handed over to their servants or relatives or friends, who then took over the buildings and lived inside. During the 1950s, the Communist Party, as the party in power, has carried out a series of high speed processes of industrialisation and urbanisation. During the same period, the Korean War broke out. China turned its investment priority towards industrial construction and wartime military industry, while the financial support of large-scale housing development was left behind. As a large number of rural populations migrated to big cities like Beijing, Shanghai, and Guangzhou, the city governments were not able to provide housing. There were critical “housing shortages” in almost all the big cities in China. The state then determined to implement “socialist reform” following the example of Soviet model of that time. By transforming the nature of property rights – from private to public – of the housing once owned by the middle class elite, mainly in Xi Guan area, the government has “obtained” extra space for housing the growing population without paying for apartment construction. However, this transformation of property rights have radically changed the original relationship between people and space. With the departure of the middle-class and elite, the position of Xi Guan as the centre of city began to decline. First, when the government confiscated a private house, only a certain proportion of the whole property was reserved for the house’s original owner. In general, that allocation measured about 50 to 100m² per household, no matter how big the house was. The housing management sectors of the government took the remainder and distributed it to workers or employees who applied for an apartment. The former owners were given very little compensation (Secretariat of Central Committee of the Communist Party of China, 1955). In 1964, according to documents of National Supreme People’s Court (1964), the private ownership of building property rights was denied, and complete nationalisation by the national government of the original private building property rights was confirmed. Eventually, private houses were all turned into national ownership in order to be uniformly arranged to resolve the housing shortage problem. Even as significant economic reform has been carried out since 1978, the government confirmed its justification of the nationalisation of prior privately owned houses (China National Urban and Rural Construction Environment Protection Ministry). The authorities insisted that “joint state-private enterprises should be transformed into state-owned enterprises and all the capitalists’ properties should be cleared up” (Central Committee of CPC), and that “all the private rent houses should conform to policy provisions of governments at all levels, they all belong to the nation and are uniformly managed by the state’s real estate management sectors.” With this strong legal support, most of the city governments refused requests from house owners to have their house property rights returned.

Changing process of Xi Guan from “centre” to “margin” Within this context of property rights conflicts, the urban form of Xi Guan area underwent a series of changes together with the changes of the community’s social structure. With middle-class elites leaving, the position of Xi Guan as the centre of city began to decline. As a result, Xi Guan area gradually lost its position of being the important centre for the city, both geographically and functionally. This was manifest in five ways. i)

Change of residential housing function: from living style of single household to multi-family dormitory Prior to 1949, and the transition to a new regime and new institutions that controlled cities, the urban house units in Xi Guan area were mainly for private family living function, and were

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

129

consistent with the modern living style. There were mainly two types of architecture: “bamboo house” and multi-storey apartment inhabitant. After 1949, in order to solve the problem of housing shortages, the authorities took over these single family houses and divided the house space into many parts. These were rented at a very low price to many citizen families, who then crowded into these single family houses and changed their spatial status, as well as the original living style and culture they used to present (Fig. 4).

Figure 4. The transformation of the social-spatial structure of houses in Xi Guan Source: Analyses and photos by author. .

ii)

Change of residential house function has led to change of internal form of the building Since the original family unit residential houses were changed into multi-family dormitories, the traditional residential model -- that was the reason the house had been built in such a way – was destroyed. The correspondence of “space and people”, the consistency of “one household – one residential house” no longer worked. There was quite limited space for those families that moved into the crowded conditions of these houses. In order to better use the limited space, every family carried out internal space re-configuration and extension as much as possible, including occupying the public corridors, by enclosing the balconies so as to transform them into bedrooms, and by adding rooms on the roofs of the house. From inside to outside, from parts to the whole, the traditional house building units were radically transformed. This self-transformation by the multifamily inhabitants has led to a more important change of the buildings’ outlook of the overall street façade as well as the general urban form.

iii)

Change of building property rights has led to damage and decline of building maintenance When the traditional houses were built in Xi Guan area, their owners, most being middle class elites or entrepreneurs, held these houses as long-term property to be inherited. Large sums were paid for the initial construction, then ongoing maintenance of these houses was carried out carefully. However, when these private houses were taken over by the authorities via the nationalisation policy in the 1950s and 1960s, they became an important component of the social housing system

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

130

to accommodate the increasing urban population. The change of building property rights from private to state-owned has resulted in drastic change in the spatial quality of these houses. Although the government provided some maintenance for these residential houses, the selftransformation by the new low-income inhabitants has resulted in serious damage and severe decline of the buildings and their environment (Fig. 5).

Figure 5. Decline of built environment in Xi Guan area.

iv)

Damage and decline of building has led to destruction of building and transformation of plots: from “small and open” to “big and enclose”. In Xi Guan area, most of the plots for house construction were slim, around 4m wide and 16-24m long. These pieces of land conformed to Xi Guan’s typical traditional architectural mode, which had followed the commercial principle. During the 1960s and 1980s, with the nationalisation of plot property rights, the private land plots were also confiscated by the states. Since the original buildings were seriously damaged and the cost for maintenance was increasing, the government decided to pull down these traditional two to three storey houses to integrate the small pieces of plots. Five to six storey collective apartments were built on larger pieces of land in order to meet the new space requirement.

Figure 6. Newly inserted buildings change the spatial relationship of the neighbourhood.

The integration of small plots into a bigger plot has resulted in additional changes of urban form. Construction on the newly integrated plot often presents as new, big volumes growing up from the original urban tissue, although it is mostly enclosed by fences. Within the fence, the most standard way to finish the building’s surrounding space is to plant a greenbelt (Fig. 6). This form of new collective apartment shows the attitude of estrangement against the “old neighbours” who still live in the traditional houses.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

131

During high speed urbanisation period, house marketisation policy has led to more and more oversize and over high reconstruction. After 1978, the Chinese government carried out important economic reform and then implemented large-scale construction of residential house. The National Ministry of Construction issued “Opinion about further implementing private house policy and processing opinion about left issues of socialist transformation to urban private rent houses” in 1982 and 1985. The nationalisation of property rights of private residential house in the 1950s remains unchanged. As those nationalised former private houses been overused for 50 years, from 1949 to 1999, they were already suffering from very poor maintenance and a humble spatial condition. These residential houses, built before 1949, which had represented the lifestyle of the former middle class elite with their materialised presentation for traditional culture and living style, have suffered from dramatic and radical changes ever since the revolution in 1949 (Fig. 7).

Figure 7. Today’s Xi Guan area, a “broken heart” in the historical city centre.

Large-scale demolitions were increasingly carried out in Xi Guan Area. To make way for new and major real estate development, more traditional houses were pulled down, and their small pieces of land were integrated into a much bigger plot, from 36m by 50m, to 40m by 60m, even to 80m by 120m. The height of new towers also increased radically, from 15 storeys to 33 storeys. This means that the traditional skyline height of less than 10m was lifted to more than 100m. The original and traditional plot size and street texture were thoroughly changed (Fig. 8).

Figure 8. New skyline in Xi Guan area during the 2000s.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

132

Xi Guan becomes the margin of newer city centre after economic reform Since 1949, Xi Guan area has witnessed the breakdown of harmony in its social-spatial structure. Xi Guan lost its urban function as the former international trade centre. The revolutions have thoroughly destroyed the middle class elite who had supported the social-spatial form of Xi Guan in earlier times. In 1986, the 6th National Games were held in Guangzhou. The main urban development was heading eastwards in the newly created Tianhe District in favour of the National Games, leaving Xi Guan behind in the west of Guangzhou city. Being further away from the living centre of the urban elite class, Xi Guan is being increasingly marginalised. This result is reflected not only in the spatial quality and urban form, but also in population aging, impoverishment, decrease of employment opportunity and other more severe social problems.

Conclusion In the 20th century, grand revolutions or reforms took place in many countries such as Russia, France and Turkey. In former Soviet Union and Eastern European countries, revolutions often resulted in severe social transformation including nationalisation of house property rights and transformation of the urban elite class. The emergence of a new elite class, replacing the previous elite class, filled in the traditional city centre spaces and somehow helped to preserve the urban form. Even though the regime changed, the original places retained their central status without drastic marginalisation. In this way, the relationship between the so-called space of elite and community of elite maintains a mutual correspondence even in the new era of regime.

Figure 9. Low rise traditional urban fabric mixed with newly inserted high rise towers in Chinese metropolis as typical urban form of today.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

133

Comparative studies of the influence of worldwide revolutions on urban form shows that the Chinese revolution is among those that have led to important in urban development. The escape of the original urban middle class elites away from their property and the nationalisation of private house property rights resulted in drastic changes of urban form in Chinese cities. Such drastic changes appeared not only in Xi Guan area of Guangzhou city, but also in almost all the large Chinese cities, including Beijing, Shanghai, Nanjing, Wuhan and Tianjin (Fig. 9). Revolutions that brought sudden change to social structure have eventually brought radical changes of urban form in cities.

References M. Fujita, P. Krugman, P. & Venables, A.J. (2001) The spatial economy: cities, regions and international trade, (MIT Press, Cambridge). Tarkhanov, A., Kavtaradze, S., and Anikst, M. (1992) Architecture of the Stalin Era (Rizzoli, New York). Second Office of the Secretariat of Central Committee of the Communist Party of China (1955) “Elementary Situation about Current Urban Private Property and Opinions to Carry out Socialist Transformation” (December 16, 1955) (China) Supreme People’s Court (1964) “Reply on the Actual Lost Ownership of the Owners of National Rented Houses” (China). China National Urban and Rural Construction Environment Protection Ministry (1982, 1985) “Opinion about further implementing private house policy and processing opinion about remaining problems of socialist transformation to urban private rent houses” (China). Central Committee of CPC, No. 507 Document. Item 2 (China).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

134

The compact city neighbourhood: an emerging new stereotype and model to redevelop the industrial fringe of the historical European city and develop new sustainable suburbs Todor Stojanovski Urban and Regional Studies, KTH Royal Institute of Technology, Stockholm, Sweden Email: todor@kth.se

Abstract. The compact city surfaced in the European sustainable cities debate in the 1990s and since then many experiments have occurred with new sustainable suburbs based on a recognisable urban pattern and character. This has emerged as a new European stereotype and model for urban redevelopment of abandoned industrial zones. A similar concept for dense and walkable neighbourhoods oriented to public transportation, Transit-Oriented Development (TOD), is promoted in the USA, but in a different context. Transmissions and convergences of urban images happen continuously throughout this history. Paris with its lavish cityscapes inspired urban reflections throughout Europe in the end of the 19th century. With mainstreaming modernist movement in architecture and urbanism in the 20th century, urban replications appeared all around the world almost instantly. A similar phenomenon is happening with the compact city model today. Do we need more replications of urban models in Europe? How can this stereotype be broken? Keywords: compact city, neighbourhood, stereotype, replication, urban form, transit-oriented development (TOD)

Postmodernism refers both to a style in architecture and urbanism and to the economic and social conditions after the modernist era in the mid-20th century. In this ongoing condition, many European cities are experiencing deindustrialisation and change towards knowledge-based economies and postindustrial society (Bell, 1973). With accumulation of global capital in creative businesses and startups there is a new demand for historicity and innovative neighbourhoods that are sustainable. The traditional urban cores of the larger European cities are becoming increasingly popular. Many developers focus on abandoned industrial zones, which are the first fringe belt around the historical urban cores from the industrialisation period in the 19th century (Whitehand, 1967). These historical urban cores are extended in large European cities like London, Stockholm, Helsinki, Amsterdam or Copenhagen by replicating a recognisable urban pattern which has emerged as a new European stereotype and model for urban redevelopment of abandoned industrial zones. The same model is also applied to development of new sustainable suburbs on greenfields in Stockholm, Copenhagen, Amsterdam, Eindhoven, Cambridge and Paris. Hammarby Sjรถstad, a Swedish prototype for a sustainable neighbourhood in Stockholm, is analysed in this paper and compared with similar neighbourhoods in the UK, Finland, Denmark and The Netherlands. The history of the compact city is described together with discussion of the transmission, similarities and differences of the compact city model in different cities. The paper also addresses sustainability from the perspective of transportation and accessibility, since mobility is becoming fundamental in urban life. How

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

135

sustainable are these neighbourhoods in relation to mobility? Do we need more replication of urban models in Europe? How can the stereotype be broken?

Compact city in Europe There are two models for European sustainable cities which can be traced in documents, papers, reports and responses on sustainable cities and mobility published by the European Commission (EC) from the 1990. The first is the compact city model and the second is the green city model. Compact city emphasises intensity through urban regeneration, densification and effective public transportation systems. Urban expansion is halted to protect the natural environment by creating a strong contrast between city and nature. The green city model is based on ecological design and natural cycles, propinquity and conservation. It presupposes creation or transformation into communities with lifestyle adapted to the natural rhythms of the surroundings (EC, 1998, pp. 6-7). While the compact city concept exists in the book, The Compact City, which offered a vision of a highly technological city with temporally and spatially organised urban activities (Dantzig and Saaty, 1973), there is no reference to this book in the EC documents. It is presented more as a critique of the Modernist movement, the sprawling and uniformity of modernist suburbs and edge cities and their mono-functionality. Many architects like Severio Muratori, Aldo Rossi, Robert Venturi, Paolo Portoghesi, Leon and Rob Krier also criticised the Modernist movement and advocated for new historicism and return to the traditional ‘cities with quartier’ (Krier, 1984). The compact city favours architectural heritage, by respecting rather than imitating the old, greater diversity by mixed uses and solving urban problems within existing boundaries of the city (EC, 1990, p. 30). It depicts the traditional European city, but draws its inspiration from high-tech architecture and new technologies. It is both neo-traditional and neo-modern. The neo-traditional tendency is also shared with American New Urbanism where Transit-Oriented Development (TOD) is promoted as a concept for dense and walkable neighbourhoods oriented to public transportation (Calthorpe, 1993). This concept shares much with European compact city, though in a different context. Compact city policies are not new in Europe, and they are prevalent in northern and western European countries including Benelux countries, Denmark, Sweden, Germany and France. Strict urban boundaries of growth were proposed to protect and manage agricultural land. Much emphasis is put on polycentric cities, urban density and ‘centralised decentralisation’. Patterns of polycentric cities with satellite suburbs (Stockholm) or suburbs in corridors as hand and fingers (Copenhagen) inspired by a diagram presented in Ebenezer Howard’s 1898 book Garden Cities of To-morrow. This had a profound effect on European urban planning in the mid-20th century via Lewis Mumford. The question of density has dominated Europe and Sweden (Rådberg, 1988) for more than a century given the 1912 publication of Raymond Unwin’s book Nothing gained by overcrowding. However, from the 1990s, the discussions changed from optimal urban densities to policies for densification and more dense and mixed urban areas. The compact city policies in UK were conceived as urban intensification and defined through changes in form and activity (Jenks and Gerhard, 2000) as increased use of the buildings. Changes of urban function lead to increased urban activity, as well as increased numbers of people living in, working in or traveling through an urban area (Williams et al., 1996). The experiences in the UK are divided. There is advocacy for an urban renaissance through more dense and diverse cities, but there are also British researchers who criticised the intensification concept. Research has shown that intensification does not always contribute to improvements (Jenks and Gerhard, 2000). Another argument proposes that the attachment to the car is too strong and causes congestion. The costs of congestion decrease the benefits of intensification (Melia et al., 2011). In Sweden the densification advocacy coincided with the emergence of a typological understanding of urbanisation. The Swedish neighbourhoods were classified in a typology that reflected historical urbanisation and captured the character of urban form and urban function (Rådberg, 1988). Densification was understood as policy of adding new buildings and increasing the number of inhabitants or activities in one urban area while retaining the character of the neighbourhood. It started with an experiment, Hammarbyhöjden, which is an early modern suburb in Stockholm that experienced densification in the 1980s. This was achieved through infill buildings that were carefully designed to complement the character of the neighbourhood. The project was a success

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

136

and densification became a new policy in the City of Stockholm (Stockholms Stad) which adopted the compact city motto “Build the city inward!” in 1999. The dense and mixed city has been a popular vision and goal in many Swedish towns and cities since the end of the 1990s. The compact city is researched from the perspective of sustainable mobility and transportation. Empirical research in The Netherlands and Denmark shows that people travel shorter distances in more dense areas and that car use is lower than in the less dense areas (Susilo and Stead, 2009; Naess, 1995, 2006). The compact city model in Hammarby Sjöstad Hammarby Sjöstad is a neighbourhood under development and located on the edge of the historical urban core of Stockholm. It is a redevelopment project for an abandoned industrial zone and regarded as an extension of the historical urban core, although it is also divided by a water channel. The neighbourhood unfolds along a wide multimodal boulevard dominated by the Tvärbana, a Light Rail Transit (LRT) system. The boulevard includes bike lanes, sidewalks along attractive façades with storefronts and ribbons of greenery and landscaping to beautify the cityscape and segregate the LRT line. The urban design of Hammarby Sjöstad revives the image of the industrial European urban cores with wider blocks and partial enclosure to allow prospects and circulation of air. Through high-tech architecture and use of environmentally friendly technologies, like advanced materials for insulation and other methods, to improve the energy efficiency of the buildings or advanced waste management and recycling (SymbioCity concept promoted by Sweco). The urban pattern is both novel and eclectic. Hightech and neo-traditionalism coexist within a postmodernist mix. Different elements draw inspiration from the traditional city prior to industrialisation, the industrial city of the 19th century and the modern city, but the mix is original (Fig. 1).

Figure 1. The urban pattern and inspiration of the compact city model in Stockholm. The traditional urban cores are from Stockholm and Norrtälje, the industrial cores are from Norrköping and Stockholm, the modernist suburbs are from Stockholm and Karlstad and the postmodernist suburbs are both located in Stockholm.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

137

Images of the traditional city emerge in Hammarby Sjöstad firstly along the multimodal boulevard as historically proportioned façades with storefronts, streets and sidewalks oriented towards the pedestrian, and the streetcar. The boulevard is public space and main axis of the neighbourhood. The courtyards are reminiscent of historical cities that unveil semi-public spaces in the urban tissue. The modernist city exists though high-tech architecture and urban landscape design. The apartment blocks are not fully enclosed in a perimeter. The partial segregation of the railways and busways on the avenue and the network of streets and cul-de-sacs from the avenue causes a super-block effect over the traditional urban blocks. It is not a superblock in nature as in the modern suburbs, but it is superimposed over a traditional urban environment. The tendency in the European traditional city is to interconnect all the streets and interweave squares as focal points in the network. The popularity and advertising (including Sweco’s SymbioCity campaign) of ‘sjöstad-s’ (translated as waterfront cities) resulted in a number of them emerging as fragments in different cities in Sweden and elsewhere. In many northern and western European countries many similar neighbourhoods were developed on abandoned industrial brownfields with a similar model of multimodal boulevards dominated by LRT or Bus Rapid Transit (BRT). These BRT systems are sometimes referred as Buses with High Level (quality) of Service (BHQS). The problem from mobility perspective is that these public transportation systems cannot compete with the car (their operational speed is often too low). Sometimes, as in Hammarby Sjöstad, the LRT system is orbital and does not connect directly to the urban core. The passengers need to make an additional transfer to the regional system. How does the public transportation work in this compact city model? Does this model contribute to more sustainable mobility? What is the prospect of replicating these neighbourhoods if there is a ‘mobility problem’? How can the stereotype be broken to change the model and solve this fundamental urban problem?

Methods The urban challenge in sustainable development is to fit cities into the pattern of oceans, land and clouds on the planet. It mixes physical aspects – impacted by finite natural resources and environmental degradation due the human consumption – with social cohesion, justice and economic viability and growth (WCED, 1987). In a context of cities, sustainability often tangles compactness and density. The compact city is a concept for European sustainable city promoted by the EC where the argument is that urban density and diversity are more likely to result in people living close to work places and services that are required for everyday life (EC, 1990, p. 40). Cities today are conceived as “extraordinary agglomerations of flows”, not only of people on the move, but as other forms of mobility like flows of information, capital, values, norms and lifestyles (Ash and Thrift, 2002, p. 42). Density per se is inelastic and does not always contribute to less travel (Ewing and Cervero, 2010). However the objective is also not to travel less. The sustainable transportation challenge in European cities is to break the dependence on oil without compromising mobility (EC, 2011, p. 5) by creating integrated, multimodal transportation systems that fully exploit the potential of public transportation (EC, 1997, pp. 11-12). The emphasis is on more energy efficient transportation modes where greater numbers of passengers are carried jointly to their destination by the most efficient (combination of) modes (EC, 2011, p. 5). This research focuses on urban form, or elements in the urban pattern of the compact city neighbourhood, and public transportation use as a measurement of sustainable mobility in the cities. Urban morphology focuses on the physical form of cities and their evolution from formation to subsequent transformations by identifying and dissecting its various elements (Moudon, 1997, p. 3), or recognising and abstracting urban forms and patterns (Marshall and Çalişkan, 2011, p. 421). Urban form is traditionally defined through pattern and process (Kostof, 2005), a historical layering and change of urban arrangement. The urban society, with its social structure, economy and culture, is an agent of conservation or transformation (Conzen and Conzen, 2004, p. 275). The urban form in the Conzenian school (Whitehand, 2001) is investigated as a historical change of patterns of: 1) streets, 2) lots and their aggregation in blocks, 3) buildings; and 4) land use (Conzen, 1960; Birkhamshaw and Whitehand, 2012). The street frontage is an important element the Muratorian School of urban morphology (Cataldi, 2002).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

138

Severio Muratori and later Gianfranco Caniggia were Italian architects who analysed urban form though building typologies. The urban form in the Muratorian School is seen from an architectural perspective as relationship of the buildings and plots their alignment to the street as centre of an urban block (Caniggia and Maffei, 2001). The urban block is defined by 1) street or urban space, 2) street frontages or façades of buildings, and 3) lots. Urban spaces become public, private, semi-private or semi-public realm depending on the types of buildings, orientation and character (private or public) of their façades and enclosure or openness of the streets, squares or courtyards (Krier, 2009, p. 159). Both schools of urban morphology (Conzenian and Muratorian) address the urban spaces from two perspectives: as urban plan (top perspective) and as urban space (within perspective, seeing from the street). These perspectives (Table 1) can be further expanded in analyses of urban flows within a theoretical framework where the physical form of cities is conceived through flow system and adapted spaces (Lynch and Rodwin, 1958; McLoughlin, 1969). Humans have two principal modes of experiencing urban space: 1) sensing urban space; and 2) being in a flow. Being in a flow mode is dominant in vehicles. For example, driving triggers a flow mode of thinking (Csikszentmihalyi, 1990). Drivers neglect the urban environment to focus on traffic flow, signs, turns and stops. The sensing space mode is more active during walking, but sometimes crowds of people evoke the flow mode as a consequence of crowding (‘penguin effect’). The sensing space mode also awakens when there are stops in transportation, like traffic jams or during waiting in public transportation. Table 1. Methods and perspectives on urban form and flow

Top perspective

Within perspective Transitions from top to within perspective

Urban form Urban flow Urban patterns (arrangement of Travel patterns (networks, flows and urban spaces; streets, lots, buildings, matrices between urban spaces) functions or land-uses; public and private spaces etc.) Cityscapes (sensing and orientation in Cityscapes or void (being in a flow) urban space) Exits and entrances between urban spaces (shift from sensing urban space to being in a flow)

The analysis of urban pattern of Hammarby Sjöstad focuses on architectural styles and arrangements of buildings, public spaces and private spaces. Arrangements of buildings are analysed in comparison to other typical neighbourhoods from Swedish cities. The analysis of the urban pattern of private and public spaces is inspired by the writings of Muratorian School via Aldo Rossi (1984) and Rob and Leon Krier (1979, 1984). The morphological character is conceived in the sense of Christopher Alexander (1979, p. 90) who also argues that the urban patterns are dynamic complexes of underlying elements and relationships between elements. Elements from the Conzenian and Muratorian School are joined for the analysis: 1) street or urban space between buildings, 2) lot, 3) street frontage/building façade, 4) building, and 5) building utilisation (function or openness to public/private). A ‘within perspective method’ from the British Townscape is used to analyse the urban pattern of public and private spaces. It includes serial vision, as a sequence of photographs along vantage points in urban space (Cullen, 1961) along important directions and spaces (main street, side streets, courtyards, etc.) that give the morphological character of Hammarby Sjöstad. The walkthroughs are along: 1) the avenue, 2) the waterfront promenade, 3) the elongated park, 4) one of the courtyards, 5) a secondary street that intersect the avenue and where car traffic is allowed on the intersection, and 6) a secondary street where the intersection is closed for cars. The walkthroughs are shown on the map below (Fig. 2). The Modernist city was envisioned and realised as a system of urban spaces that are interconnected by motorised transportation systems (Le Corbusier, 1929/1987). As a consequence of a half a century of modernisation of the European cities, the urban life is mobile and dependent on travel to work and experience attractive ‘third places’. With decentralisation of jobs and residences in the Swedish cities the work place and ‘third places’ are also dispersed in the urban regions. Public transportation is the only

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

139

Figure 2. Map of the walkthroughs in Hammarby Sjöstad

mode competitive to the private car for long journeys. Urban mobility is more efficient if the share of public transportation is larger in one urban region. The underlying assumption (as invariants of human mobility) is that an average person makes 1000 annual journeys, or slightly less than three personal journeys every day, and spends an hour traveling each day (Zahavi, 1974; Marchetti, 1994; Banister, 2011). There is much diversity in the 1000 journey invariant. The young and older travel less, whereas persons in middle age are more mobile (White, 2009). In the highly motorised countries with sprawled cities, such as as the USA, the number of personal journeys is higher. This is the result of a century long decentralisation of residences and jobs in American cities. An average American makes four personal journeys every day (NHTS, 2011), but the principle of aggregated studies is the same. Travel is a fixed sum game where different modes compete for a limited number of journeys (1000 or up to 1500 in car dependent cities) and higher share in the modal split. Every journey by private car means one journey less by walking, cycling and public transportation. With increased car ownership and decentralisation of residences and jobs, the number of journeys tends to increase. This number can be seen also as a variable from aggregated number of inhabitants and work places, since the urban activities depend on work and urban activity that is captured by this measure. There are neighbourhoods without inhabitants that generate high volumes of traffic. The generation of trips varies even more for different work places. The distances traveled by different transportation modes vary also. The 1000 journeys by inhabitants per year translate to roughly 650 journeys per resident and job. The method includes analysis of urban pattern of Hammarby Sjöstad, the Swedish prototype for sustainable compact city neighbourhood and the travel pattern (annual use of public transportation in respect to the 1000 journey invariant that works for Sweden). The number of annual journeys by public transportation is a simple measure for more sustainable mobility. The trips by private car are in average four times longer than by public transportation in Sweden (Trafikanalys, 2012). The number of annual journeys is calculated by multiplying the average number of passengers boarding daily at the stations in Hammarby Sjöstad and dividing it with the number of residents and jobs. In the centre of each area there is a LRT station with the same name. In Luma there are two stations Luma and Sickla Kej and the numbers of passengers boarding at the two stations were tallied. The population data comes from Statistik om Stockholm, Statistics Sweden (SCB) and Sweco, whereas the travel statistics are sourced from the Swedish Agency for Transport Analysis (Trafikanalys) and Stockholm’s Public Transportation Authority (SL). There was no data collection specifically for this research and there are some uncertainties about passenger data from automatic counts in vehicles (SL’s data collection system). But the alternatives like travel surveys and passenger counts on buses are time consuming and expensive alternatives. The statistical data and automatic passenger counts suffice.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

140

Urban analysis of Hammarby Sjöstad The urban pattern of Hammarby Sjöstad can be divided into different elements. The underlying element in the pattern is the street. The street layout is hierarchical. The main street (the boulevard) and waterfront promenade act as edges in urban space. They cut off super blocks with secondary side streets (often cul-de-sacs). The existence of the superblock is also justified by lack of clear plot division. There are a few large lots in each super block that are developed as single complex urban spaces with buildings, side streets, park and courtyards. The buildings are arranged as quasi-enclosed blocks with partially open courtyards. Within the new urban superblock there are a wide range of public, semi-public, semi-private and private places that evolve in cityscapes and evoke different experiences. The edges between and the entrances and exits to and from these places are clearly defined by urban design or landscaping. There are three types of public spaces: the boulevard or alle in Swedish, the main square, and the waterfront promenade. The boulevard is a main transportation axis and a shopping strip that extends to two thirds of the length of the neighbourhood. The main square is monumental and a stage for public events. The car traffic follows the façade, whereas the railway crosses the square. The waterfront promenade is a public space along the waterfront where limited car traffic is allowed. The semi-public place is an elongated park in the centre of the super block. The park is isolated by buildings or landscaping and not completely visible to the pedestrians on the boulevard. The side streets along the boulevard are also semi-public. They are open and visible, but not welcoming (with no storefronts, only building entrances, and many parked cars). The semi-private places are within the courtyards shaped by the buildings arranged in quasi-enclosed urban blocks. These urban spaces are accessible by narrow arcades and passages in the landscaping and are almost completely isolated, but visible to the residents (Fig. 3).

Figure 3. Analysis of the gradation between public and private urban spaces in the Hammarby Sjöstad’s model. The public spaces, the boulevard and the waterfront promenade, are circulated by pedestrians and cars. With urban design measures (no storefronts, blocked vistas) the privacy of space increased to the courtyards that are semi-private.

To experience the architecture styles, the gradation between public and private spaces and character of different urban spaces is shown on the walkthroughs along the avenue. The façades along the boulevard are modern, reminiscent of the shopping streets and boulevards in the historical urban cores of the 19 th century. The serial vision reveals two types of urban spaces on the boulevard: the railway is a very empty strip of transportation infrastructure (Fig. 4), and the sidewalks are rich with experiences (storefronts, pedestrians etc.) and lively (Fig. 5).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

141

Figure 4. Walk through along the partially segregated railway as median of the multimodal boulevard

Figure 5. Walk through along sidewalks on the multimodal boulevard

The waterfront promenade is a public space with fewer commercial activities on the ground floors and limited traffic. It has an unfavourable northern orientation. The building faรงades follow the perimeter of the block only sometimes. The neo-traditional character of the street frontage (existing along the boulevard) is broken with extensions outwards or balconies, contrary to the boulevard. The faรงades along the waterfront promenade are often more open and private and that has impact on the character of the urban space (Fig. 6).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

142

Figure 6. Walk through the waterfront promenade

The faรงades along the elongated park are similar as in the late modernist Swedish suburbs. They are open with big windows and balconies. The modernist suburbs are known as garden cities (Le Corbusier, 1929) and the buildings are arranged and oriented in a way that the windows have broad vistas of parks with a feeling of living in nature. This character exists in the elongated park. The building faรงades are open and the character of the urban space is more private (Fig. 7).

Figure 7. Walk through the elongated park

The buildings also form partially enclosed courtyards that open towards the elongated park as semipublic space and to passages to the side streets. The urban design and landscaping of the quasi courtyards varies and they are distributed in a regular pattern. They are carefully designed and landscaped characterised

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

143

by a tendency to enclose, isolate and ‘privatise the urban space’. Some urban spaces are designed as living rooms outside of the buildings (Fig. 8).

Figure 8. Walk through the semi-private courtyards

The side streets are second semipublic spaces with a different character to the elongated park. The neighbourhood has inherited a road hierarchy from modernist traffic engineering with a main street as primary artery and secondary side streets, often cul-de-sacs equipped with traffic calming measures. The buildings along the secondary streets are predominantly residential and the building façades follow the perimeter of the block. There is a neo-traditional character of the urban space without commercial street frontage and that deters pedestrians. These streets are semi-public places for transit or parking. The façades are different, sometimes more modernist with wide windows and large balconies and sometimes more traditional with smaller windows and narrower balconies (Fig. 9). A summary of the walkthroughs is shown in Table 2. Table 2. Summary about the character of urban spaces in respect to the character of the underlying elements (streets/urban space, building façades/street frontage) Street/urban space Façade/street frontage Building Character utilisation Boulevard Arterial car traffic, Closed façade with open street Commercial Public space parking, cycle ways, frontage, no balconies, following ground floor along (activities) sidewalks, LRT the perimeter of the block the entire boulevard, residential above Waterfront Limited car traffic, More open façade with sections of Commercial Public space promenade sidewalks open street frontage, following ground floor in (glimpses of the perimeter of the block in sections, residential some sections, balconies and residential above privacy and extensions outwards in some activities) sections Elongated No car traffic, Open façade, no street frontage, Residential Semipublic space park pedestrian zone not following the perimeter of the (residential block, balconies privacy) Courtyards No car traffic, Open façade, no street frontage, Residential Semiprivate pedestrian zone not following the perimeter of the space block, balconies (residential privacy) Side Limited car traffic, Closed façade with closed street Residential Semipublic space streets parking, sidewalks, frontage, following the perimeter (lacking cul-de-sacs of the block, balconies in some activities) sections

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

144

Figure 9. Walk through the side streets (Bebordgatan)

The compact city argument is that increasing density and mixing of functions (residents and jobs) results in less travel. The tendency with the development of Hammarby Sjöstad was to create mixed and dense areas. The residential density in Hammarby Sjöstad increased seven times, whereas the number of jobs increased 50%. Luma was an office park in an industrial brownfield before the redevelopment started, and Mårtensdal is a mix of office park and industrial zone (with district heating plant). The three areas in Hammarby Sjöstad (Mårtensdal, Luma and Sickla Udde) each have a different mix of functions through the redevelopment. Sickla Udde is predominantly residential and the compact city neighbourhood pattern (commercial ground floor with three to five storeys residential above) is most complete. In Mårtensdal and Luma office parks and some industries were retained, while being surrounded by a neighbourhood type of development as in Sickla Udde. Statistically, Sickla Udde best represents the compact city neighbourhood stereotype (Table 3). Table 3. Densification in the three areas in Hammarby Sjöstad (large part of Mårtensdal is currently under Mårtensdal

development). Sickla udde Hammarby Sjöstad

Luma

Residents Jobs Residents Jobs Residents Jobs

Stockholm region

Residents

Jobs

2003

453

2862

363

1697

1583

159

2399

4718

Residents (1000s) Jobs (1000s) 1 861

953

2004

519

2740

1330

1905

2422

183

4271

4828

1 873

966

2005

520

2555

2369

1775

2718

227

5607

4557

1890

972

2006

201

2654

3387

2038

3278

323

6866

5015

1 904

1000

2007

139

3050

4597

2250

3512

339

8248

5639

1 934

1035

2008

969

2922

5080

2328

4095

364

10144

5614

1 965

1054

2009

3125

3176

5339

2497

4841

402

13305

6075

2 000

1043

2010

3237

3293

5977

2703

4959

453

14173

6449

2 037

1077

Source. Statistics about Stockholm (Statistik om Stockholm), Sweco and Statistics Sweden (SCB).

One of the characteristics of the sustainable neighbourhoods designed in the compact city model is multimodal boulevards. Multimodality is achieved by prioritising public transportation systems on the ground like BRT (often BHLS in Europe) and LRT operating as medians on partially segregated busways and railways. BRT and LRT are future public transportation systems that should contribute to more

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

145

sustainable mobility. The analysis of passenger numbers boarding in the statistical areas of Mårtensdal, Luma and Sickla Udde, Hammarby Sjöstad and Stockholm regions shows that the numbers of journeys on the Tvärbana LRT system, were either steady or decreasing per aggregated number of residents and jobs during the development (Table 4). The average number of annual journeys by public transportation for Hammarby Sjöstad is slightly higher than the average in the Stockholm region, but does not indicate a major shift towards public transportation and sustainable mobility. However there are urban quarters like Luma and Sickla Udde that outperform Mårtensdal, which is still dominated by industries and is currently under development. The pattern in Luma in 2010 did not improve from 2003. The densification added new journeys by public transportation as total number, but it did not produce a dramatic shift in mobility in the 650-sum game of aggregated number of journeys by resident and jobs (Table 4). Table 4. Annual number of passengers boarding in the stations in Hammarby Sjöstad per aggregated number of residents and jobs from 2003 to 2010. The number in parenthesis shows number of journeys in a context of 1000 annual journey person as mobility invariant: it translates to 650 annual journeys per resident and job. The data about Stockholm region is from Trafikanalys, whereas the number of journeys is calculated from statistics from Statistics about Stockholm and SL. Mårtensdal Luma Sickla Udde Hammarby Sjöstad Stockholm region The number of residents The number of residents The number of residents The number of increased by roughly 4000 increased by roughly 6000 increased by roughly 3500 residents increased with 1000 new jobs by roughly 13500 with 1000 new jobs LRT LRT LRT Buses Buses Buses (Tvärba Total (Tvärba Total (Tvärba (71, 74, Total Total Total (74, 96) (74, 96) na) na) na) 96) 170 370 130 225 2003 (260) (560) (200) (341) 190 290 110 220 2004 (290) (440) (170) (334) 200 260 120 219 2005 (300) (390) (180) (332) 220 260 130 229 2006 (340) (400) (200) (349) 190 260 150 228 2007 (290) (400) (230) (350) 180 20 200 250 30 280 150 110 260 260 226 2008 (280) (30) (310) (380) (50) (430) (230) (170) (400) (400) (347) 110 30 140 240 70 310 130 120 250 240 227 2009 (170) (50) (210) (360) (110) (470) (200) (180) (380) (340) (345) 130 30 160 260 80 340 150 170 320 280 227 2010 (200) (50) (240) (400) (120) (520) (230) (260) (490) (400) (347) Source: Statistics about Stockholm (Statistik om Stockholm), Statistics Sweden (SCB), Sweco, Swedish government agency for transport analysis (Trafikanalys) and Stockholm’s public transport (SL).

Other European compact city neighbourhoods In many other Swedish cities there is a wide replication of the Hammarby Sjöstad model. The replication is not limited to Sweden, and the compact city model conceptually is used widely in all northern and western Europe countries. The neighbourhoods along the BRT and LRT lines in Gothenburg and Stockholm are not very conceptually different to the neighbourhoods along the BRT or LRT lines in Copenhagen, Helsinki, Paris, Eindhoven, Amsterdam or Cambridge. It is predominantly executed in the brownfields of the historical, now abandoned, industrial urban fringe, and there are also many new suburbs based on the same model on greenfields. There are both similarities and differences between the building façades and the urban designs of the compact city model. More experimentation with high-tech architecture is common for the towns and cities in Sweden, Denmark and Finland as well as in the large cities such as Amsterdam and London. In

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

146

Paris the architecture leans towards â&#x20AC;&#x2DC;classicalâ&#x20AC;&#x2122; modernism rather than neo-modernism, but the urban pattern of development along a partially segregated busway or railway is consistent. Many busways are drivers for development of new suburbs with moderate density housing especially in UK and The Netherlands on the outskirts of Amsterdam and London and smaller towns or cities such as Cambridge or Eindhoven. In UK and The Netherlands, with a tradition of living in row houses, the compact city model includes neighbourhoods with row houses that are organised along BRT or LRT lines. Development in Eindhoven is near the airport which is a Dutch hub for low-fare airways (Fig. 10).

Figure 10. There are similarities and differences between the cityscapes in different towns and cities in Northern and Western Europe (photographs 1-3 from up and left to right and down are from Stockholm, 4-8 are from Helsinki, 9 is from Copenhagen, 10-11 from Gothenburg, 12-14 from Amsterdam and 15-16 from Eindhoven)

BRT and LRT are entangled with urban development in almost every city in northern and western Europe. The partially segregated busways and railways are constant features of the development model. These systems are disadvantaged in a sense of speed (Fig. 11).

Discussion and conclusions Stereotype and innovativeness The society in northern and western Europe is in a period of transition to postmodernity, from welfare to knowledge and from industry to high tech, characterised by an innovation driven economy and knowledge society. Deindustrialisation is a consequence of global business and transnational capital moving industry to the new global manufacturing fringe and catalyses a process of industrialisation. The cities in northern and western Europe are specialising in knowledge and innovation oriented business and there is a demand for urbanity and access to networks with global reach (Table 5). The abandoned brownfields in proximity to the attractive industrial urban cores are transformed into habitats of the

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

147

future knowledge society. The compact city model has an explicit role as developer of the postmodern city and targets this specific urban edge of the historical urbanisation in northern and western Europe. The earliest industrial zone is a new front for development and the city centres are slowly extending their edges. The urbanisation in a Swedish context that equally concerns rest of northern and western European countries emerged in eras characterised by traditional society, industrialisation, welfare state and knowledge society (Engström and Cars, 2008).

Figure 11. Partially segregated busways and railways in different towns and cities as constant in the compact city model (photographs 1-2 are from Helsinki, 3-4 from Gothenburg, 5 is from Paris, 6 from Douai, 7 from Cambridge and 8 from Amsterdam) Table 5. Epoch of the Swedish society in respect to development of the economy and industry Source: expanded from Cars and Engström, 2008 Economy and Need of Communication and Epoch Work character industry access transportation technology Work by demand and Traditional Spatial Narratives and stories, private Labor division wages by product or society proximity carts and carriages harvest Newspapers and telegraph, Industrial Mechanisation Wages by hour and Spatial public omnibuses, trams and society of labor maximum work hours proximity trains Specialisation Wages by month with Television, radio and telephone, Welfare Temporal of processes limited 40 work hours per private cars and public buses state proximity and automation week and trains Knowledge Specialisation society of services

Flexible work hours and wages by product

Proximity to a network

World wide web, private and public cars and jets and high speed buses and trains

Urban structure City and villages City and region Urban region Urban network with global reach

Standardisation and replication produces uniformity and as a consequence there are many ‘Sjöstads’ in Sweden and around northern and western Europe. When a neighbourhood becomes typical there are always risks. The attractiveness of the original can sometimes decrease if there are many similar copies. The experience of multiplied modernist suburbs in Sweden resulted in low attractiveness (Rådberg, 1997). The copied neighbourhood might lose its appeal in a postmodernist paradigm of uniqueness. On the other hand the mix of traditional and progressive, and high urbanity might be cherished. The question of stereotype and innovativeness has deep roots in architecture. In the “Books of architecture” from the 1st century BC, Vitruvius writes about the origins of architecture, inventions and methods of transmission: They began, some to make shelters of leaves, some to dig caves under the hills, some to make of mud and wattles places for shelter, imitating the nests of swallows and their methods of building.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

148

Then observing the houses of others and adding to their ideas new things from day to day, they produced better kinds of huts. Since men were of an imitative and docile nature, they boasted of their inventions as they daily showed their various achievements in building (1931, pp. 79-81). Novel and unique architecture and urban designs, images and diagrams often evoke responses as copies, interpretations and critique. It drives innovation and establishes traditions, styles or paradigms. The surgery in Paris from the mid-19th century is a historical event that caused strong reactions in Europe. Baron Haussmann interposed new boulevards in the traditional urban fabric and regulated the façades into lavish cityscapes. In the spirit of the 19 th century, he conserved a part of the past (Choay, 1969) and enriched it with new infrastructure. The European society was fascinated and perturbed by this novel and contradictory mix caused by the injection of modern technology in the traditional city (Benevolo, 1980). Internationalisation by increased communications between architects from different countries started with the modern movement and CIAM (Congrès International d'Architecture Moderne) at the end of the 1920s. The architectural standards and norms grew in popularity during the modern society. It was advocated as uniformity in detail by Le Corbusier (1931), but also by Bauhaus Walter Gropius and Ernst Neufert. The European compact city model needs variety. Master architects can experiment within neomodernism and new-traditionalism in traditional-modern synthesis. It is possible to experiment more with traditional street patterns and modern blocks. An example of such mix is a project by BIG in New York that fuses skyscraper design and European urban block in Manhattan’s street network (Ingels, 2012). Much can be learned from the master architects in American New Urbanism. Transit-oriented development (TOD) and the compact city are conceptually very similar, but the experiences in the USA are much more varied than in Europe. One of the reasons for this is high quality urban design through a diversity of urban spaces. For example, TOD in Portland is similar in urban structure to the European compact city neighbourhoods, but the architectural styles and heights of building are different. Every LRT stop is clearly defined by a tower which is embedded in the structure of the neighbourhoods. In Los Angeles, TOD is very broadly interpreted from very dense urban environments along the Red Line, a subway line, to simply accentuations by landmark buildings along the Gold Line, a LRT system. San Diego is the city that was first in USA to adopt TOD guidelines and the TODs along the San Diego trolleys are very varied (Fig. 12). Another direction is to highlight the green city in the European sustainable cities debate and discuss hightech architecture in regard to more ecological design experiments with ecomorphic, mimicking creations of natural environments, instead of anthropomorphic architecture. Why not mix the compact and green city into a hybrid? It is possible to design high-tech arcologies or underground cities – cities that imitate hills and termite mounds (Stojanovski, 2007). In a same direction Malcolm Wells advocates earth-covered architecture in a greener context. Wells (1981, pp. 34-40) identifies how the architecture fits into the wilderness. He observes that buildings consume water, food, air, soil, energy, resources but they do not clean their own waste, they do not produce food, they pollute and they destroy the wildlife. The mix of high-tech and ecological architecture in a traditional-modern struggle is also a possibility for the futurehistoric European city. In a very futuristic direction the high-tech and ecological architecture exists as cosmomorphic design including artificial Suns, photovoltaics and urban agriculture. The ‘Sjöstad’, literally translated from Swedish a waterfront city, is emerging as a stereotype. The model is a neo-traditional neighbourhood with modern high-tech architecture as redevelopment of the abandoned fringe of factories around the industrial urban core. The ‘waterfront’ is equally important as a zone of factories in the industrial city, or as a fringe of sustainable neighbourhoods in the postmodern city. The sustainable neighbourhoods in Stockholm like Hammarby Sjöstad is very attractive today, but in the future they will attract competition from other compact city neighbourhoods on the waterfront with strong environmental profile. Norra Djurgårdsstaden is a new neighbourhood in Stockholm that can cast a shadow on Hammarby Sjöstad. Do we need more replications and more stereotypes in Europe when the postmodern revolves around uniqueness and originality? Why not experiment more? Why not break the stereotype? Malcolm Gladwell puts this in a postmodern perspective: It is not about Pepsi, but about

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

149

Pepsis (the original, light, cherry, green, retro, fluffy, happy, super, etc.)! Do we need compact city or compact cities (the original, light, cherry, green, retro, fluffy, happy, super, etc.)?

Figure 12. The American TOD experiences (photographs 1-7 from up and left to right and down are from San Diego, 8-9 are from Portland and 10-12 from Los Angeles)

The stereotypes can be broken by innovative architecture and urban form. One solution is master architecture and the TOD projects in the USA designed by master architects are good examples. Another direction is to pursue the green city as rigorously as the compact city, and through experiments. High-tech architecture can be discussed in respect to ecological design and â&#x20AC;&#x2DC;ecomorphicâ&#x20AC;&#x2122; (imitating creations of natural environment) experiments, instead of anthropomorphic architecture. The fixity and completeness of the model The problem inherited from the Modernist movement manifests when the compact city model is standardised and industrialised for development at a large scale. The models are finished and it is difficult to make adjustments. In the pattern of buildings, private spaces and public spaces there is no space for additions or changes. There is a need to address the disadvantages of the completeness of the model and the experiences with modernism from the 20th century. Swedish modernism revolved around ABC cities

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

150

and suburbs, where A stands for arbete or work, B for bostad or residence and C for centrum or centre. The prototypes in Vällingby and Farsta had unique modern architecture and continued to develop throughout 1990s and 2000s. There was a space left. The latter ABC cities were fixed and completed. Tensta or Rinkeby did not change and slowly deprived in the following decades. Hammarby Sjöstad is a success today. It is livable, attractive and lively, but it emerges as a finished and universal model of development at large scale in northern and western Europe. Do we really need more models? What is the alternative? The rules are procedures for conceiving and generating space and models are prototypes, a model space or a model of space (Chaoy, 1997). The alternative is to better understand regularities in urban form and this knowledge is presented as rules. One rule is to have streets or boulevards narrower than 30m to achieve urban feeling and enclosure. The BRT or LRT lines are often designed as medians on multimodal avenues that are wider than 30m. The heuristic width of anthropomorphic urban space is within a hearing range of 30 (Gehl, 1987) and beyond that width the social distance disappears. The boulevards in the compact city model are often wider than 30m. There is a need for rules about urban form in their historical and future development and urbanisation. The compact city model with its high-tech architecture, multimodal boulevards, sophisticated LRT or BRT systems, emerges as a standardised and universal postmodern solution, but it is not applicable in all historical fringes. It is a vision of new neighbourhoods that integrates historical landmarks, but requires width for development either as greenfields or bulldozed brownfields. The wide perspective of transformation and introduction of more efficient transport technologies in different urban environments in the contexts of societies, fringes and typologies is neglected. Breaking the stereotype is sometimes an escape from the focus on the urban by looking at the global and changes in perspectives in the society and economy. From another perspective changes can be small scale. The model of physical form at large scale is exchangeable with a model of processes and rules at very small scale. Then the power of global trends, information flows and knowledge is only peripheral to the premise for future architecture and urban design. How sustainable are these neighbourhoods? There is much focus on sustainability of urban form, but less on urban flows. Hammarby Sjöstad is a lively and attractive neighbourhood in Stockholm. It is a successful and innovative experiment in a direction of high-tech architecture and urban design, recycling and green materials, but the performance as sustainable transportation and mobility is not convincing. The Tvärbana acts as primary public transportation axis in the neighbourhood, but is a secondary, orbital link in the regional public transportation system in Stockholm. There is a need to do an additional transfer to the Tunnelbana, the metro system, or the Pendeltåg, the commuter rail, to reach the primary public transportation network. With an average speed of 20-25km/h BRT and LRT are not very competitive in the urban regions where the average speed of car traffic is 35-50km/h. The commuter rail systems which usually define the primary public transportation network operate with 50km/h or higher speeds. With longer distances the attractiveness decreases and even the smaller towns and cities in northern and western Europe are sprawled. A person in Sweden travelled an average 45km and 70 minutes each day in 2011 and 2012 (Trafikanalys, 2012). At aggregate level BRT and LRT operating on partially segregated busways or railways cannot offer accessibility within that distance and time travelled. They also cannot achieve high frequencies because they will block the traffic on the intersections on the boulevards. To achieve multimodality there is a need for a continuous conceptualisation of and debates about public transportation, the urban form of their infrastructures and their capacity to facilitate flows. There are different public transportation infrastructures that are morphologically similar and facilitate consistent urban flows. In respect to the streets as traditional element of urban form (Conzen and Conzen, 2004; Krier, 1979, 1984) there is: 1) public transportation on streets (buses and trams); 2) completely separated, either elevated or on the ground (heavy railways or heavy busways); 3) underground (subways); and: 4) partially separated on ground (light railways or light busways) (Stojanovski, 2013). Each principal

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

151

infrastructure facilitates linear or nodal accessibility, and depends on the traffic flow and the capacity of the transportation system. To meet the urban challenge of the “Our common future” report there is a need for holistic understanding of urban form as liquid and urban flow as fixed. The towns and cities in northern and western Europe enjoy modernity through automobility and life at a distance. The challenge in the postmodernist future is not to be immobile, but to be sustainably mobile. There is a need to radically transform and equip the modernist towns and cities with more efficient public transportation systems than LRT and BRT as medians on boulevards (even though they suffice in certain contexts). The morphological understanding of the urban form as elements (typologies of streets, transportation infrastructures, lots, buildings) can both help in understanding and breaking the stereotype (not focusing only on BRT and LRT in the compact city model, because they lack speed). It can also result in the choice of the right elements in the right context to achieve better connectivity with public transportation, and thus contribute to more sustainable mobility.

References Alexander, C. (1979) The timeless way of building (Oxford University Press, New York). Ash, A., & Thrift, N. (2002) Cities: reimagining the urban (Polity, Malden). Benevolo, L. (1980) The history of the city (Scolar P, London). Birkhamshaw, A. J., & Whitehand, J. W. R. (2012) Conzenian urban morphology and the character areas of planners and residents. Urban Design International, 17(1), 4-17. Caniggia, G., & Maffei, G. L. (2001) Architectural composition and building typology: interpreting basic building (Alinea Editrice). Cars, G. & Engström, C.-J. (eds.) (2008) Stadsregioners utvecklingskraft: trender och nya perspektiv (Kungliga tekniska högskolan, Stockholm). Cataldi, G., Maffei, G. L., & Vaccaro, P. (2002) Saverio Muratori and the Italian school of planning typology. Urban Morphology, 6(1), 3-14. Choay, Françoise (1969) The modern city: planning in the 19th century (George Brazillier, New York). Choay, F. (1997) The rule and the model: on the theory of architecture and urbanism (MIT Press, Cambridge). Conzen, M. R. G. (1960) Alnwick, Northumberland: a study in town-plan analysis. Transactions and Papers (Institute of British Geographers), iii-122. Conzen, M. R. G., & Conzen, M. P. (2004) Thinking about urban form: papers on urban morphology, 1932-1998 (Peter Lang Pub Incorporated). Csíkszentmihályi, M. (1990) Flow: the psychology of optimal experience (Harper & Row, New York). Cullen, G. (1961) Townscape (Architectural Press, London). EC. (1990) Green paper on the urban environment (EC). EC. (1997) Towards an urban agenda in the European Union (EC). EC. (1998) Response of the EC Expert Group on the Urban Environment on the Communication “Towards an Urban Agenda in the European Union” (EC). EC, (2011) White paper 2011. Roadmap to a Single European Transport Area – Towards a competitive and resource efficient transport system (EC) Dantzig, G. B., & Saaty, T. L. (1973) Compact city: A plan for a liveable urban environment. San Francisco: WH Freeman. Gehl, J. (1987). Life between buildings: using public space (Van Nostrand Reinhold, New York). Ewing, R. & Cervero, R. (2010) Travel and the Built Environment. Journal of the American Planning Association, 76(3), 265-294. Harvey, D. (1973/2009) Social justice and the city. (revised edition) (University of Georgia Press, Athens). Ingels B. (2012) Hedonistic sustainability. Lecture at KTH Royal Institute of technology, February 15. Jenks, M., & Gerhardt, W. (2000). Urban intensification: the potential of decision support for assessing brownfield sites. Urban Design International, 5(3-4), 3-4. Kostof, S. (2005) The city assembled: the elements of urban form through history (Thames & Hudson, London). Krier, R. (1979) Urban space (Academy Editions, London). Krier, L. (1984) Leon Krier: houses, palaces, cities (Architectural Design, London). Le Corbusier. (1931/1986) Towards a new architecture (Dover publications, New York). Lynch, K., & Rodwin, L. (1958) “A theory of urban form”, Journal of the American institute of planners, 24(4). Marchetti, C. (1994). “Anthropological invariants in travel behavior”. Technological forecasting and social change, 47(1), 75-88.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

152

Marshall, S., & Çalişkan, O. (2011) A joint framework for urban morphology and design. Built Environment, 37(4), 409-426.McLoughlin, J.B. (1969). Urban and regional planning: a systems approach (Faber and Faber, London). Melia, S., Parkhurst, G. and Barton, H. (2011) “The paradox of intensification”. Transport Policy, 18(1), 46-52. Moudon, A. V. (1997). Urban morphology as an emerging interdisciplinary field. Urban morphology, 1, 3-10. Mumford, L. (1938) The culture of cities (Harcourt, Brace and Company, New York). Naess, P. (1995) “Travelling distances, modal split and transportation energy in thirty residential areas in Oslo”, Journal of Environmental Planning and Management, 38(3), 349-370. Naess, P. (2006) Urban structure matters: residential location, car dependence and travel behavior (Routledge, London). NHTS (National Household Travel Survey, U.S. Department of Transportation, Bureau of Transportation Statistics). (2011) SUMMARY OF TRAVEL TRENDS 2009. Retrieved from http://nhts.ornl.gov/2009/pub/stt.pdf Park, R., Burgess, E., & Mckenzie, R. (1925) The city (University of Chicago Press, Chicago). Reclus, E. (1905) L”homme et la terre, tome cinquième. Paris: Librarie universelle. Rossi, A. (1984) The architecture of the city (MIT Press, Cambridge). Rådberg, J. (1988) Doktrin och täthet i svenskt stadsbyggande 1875-1975 (Statens råd för byggnadsforskning, Stockholm). Rådberg, J. (1997) Stadstyper i Västerås: bebyggelseanalys med hjälp av geografiskt informationssystem (Kungliga Tekniska högskolan, Stockholm). Susilo, Y. O., & Stead, D. (2009) “Individual carbon dioxide emissions and potential for reduction in the Netherlands and the United Kingdom”, Transportation Research Record: Journal of the Transportation Research Board, 2139(1), 142-152. Stojanovski, T. (2007) Sustainable cityscape: future study of one neighbourhood in the city of Skopje with a story and three essays (Skolan för arkitektur och samhällsbyggnad, Kungliga Tekniska högskolan, Stockholm). Stojanovski T., (2013) Public transportation systems for urban planners and designers: the urban morphology of public transportation. Paper presented on the 3rd International Conference on Urban Transportation Systems (Paris, France). Trafikanalys. (2012) The Swedish national travel survey 2011–2012. Accessed on 19th August 2013 from: http://trafa.se/PageDocuments/RVU_Sverige_2012.xlsx Unwin, R. (1912) “Nothing gained by overcrowding”. Garden Cities and Town Planning Association, London, 23. Vitruvius. (1931) On architecture, Volume 1 (Willian Heinemann, London). Wells, M. (1981) Gentle architecture (McGraw-Hill, New York). White, P. R. (2008) Public transport: its planning, management and operation (5th edition) (Routledge, London). Whitehand, J. W. (1967). “Fringe belts: a neglected aspect of urban geography”, Transactions of the Institute of British Geographers, 223-233. Williams, K., Burton, E., & Jenks, M. (1996). “Achieving the compact city through intensification: an acceptable option”. in The compact city: A sustainable urban form, 83-96. WCED. (1987) Our common future (Oxford University Press, Oxford). Zahavi, Y. (1974) Traveltime budgets and mobility in urban areas (Federal Highway Administration, Washington).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

153

[This page is intentionally left blank]

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

154

REGIONAL CENTRES cities and towns with local importance, at the edge of national and regional networks

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

155

The evolution of urban grid of Indus Civilisation: case studies of Mehar Garh, Mohan Jo Daro, Cave City ‘Rakas Jo Rohro’ at Rohri and ‘Sat Gharhi’ at Sehwan, Sindh, Pakistan Javeria Shaikh Division of Architecture, Hanyang University, South Korea Email: javeria@hanyang.ac.kr

Abstract. The relationship between urban form and sustainability is one of the most intensely debated issues on the international urban morphological agenda. The way that cities were formed and the effects of their form on resource depletion and social and economic sustainability are central to this debate. This research provides an overview of the theoretical advances in the compact city debate, and the present thinking on the first historic compact city development. The urban grid used at Indus Civilisation has a long history and it was first used at the sites of Mehar Garh and Mohan Jo Daro. This paper revises the archaeological findings concerning urban grid and pushes back the occurrence of the first urbanism to 12,000 year old Mesolithic industrial cities to Late Stone Age planned caved city settlements. Four precedent areas were selected. These include Mehar Garh, Balochistan, Mohan Jo Daro Indus Valley Civilisation, Sat Gharhi at Sehwan, and Cave City Rakas Jo Rohro at City of Rohri Sindh. These areas were comparatively analysed on the basis of six variables i.e. movement of people, shape geometry, sanitation grid, public and private domain, security systems and visual connectivity. The analyses was carried out via space syntax or ‘depth map analysis’, ‘graph analysis’ and ‘attributes explorer’ analysis, and statistical analyses were executed with the use of ANOVA online software application. Conclusively it was found that all these settlements qualify to be called a city with an urban grid that makes world’s older compact cities. Keywords: Urban grid origin, urban form, urban morphology, Mohan Jo Daro, ANOVA

The study of the urban grid is not new for researchers of urban morphology. Various attempts have been made by urban designers and planners to study and analyse urban grids of cities to identify a perfect grid plan of city that suits the city dwellers and satisfy the functions of city. The urban grid is basically understood as an orthogonal grid that is used as an arrangement of urban form and its four-dimensional formation. Various algorithms concerning urban grid can be found throughout the history of urban form. The significant element of urban grids is that it makes a basic cartographic diagram to produce functionally well-organised spaces. The origins of the urban grid are evident in different civilisations such as Greeks and Romans who have used the orthogonal grid in their planned cities where the grid design was developed for planning of cities and towns. It is observed by different researchers of urban morphology that, in the beginning of human civilizations, the urban grid used rectilinear horizontal and vertical lines as a tool for spatial development of an urban context. The urban grid formation has numerous geological and geographical qualities that lead to development of chronological-three-dimensional forms in different civilisations.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

156

It is evaluated by urban morphologists that urban grid play a significant role with respect to land form and geographical features of the context. The generic expectation from the urban grid is manifold such as a city grid must have efficient circulations of people and goods. The urban grid may depict a proper shape geometry that fulfils the aesthetic function of city. The urban grid must respond to the needs of proper disposal of solid and liquid waste to achieve public health. The urban grid is also expected to integrate public spaces with private places to create an urban place value. The urban grid is also expected to provide safety and security function to the city. Finally, urban grid is also anticipated to deliver a visual connectivity of urban elements to depict the urban character. This study is also one such attempt where the context of study was found in the history with the objective to conduct an urban morphological analysis of selected historical urban contexts of Mehar Garh in Balochistan and Mohan Jo Daro of Indus Valley Civilisation at Larkana District of Sindh Province of Pakistan. The other contexts as part of this analysis are Sat Gharhi at Sehwan Sharif City Sindh, and Cave City of ‘Rakas Jo Rohro’ popularly interpreted as ‘Devil’s Mouth’ located at the City of Rohri Sindh. This morphological analyses is not done in isolation from theoretical underpinning but it is based upon the compact city debate, and the present thinking on the first historic compact city development. In this respect case studies of each selected context is carried out and its theoretical and visual findings are outlined. The evolution of urban grid presented in this research paper is also based upon the archaeological findings concerning urban grids of Mesolithic industrial cities to Late Stone Age planned caved city settlements. The morphological analysis of selected historical urban contexts used six variables i.e. movement of people, shape geometry, sanitation grid, public and private domain, security systems and visual connectivity. The variables used in the analyses are basically emerged out of review of existing literature on urban grids. These variables are then applied to all the selected historical urban contexts and developed its grading via space syntax software and converted the physical realities of urban grids into numerical data. This numerical data was further analysed statistically with the use of ANOVA online software applications. The findings of research are outlined below.

Literature review In this urban morphological research a very deep literature review carried out concerning history of urban grid in old civilisations of the world. These include Indus Valley Civilisation, Mehar Garh Civilisation, Greek civilisation, Roman Civilisation, and Chinese Civilisation. Although urban grid is studied in various cities of different civilisations but, the main focus of studies was the evolution of urban grid in Pakistan because it was found to be one of the oldest urban grids as evident from the Stone Age Cave Men city of Rohri. Secondly a critical review is also carried out about the concept of compact cities and the current discussions across the globe about its role in creating sustainable cities. Then the concept of compact city is explored within the historic and prehistoric cities of Pakistan and their urban grids. As there was a natural relationship between the concept of urban grids and compact urban structure of cities of Indus Valley and Mehar Garh in Pakistan. Therefore detailed case studies were undertaken and presented here. This study categorises the foundation of urban grid arrangement which is a correlational and time based research study which was carried out on regressive archetypal model by using the factors and variables obtained from literature review. There is a visual, descriptive and statistical analysis that illustrates the possible linkage between cavemen in the Late Stone Age urban grid system and compact cities of the Indus Valley and Mehar Garh Civilisation. It is anticipated that this research make a major influence on the evolution of urban grid system and compact cities within the history of urban morphology.

Methodology and approach Four morphological case studies were carried out and a comparative analysis was done with theoretical explanation of the realities of each context as well as their collective statistical comparison. A

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

157

correlational and time based research study was carried out on regressive archetypal model. Whereas, a cross-sectional research method is also applied on this archetypal model, so as to compare the earlier urban grid system with the latter along with different research variables. Furthermore space syntax or ‘depth map analyses, ‘graph analyses and ‘attributes explorer analysis was also carried out. As a result it lead to reach at the nature of urban grid system and compact city qualities respectively. Basically this paper have found out the archeological evidence with a primary source of physical documentation of the context which leads to a further earlier onetime invention of the urban grid and a compact city i.e. Rakas Jo Rohro Hills and cave city of 10,000 BC. Here the prototype for the urban gird and compact city planning for Mohan Jo Daro is suggested to be the orthogonal grid plan for the 10,000 BC Seven Caves at Kai. Another significant aspect is that the research method applied in this research paper also coincides with the research conducted by Rose-Redwood (2008). In his research, he explains that the urban orthogonal grid is neither an amicable model nor a timeless form. It only has similar urban geometry, whereas the urban grid is an incongruity for a variety of morphologies. He also suggested documenting the similar appearance of grids then critically analyzing the spaces generated by them. An urban grid is basically a partition that pinpoints the origin therefore this paper traces the prototype for the origin of the urban grid. The ancient urban contexts were planned on the logic of urban morphology and they practiced common sense ideas for urban grids, Smith (2007). Similarly Spiro Kostof (1993) is another research scholar of urban morphology who has used the term urban grid with reference to orthogonal planning, in his study of Jericho, an 8000-6000 BC compact city. Later on, the other advanced civilisations of Greeks and Romans arose and their compact cities were also planned on orthogonal grid systems.

Evolution of urban grid in Pakistan Around 10,000 BC, major development towards a settled form of life took place in the region. From dependence on wild plants and animals, a move was made towards organised farming in the river valleys in the hilly regions of Northwestern Balochistan and in the plains. These hunters and the gatherers selected spots along the banks of mountain stead in the plains, where they selected cultivation of crops such as cereals and domestication of sheep and goats. For some time, instead of constant wanderings from place to place, these semi-nomads undertook seasonal migration. They spent their winters in the plains and returned to the settlements in the hills during summer. This semi-nomadic life relied on the cultivation of crops, rearing livestock, making vessels out of clay and building houses out of mud and reeds. In the early period, when the practice of agriculture was still in the elementary stage, these seminomads continued with their hunting and gathering to supplement the local food resources. In due course some of these nomads gave up their seasonal migrations altogether and established the first permanent settlement in the Indus Plains. Here, the first urban planning of compact city model emerged. They lived in caves i.e. Rakas Jo Rohro near Rohri and Sat Garhi at Sehwan City of Sindh. The Congress of the Asian Archaeology in New Delhi in 1961 decided that in the existing state of knowledge, the Stone Age may be divided into Early, Middle and Late Stone Ages. In Baluchistan and Sind, for the same period of settled life, animal husbandry and some form of cultivation depended solely on the use of stone. This is supposed to constitute primary Neolithic phase or Mesolithic or Paleolithic Phase or 9500-5500 BC. Therefore, the urban settlements of Sindh and Balochistan were the oldest Late Stone Age cities. Case study of Mohan Jo Daro Mohan Jo Daro, a perfectly patterned gridiron and as precise as modern day New York City, served 35,000 inhabitants (Bulliet et al.). This gridiron contained the covered ducts to refrigerate the grains. A plumbing substructure which achieved premium of hygiene with parallel underground sewage system moving towards Indus River independent of the city’s drinking water supply (Higgins 2009).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

158

Gangal et al. (2010) carried out the spacio-temporal study of Indus Urbanisation and gives an account with rough deviation probability about the spread of the Indus sites, the Neolithic site, spread over the temporal period of 2000 years. The compact city plan of Mohan Jo Daro was not based on religious orientation as in Egypt and Mesopotamia. At stages of historic evaluation of Mohan Jo Daro, its urban grid evidently represented political and economic regimes. It was associated with diverse cultural interpretation and geographical development practices as observed in the map of the cityâ&#x20AC;&#x2122;s public and private domain (Fig. 1).

0 3m 6m

Figure 1. Compact City of Mohan Jo Daro and its Urban Grid (Mohenjo Daro Archives, Larkana Division)

The urban morphology of this city clearly depicts that the city has a most advanced compact urban grid. The conscious planning of city axis and an interwoven urban grid pattern of public and private courtyards with proper accessibility made this city of refined citizens. The compact open and built up spaces present an orthogonal shape geometry. The prominent hydraulic engineering of Mohan Jo Daro makes this city a great example of prehistoric civic advancements. The city had the installation of water supply and sanitation devices in residences as well as worldâ&#x20AC;&#x2122;s earliest flushed toilets which were connected to a common sewerage system. This fact clearly spell out that the city was inhabited by highly civilised people. Case study of Mehar Garh The genesis of the Indus Civilisation can also be tracked to the early Neolithic settlement of Mehar Garh. The settlement is located on the Kacchi Plains of Balochistan Pakistan along the western borders of the major alluvial plain of the Indus River (Samad, 2000). The site was discovered by French Archaeologists Jean Francois Jarrige and Catherine Jarrige in the year 1974. This area was excavated, layers by layers with radiocarbon dating of each layer from 1974 to 1986 and later 1997 to 2000. The studies of L. Constantine and R.H. Meadow have brought to light a process of transformation of the subsistence economy in Mehar Garh. The discovery of several groups of compartmented buildings corresponds to the proto-agricultural age, at a time when meat was provided chiefly by hunting. Numerous burials found from this site such as baskets, stone and bone tools as well as place of animal sacrifice. Simple figurines of women and animals were also found from the site. A single ground stone axe was also discovered. These ground stone axes depicts that the settlement belongs to late Stone Age. It is probable that some of the relatively unusual characteristics of Mehar Garh when compared with the first villages in the area of West and Central Asia, in particular with its size, can be explained by links to an Upper Paleolithic pattern of

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

159

social organisation perhaps characterised by relatively large seasonal concentrations of population with the context.

Figure 2. Traced by Author from: Mehar Garh field report 1994-1985 Source 1d: Mehar Garh field reports 1974-1985 From Neolithic Times to the Indus Civilisation, Area MR 1 period V, structure in Sector MR C, Excavation at Mehran The ninth season 1982-1883 (p. 134)

These are mainly four building layouts found in Mehar Garh i.e. two rooms, four rooms, six rooms and ten rooms structures. The whole urban grid is composed of these structures. Fig. 3 depicts ten rooms arrangement with the seven longer rooms each about 3m x 1.5m. The walls, very thin are composed of a single row of bricks. In the northern part of Mehar Garh, two orthogonally grid planned structures were excavated one consisting three rectangular rooms, the other measuring 8x5m (Fig. 2 and Fig. 3). The geomorphologic evolution of the region, transformations of the vegetal cover, and profound changes in the agricultural and pastoral economy have certainly deeply modified the landscape and the relations between humans and the current natural environment of the excavated site of Mehar Garh.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

160

Figure 3: Traced by Author from: Detailed Map of Mehar Garh Built Up Structures Source 1d: Mehar Garh field reports 1974-1985 From Neolithic Times to the Indus Civilisation, Area MR 1 period V, structure in Sector MR C, Excavation at Mehran The ninth season 1982-1883 (p. 441)

Case study of Sat Ghari (Seven Caves) at Kai Valley Sehwan, Sindh Pakistan: Sindh province of Pakistan has many prehistoric structures located in different remote places and less explored areas due to inadequacy of expertise and patronage to subjects of historical urban landscapes. One such place is this Stone Age seven chambered cave located at Kai Valley Sehwan, Sindh (Fig. 5). The Stone Age was connected with simple hunting and food gathering, but no provision for future requirements. The men lived in small groups- always in danger of extinction with high rate of mortality and low life expectancy. Early human showed major progress, during evolution and become more humane, and discovering more material and career and more formalising planning techniques (Snooks, 2002). Furthermore as (Rousseau, 1762) refers that human become skilled from each other and there is a necessity to have supplementary neighbourhood and need for proximity between humans, this was applicable from pre-historic time. Shakespeare’s line, “what the city but the people” is true for this Stone Age settlement. Cities grow by building up, or out, and when the city doesn’t put together, people are taboo from understanding the magnetism of small hamlets that exist at remote places. The essential message in this urban grid of Sat Gharhi is that the techniques of orthogonal grid is developed in Stone Age era, and this practice on space syntax is considered as special case with respect to accessibility to safety. The urban morphology shown in Fig. 4 as depth map indicates that, the caves of Sat Garhi are usually naturally formed as a consequence of natural erosion process in calcareous environment. The caveman used these for residence, by amendment of the site according to their needs. Case Study Cave City ‘Rakas Jo Rohro’ at Shahdee Shaheed, Rohri, Sindh Pakistan: The cave city “Rakas Jo Rohro” is located at Shahdee Shaheed, Rohri city in the province of Sindh Pakistan. Locally termed as ‘Rakas Jo Rohro’ is interpreted as “Devil’s Place” or “Devil’s Mouth”. This is a prehistoric place in Sindh. Represented in Fig. 5 the morphological depth map of “Rakas Jo Rohro” or “Devil’s Mouth”. This cave is one of the biggest cave in this area, with five varying size chambers, from the late Stone Age. This cave is designed to penetrate all the way through the entire hill section, it has two entrances (Abro, 1998). This cave is 50m deep, it starts from Shadee Shaheed and culminates at the “Pug Waro Takar” or a

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

161

“Turban Shaped Hill”. Rakas Jo Rohro, grid symbolises the arrangement for early human’s safe and secure place, which is carved by Late Stone Age people and semi-sculpted through wind erosion. This is first example of bio-mimicry or naturally sculpted compact city which refers to studying nature's processes and then replicates these designs and techniques to solve current human problems of shelter.

Figure 4: Sketch Plan of Sat Ghari (Seven Caves) at Kai Valley Sehwan, Sindh Pakistan (Drawn by Author)

0 3m 6m

Figure 5. Morphological Depth Map of Rakas Jo Rohro (Devil’s mouth) at Rohri, Sindh Pakistan. (Sketch drawn by author)

Today planners suggest that in this way ecologically performing cities could be built (Antonio Gnisci 2011). However, the 10,000 year old compact cities were based on bio-mimicry because they were semisculpted by the air current. Since this space is design and the site by the caveman was chosen on the hill such that it was already fit to be occupied and then amendments could be made further in the site. The entrance here is 2.5m wide diameter and height is 2m, the total depth is 14m, the third meeting room is 4m high, the second cavity depth is 4m long, width is 6m, and high is 3m. The fourth compartment is 12m deep, 6m wide and 9m high. The fifth hollow’s height is 3m and width is 6m at the Rohri Hills, Sindh. The morphology of this compact city plan resulted to be a tubular orthogonal axial compact city. Comparative Analysis of Case Studies: The core argument of analysis is that there is a regular occurrence of the urban grid pattern ubiquitously throughout the history of cities in Sindh. The four folded determinants are applied to analyse each settlement through a cross-sectional regression model and it lead to explain the nature of urban grid

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

162

system in each studied area. The urban grid was also evaluated via six variables as narrated earlier within Mohan Jo Daro, Caves near Mehar Garh, Rakas Jo Rohro and Seven Caves at Kai. Determinant 01: Mixed wse and public private activities The first determinant is to scale the studied settlements is the nature of land use and activities of people in the urban grid. It was evaluated from the case studies that a grid for a city defines special and functional nature as a whole. Without grid, city would eradicate all the characteristics of concentration, high-quality spatial scale, controlled juxtaposition and utilisation of spaces. The continuity and incorporation of precise-ordering for the city depends on the grid (Hillier, 2007). This characteristic was fulfilled by well designed compact cities at different times as shown in the figure.1 and table 1 that synthesised the different aspects in each studied settlement. Determinant 02: Concentration of surplus The second determinant to evaluate all settlements is the process of substance economy in each settlement and concentration of surplus corresponding to urban grid. The case studies show that in the Middle Stone Age, hunting and food gathering methods were improved and some provision was made for the next few days. It differed from the Late Stone Age in which hunting and food gathering methods were more efficient and there was some provision for future needs but not self-sufficiency. It was not until the Neolithic and Chalcolithic cultures that self-sufficiency in the food producing economy was achieved with reserves for future use. It is also evident from each urban grid of settlements premeditated in this research. Determinant 03: Geometry and changeability Third determinant is the geometry and the geometrical components which make the urban grid in each studied settlement. The study found out grid in study areas as of right angles and straight lines which supports growth and change, finally density and efficient security. The framework of space syntax of pedestrian route within studied urban grids is associated with accessibility lines. The urban forms defined location or point through straight line. A line of conduct is essential to become a unit for the entire mesh of the grid, continuity and the spatial scale, order, and integration. The ability of space syntax used here is to describe the relationships of part-to-whole quantitatively and syntactical accessibility as applied to all case study settlements. Similarly, the connectivity and directness of utilitarian and recreational walking grid for the streetscape are examined which produced pure utilitarian geometry in Mohan Jo Daro and Mehar Garh. Determinant 04: Circulation analysis The fourth determinant is circulation of people that explains how grid can facilitate people to reach from one point of city to other, and this also includes how people visually connect. The study found out that, in each axially connected grid form on the temporal scale presented the flow of people, goods and mode of transport. The circulation produces an axial pattern of grid with its own fashion, representing a unique grid in each townscape period. This is similar to present day axial patterns which are recognisable one in language of the design fabric, with its distinctive architectural building style, street space, squares and plot pattern. The past urban land and building utilisation and circulation patterns have left its traces on the much altered present day grid utilisation and circulation patterns. The incidence of atypical planning periods reveals to the urban morphology what is known symbolically as â&#x20AC;&#x2DC;sequential layeringâ&#x20AC;&#x2122; and indicated the historical point in time depth concerned in the assembly of the axial grid plan. The diagrammatic one accumulates different period which features and display the diversity and strength of social and cultural contents in the axial grid pattern history.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

163

Summary and findings From the analysis of the case study on four determinants established that grid planning strategy uses point-to-point information. For example, using route or linear strategy, a person would obtain directions from point A to point B. If they needed to go farther, they would obtain directions from point B to point C, continuing in a linear fashion. The linear grid effects on Indus Valley Civilisation Mohan Jo Daro City and Mehar Garh settlement Balochistan, and “Rakas Jo Rohro 1” Devil’s Mouth 1 at Shadee Shaheed, Hills of Rohri City, Sindh, Pakistan, This was simulated on depth map analysis software developed at UCL which simulates space syntax by theoretical underpinning of connectivity and how grid can facilitate, people to reach from one point of city to the other, and how people visually connect. To calculate the urban grid evolution and transition of grid planning development, from Late Stone Age to Mohan Jo Daro, at first examined the theoretical backgrounds in linear fashion. In this way grid theory is applied for all studied areas and it was found out that different eras at Mohenjo Daro grid used according to a fixation line to create a mature-agricultural city with the support of mighty Indus River. For Mehar Garh Bolan River was the urban fixation in order to create a semi-agricultural and semi- hunter and gatherer city. The finding about the fixation line of Rakas Jo Rohro it was analysed that how these caves were used and worked. The finding in this regard is that at first it was a hunter and gatherer settlement, which was developed as a result of wind erosion. The people used it as it existed. However; later on human intervention came into practice to the natural cave. It was evident through central spine and the horizontal core of the hill which has acted as a fixation line. In addition one can decipher the city structure as its grid shows the use of this cave through control of an administration in a primitive format. Statistical findings as per space syntax simulations and use of ANOVA Graph 1 depicts the statistical result of space syntax or ‘depth map analysis’, ‘graph analysis’ and ‘attributes explorer’ analysis on six variables that came out of the literature review. Based on these variables every studied context was given numerical values by space syntax software. These values were further analysed via ANOVA software that produced following result of whole research.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

164

Graph 1. Showing the ANOVA result

ANOVA: Results

Group A: Movement of people: Number of items= 5 1.00 2.00 3.00 4.00 5.00 Mean = 3.00 95% confidence interval for Mean: 1.647 thru 4.353 Standard Deviation = 1.58 Hi = 5.00 Low = 1.00 Median = 3.00 Average Absolute Deviation from Median = 1.20

Geometry: Number of items= 5 1.00 2.00 3.00 4.00 5.00 Mean = 3.00 95% confidence interval for Mean: 1.647 thru 4.353 Standard Deviation = 1.58 Hi = 5.00 Low = 1.00 Median = 3.00 Average Absolute Deviation from Median = 1.20

Sanitation Grid: Number of items= 5 1.00 3.00 3.00 4.00 5.00 Mean = 3.20 95% confidence interval for Mean: 1.847 thru 4.553 Standard Deviation = 1.48 Hi = 5.00 Low = 1.00

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

165

Median = 3.00 Average Absolute Deviation from Median = 1.00

Public and Private: Number of items= 5 1.00 3.00 3.00 4.00 5.00 Mean = 3.20 95% confidence interval for Mean: 1.847 thru 4.553 Standard Deviation = 1.48 Hi = 5.00 Low = 1.00 Median = 3.00 Average Absolute Deviation from Median = 1.00

Security: Number of items= 5 1.00 2.00 3.00 4.00 5.00 Mean = 3.00 95% confidence interval for Mean: 1.647 thru 4.353 Standard Deviation = 1.58 Hi = 5.00 Low = 1.00 Median = 3.00 Average Absolute Deviation from Median = 1.20 Visual Connectivity: Number of items= 5 3.00 3.00 4.00 5.00 5.00 Mean = 4.00 95% confidence interval for Mean: 2.647 thru 5.353 Standard Deviation = 1.00 Hi = 5.00 Low = 3.00 Median = 4.00 Average Absolute Deviation from Median = 0.800

Conclusions Conclusively this research shown that the urban grid was used by the early humans around 10,000 years ago in the old cave civilisations, and it qualifies to be called a “City with Urban Grid” and “World’s First Pre Historic Compact City.” Furthermore, the grid was not a coincidental design that suddenly appeared in 2500 BC at Indus Valley, but it was deeply rooted in Ancient Sindh’s urban planning strategies that date to 10,000 years ago. Orthogonal grid planning is utmost important for a compact city development.

References Bulliet, R., Crossley, P., Headrick, D., Hirsch, S., and Johnson, L.. The Earth and Its Peoples, Brief Edition, Volume 1. Chapter 5: The Rise and Spread of Civilization in India c. 2500 BC- 1025 BC Hillier, B. (1996) Space is the machine (Cambridge University Press, Cambridge). Gangal, K.Vahia, M. Adhikari, R (2010) “Spatio-temporal analysis of the Indus urbanization” Current Science Academic Journal 98: 846 -852 Gnisci, Antonio. (2011) ‘Designer Cities, Next Decade’ Projects Today: 47-49 accessed on May 17, 2013 http://www.projectstoday.com/PTSpecialFeatures/PDFData/PT_KGFT_oct11.pdf Higgins, B. H. (2009) “Gridiron 2670 BCE.” In The Grid Book (The MIT Press, Cambridge). Kostof, Spiro. 1993. The City Shaped: Urban Patterns and Meanings Through History. Boston, MA: Bulfinch Rousseau, J. Émile, (1979) Cities are the abyss of the human species. (Les villes sont le gouffre de l’espèce humaine). Book 1, p. 59 originally published in France (1762)

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

166

Rose-Redwood, R. (2011) “Mythologies of the Grid in the Empire City, 1811-2011” The Geographical Review, 101: 396413. Rose-Redwood, R. S. (2008) “Genealogies of the grid: revisiting Stanislawski’s search for the origin of the grid-pattern town” The geographical review 98: 42-58 Samad, U. (2000), Ancient Indus Civilization (Royal Book Company, Karachi). Smith, M. (2007): ‘Form and meaning in the earliest cities: A new approach to ancient urban planning’, Journal of Planning History 6,1:3-47. Snooks, G. D (2002) ‘Uncovering the Laws of Global History’, Social Evolution & History 1: 25–53.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

167

[This page is intentionally left blank]

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

168

PUSHING THE EDGE new technologies and new techniques

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

169

A quantitative approach linking street spatial configuration to street pattern Wu Zhouyan School of Architecture and Urban Planning, Nanjing University, Nanjing, China Email: jewelwzy@gmail.com

Abstract. In urban morphological study, street spatial configuration plays an important role for understanding morphological pattern. For planners and architects the real-time correspondence of the perspective and the plan is essential. After Kevin Lynch, many researchers have sought to quantitative approaches describing and evaluating urban space from human eyeâ&#x20AC;&#x2122;s view. However, the suggestions given by precedent studies for design are general and difficult to be applied into practice. This paper provides the standard grid-matrix as an efficient tool which relates perspective to plan, and achieves the connection between street spatial configuration and street pattern. This research develops ideal models in Rhino and exports the generated perspectives into AutoCAD to establish a standard grid-matrix. A statistic-based analysis is adopted to find the rules linking street spatial configuration to street pattern, and the street spatial configuration and pattern within defined study range are discussed. To demonstrate the feasibility of this quantitative approach, an application on the perspective view of street is conducted. Keywords: street spatial configuration, street pattern, standard grid-matrix, quantitative approach

For both researchers and practitioners how to study urban space is a critical issue because traditional methods are limited. Traditional methods observe the city from the Godâ&#x20AC;&#x2122;s eye perspective. Undoubtedly it is an important and fundamental way to learn urban morphology. Given that the responsibility of researchers and practitioners is not only understanding urban form, but also creating high quality urban space the everyday experience of people, the traditional method has limitations. How to create urban space with specific qualities by operating on urban form is vague. Lynch was the first urban form theorist to bring human cognition into the field of urban space research. He studied the mental images of three cities held by the citizens. His research led to a new perspective of studying the city, in which one should step inside the city using cognition as the tool for assessing the quality of urban space (Lynch, 1960). Lynchâ&#x20AC;&#x2122;s research revealed that traditional ground plan has some limitations in representing a comprehensive urban space. Since then, many researchers and urban planners pay more attention to the perceptions of pedestrians in cities. Kostof (1991, p. 25) proposed two methods for urban space study: the use of the spatial characteristics of the city and these of use of the characteristics of the city in space. The former surveys urban form in a top-down view through urban ground plan and the latter from inside the city through perspectives. However, the former method is highly developed in research and practice, while the latter one is to be explored and hard to use in conducting design. Although practitioners present their projects with a series of plans, they have the image of space in their mind while working through drawings. They are trained to translate the plan into perspective view and envisage perspective view to adjust plan. Consequently, the real-time correspondence of the perspective and the plan matters to them. The interaction is achieved through practical experience and the connection between perspective and plan is qualitative, making the process repetitive and slow.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

170

Ding (2011) has raised the importance of studying images from inside urban space to bridge the gap between urban form and human cognition, and proposed a new approach in urban space representation and research using moving images. Her research has proved the potential of image or photographs being used to link street spatial configuration and street pattern, but the link is represented by a feature line rather than quantitative correspondence. Grounded in Dingâ&#x20AC;&#x2122;s study, the work presented in this paper explores the method using images in depth, and suggests a quantitative way to link street spatial configuration to street pattern conveniently and efficiently. Through a review of recent study on urban space from human eyeâ&#x20AC;&#x2122;s view, quantitative approaches are examined and a new proposition of quantitative connection between street spatial configuration and street pattern is put forward. This research concentrates on the disciplines of perspective, developing a standard grid-matrix which links street spatial configuration to street pattern. It seeks to establish a system for urban planners and architects to fulfill the real-time interaction of perspective and plan.

Review of approaches of studying urban space by cognition The approach that Lynch applied to study urban space is descriptive. Many researchers have since attempted to develop the quantitative research of urban space, representing it through the eyes of observers in different ways. Streetscape study The streetscape study researches visual elements of a street, which combine to form the street's character from the pedestrian view. The research led by Porta and Renne (2005) measured the individual components of the streetscapes taken from two cases for the street indicators through statistical analysis. As a result, the database of street indicators was linked to polygons on maps. Since Porta and Renne focus their research on systematic assessment for designers to measure and compare, street spatial configuration has been disassembled into different components and quantified in values mapped on plan rather than through a physical connection with street pattern. It is difficult for designers to adjust their work instantly according to the evaluation maps of successful streets, strengthening positive attributes, removing the negative attributes, and reflecting them in urban form or urban spatial configuration. Cooper and Oskrochi (2008) have sought to link urban design and visual perception by using the calculation of fractal dimension to gauge the level of visual variety represented in streetscapes. They attempted to assess street visas both in terms of fractality and levels of perceived visual variety, illustrating that there are strong positive associations between them. The study has established a connection between street spatial configuration and visual variety human cognised through the digitalisation of the streetscapes and quantification of the textures. It concentrates on the analysis of streetscapes without considering the pattern corresponding to the different levels of visual variety, which is actually fundamental to urban design. Digitalisation of cognised visually urban space With regard to the relationship between human cognition and urban form, many researchers have addressed visibility analysis in urban design. Since computer technology advanced, approaches in 2D and 3D visual analyses were explored computationally. Isovist is a way of depicting the visibility of urban space in 2D, in relation to which a number of authors (such as Benedikt, 1979; Batty, 2001; Turner et al., 2001) have developed quantifiable analysis of architecture and urban space. However, the 2D approaches have some limits in analysing a complicated urban environment, for they view the city as a top-down object. The research on spherical analysis of urban spatial configuration has extended the study of visibility analysis to the 3D field. The approach proposed by Fisher-Gewirtzman et al. (Fisher-Gewirtzman and Wagner, 2003; Fisher-Gewirtzman et al., 2003; 2005) is more closely related to human perception and they developed the spatial openness index

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

171

(SOI) model to measure the volume of free spaces potentially seen from a given point. This sphericalbased analysis approach achieves the manipulation of changes in human perception of the urban spatial configuration through the change of urban form, and this can help designers in decision-making. The changes are presented through the calculation of SOI, not by perspectives received by visual perception. So the research has not established a physical connection between urban spatial configuration and urban pattern. Based on former studies of isovist, viewshed, spherical approach and Gibsonâ&#x20AC;&#x2122;s theory of direct ecological perception, Yang et al. (2007) have proposed an approach of GIS-based Viewsphere visibility analysis. The viewsphere visibility analysis is a powerful computational tool that describes and analyses urban spatial configuration from the view of human cognition. Like the SOI, it provides supporting information to the urban design decision-making process. It translates the visibility of urban spatial configuration into dataset, mapping it on the plan, but doesnâ&#x20AC;&#x2122;t link urban spatial configuration directly to urban pattern. The suggestions in those approaches given for the design process are general, and they are expressed with data which is abstract and obscure for designers. Therefore a quantitative connection which offers a realtime correspondence of perspective and plan for the designers between street spatial configuration and street pattern is needed. Setting and experiment Ding (2011) argued that filming has transferred 4D (the space-time scenario) to a 2D medium (screen) that can become the research platform. This research attempts to establish a quantitative approach on that platform. The research develops ideal street models in Rhino, which can generate perspectives simulating visual perception. The camera is set at the height of 1.68m, and the perspectives generated are imported into AutoCAD for further operations (Fig. 1).

Figure 1. Development of ideal street models in Rhino Source: Author

Setting of the frame The model space in the computer is boundless, and the sizes of the perspectives generated in Rhino are determined by the sizes of windows. Therefore, a frame defining the study border of street spatial

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

172

configuration is needed. The size of the frame in the research is set as the size of the picture taken in real world. To fit the frame in computer, the crossroad of Zhongshan Road and Hanzhong Road in Nanjing is chosen to capture the same field of view. The camera heights in real world and in Rhino are both 1.68m (Fig. 2). Then the picture taken in real world and the perspective generated in Rhino are imported in AutoCAD and overlapped (Fig. 3). The frame is fixed in computer on street spatial configuration. For the convenience of measurement, the length of frame is set to 600 units in AutoCAD. Studying the image principle The 35mm camera is named after the 35mm film it uses, which is most commonly used for chemical still photography. For a 35mm camera, the most commonly used normal lens is 50mm, which reproduces a field of view that generally looks "natural" to a human observer under normal viewing conditions. The camera used in this research is a Nikon J1 and the focal length of lens adopted is 10mm, which can be converted to 35mm camera with a 27mm lens. According to camera imaging rules, the scope of the picture S1 taken by a 27mm lens has a relation with the one S2 taken by a 50mm lens: a1:a2=50:27, S1:S2=(50:27)2. The field of view taken by a 50mm lens is a normal frame located in centre (Fig. 4).

Figure 2. Perspectives of the crossroad of Nanjing Zhongshan Road and Hanzhong Road in real world and Rhino. The camera heights are both 1.68m. Source: Author

Figure 3. Perspectives imported in AutoCAD to set the frame. Source: Author

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

173

Figure 4. Scopes of the picture S1 taken by a 27mm lens and the picture S2 taken by a 50mm lens. The field of view taken by a 50mm lens is a normal frame to the study frame. Source: Author.

Experiment In order to predict how street spatial configuration changes with street pattern, two groups of ideal street models are developed in Rhino. The first group contains two models with identical built height and different street widths (Fig. 5(a)). Putting the plan together with the perspectives, when the street becomes wider, the intersection point of the skylines and the frame is further apart from the centre, while the intersection point of the street edge and the frame is approaching the horizon. The second group consists of two models with different built heights, and identical street widths (Fig. 5(b)). When the built height rises, the intersection point of the skylines and the frame is approaching the centre, while the intersection point of the street edge and the frame remains unmoved. For further exploration, another group of ideal street models is developed in Rhino, for which built heights are all 24m, and the street widths are from 3m to 100m. Overlapping the perspectives, street spatial configuration presents some rules accordingly to street pattern (Fig. 6). When the street is 3m wide, the skyline converges towards the vertical centre-line. When the street is 100m wide, the street edge converges towards the horizon. The intersection point of the skyline and the frame moves from the upper edge to the vertical edge as the street becomes wider, while the intersection point of the street edge and the frame moves from the lower edge to the vertical edge. The experiments illustrate that there are certain rules to link street spatial configuration to street pattern.

(a) (b) Figure 5. (a) The group made up of two models with identical built height, and different street widths. When the street becomes wider the skylines are further apart from the centre, while the street edges are approaching the horizon. (b) The group made up of two models with different built heights, and identical street width. When the built height rises, the skylines are approaching the centre, while the street edges are left unmoved. Source: Author

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

174

Defining the study range of the street space This research defines the study range of the street space by connecting with the observation point. The observation point is defined along the street centre-line. Human eye has a maximum limitation in vision distance beyond which people could distinguish neither objects nor their shapes. Maertens (in Moughtin 2003, p. 38) suggests that the nasal bone is a critical feature for the perception of the individual. At a distance of about 35m (115ft) the face becomes featureless. Using the analogy of the nasal bone, Maertens also suggests that it is this size that dictates the dimensions of the smallest parts of a building of human scale. We can distinguish people at 12m (40ft) and a person can be recognised at 22.5m (75ft). Body gestures can be discerned at 135m (445ft). This is also the maximum distance that a man can be distinguished from a woman (Moughtin, 2003:38). According to Maertensâ&#x20AC;&#x2122; and Moughtinâ&#x20AC;&#x2122;s respective descriptions, the best distance to observe the details of buildings is from 12m to 35m. To include more spatial configurations to the study range, this research extends the best distance, and the study range is defined as from 10m to 50m away from the observation point. Within this range the variation of street spatial configuration can be subtly described (Fig. 7).

Figure 6. Perspective overlapping the spatial configuration of streets which are from 3m to 100m wide. The change of street spatial configuration shows some rules. Source: Author

Figure 7. Study range of street space. Source: Author

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

175

Development of standard grid-matrix Basic components The standard grid-matrix is composed of study frame and normal frame, horizon HHâ&#x20AC;&#x2122;, vanishing point P and x-y coordinate, among which the horizon lies in the middle of the study frame, and the vanishing point is in the middle of the horizon, centre of the frame. The coordinate origin is in the middle of the lower edge of the study frame (Fig. 8). The spatial configuration in the study frame is divided by x axis and the horizon into four parts: the two beyond the horizon constitute the skyline area, and the two beneath the horizon constitute the street edge area.

Figure 8. Basic components of the standard grid-matrix. Source: Author

Grid and calibration Corresponding to the skyline area and street edge area, there are two grids, one is grid-matrix beyond the horizon and the other is perspective grid beneath the horizon. The research assumes that the ground is flat, so the grid-matrix which represents the width and depth of street space can be projected on the scene (Fig. 9). The calibrations on the frame mark the real width between one-sided street edge and the street centre-line, and the calibrations on the left of y axis present the real depth or, in other words, the perpendicular distance from the observation point to the horizontal line where one point of the street edge is located. The calibrations on the right of y axis are the real height corresponding to the depth, which means the height of the interface within the frame. The street pattern is projected on the scene by a perspective grid. The building height of the street pattern is placed on the grid-matrix beyond the horizon. Considering the identical pattern can generate a variety of skylines by multiple building heights, and one skyline can correspond to different patterns, a medium to project the built height and the width is required. The research develops a grid-matrix at the distance of 100m away from the observation point (Fig. 10). The level calibrations on the frame indicate the real width between one-sided street edge and the street

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

176

centre-line, and the vertical ones indicate the real building height. The skyline intersects with the gridmatrix and the coordinate of the intersection point indicates the width and height.

Figure 9. Perspective grid representing the width and depth of street space. Source: Author

Figure 10. Grid-matrix generated at the distance of 100m Source: Author

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

177

D lines The spatial configuration is delimited by the study frame. If the built height within the frame is the full height of the building, the set of the built heights is defined as the real height area. Otherwise, the set of the built heights is defined as the visual height area. When the skyline coincides with the diagonal, the visual height area vanishes. If the oblique angle of the skyline is less than the one of diagonal, the visual height area never exists. Otherwise, the visual height area appears. Therefore the diagonals in the skyline area are the critical positions beyond which the spatial configuration within the study frame is fully visible. These diagonals are named D lines (Fig. 11). The heights of buildings corresponding to different widths of street space are labeled on D lines.

Figure 11. D lines indicating the critical positions that the visual height area vanishes. Source: Author

E lines The intersection point of the skyline and the frame moves with the alteration of the width and the height. However the changing trend fluctuates, and in some conditions the trend is obscure. To find the critical values, the x-coordinate and y-coordinate are statistically analysed. The values are categorised into two groups. One group records the measured x-coordinates of the skylines with the same width and different heights. The scatter diagram is based on data of each width and different heights (Fig. 12(a)). The other group records the measured y-coordinates of the skylines with different widths and the same height. The scatter diagram is based on data of each height and different widths (Fig. 12(b)). The inflection points of all the scatter diagrams are analysed. The inflection points are located on the grid-matrix and connected by curves. Their values are labeled on the curves. These curves are named E lines (Fig. 13). The grey areas identify the conditions that the changing trend of the skylines is obscure. In addition, the ranges of the distance between the street edge and street centre-line corresponding to different heights of the buildings are listed. Within the range the skyline changes noticeably. For instance, if the built height is 3m, and the building is 1.5m to 14.1m away from the street centre-line, the variation of skylines is more visible.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

178

(a) (b) Figure 12. Examples of scatter diagram analysis. The circle indicates the inflection point. (a) Scatter diagram based on data of each width and different heights. (b) Scatter diagram based on data of different widths and the same height. Source: Author

Figure 13. E lines indicating the critical pairs of real width and built height. The grey areas identify the conditions that the changing trend of the skylines is obscure. The ranges of real width W corresponding to different height H of the buildings: W=3m, 1.5m<h<14.1m; W=6m, 1.5m<h<25.5m; W=9m, 1.5m<h<33.1m; W=12m, 1.5m<h<39.4m; W=15m, 1.5m<h<44.7m; W=18m, 1.5m<h<49.5m; W=21m, 1.8m<h<53.8m; W=24m, 2.3m<h<57.9m; W=27m, 3.0m<h<61.6m; W=30m, 3.7m<h<65.2m; W=33m, 4.5m<h<68.6m; W=36m, 5.4m<h<71.8m. Source: Author

Overlapping the basic components, grid and calibration, D lines and E lines, the standard grid-matrix is achieved (Fig. 14). It is a tool assessing the urban space as well as designing. If the observation point is not along the street centre-line, the standard grid-matrix can be modified according to descriptive geometry. The vanishing point maintains in the centre of the study frame, while the positions of calibrations are adjusted.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

179

Figure 14. Standard grid-matrix. Source: Author

Linking street spatial configuration to street pattern Statistic-based analysis linking street spatial configuration to street pattern The construction of standard grid-matrix realised the link between street spatial configuration and street pattern. To enable the link more specifically, an analysis based on the geometrical relationship is applied. The standard grid-matrix can be used to measure the street spatial configuration, and feedbacks four values: the real built height Hi, the real width between the street edge and street centre-line Wi and the xcoordinates of the endpoints Qi1, Qi2 of skyline Li1, L i2. This research assumes the angle between the street edge and the horizon to be Î˛, and the measured y-coordinates of the endpoint s Qi3, Qi4 of street edge to be y i1, y i2. The measured coordinates of vanishing point P is (0, 200.86). Extend the street edge to intersect with x axis and measure the x-coordinate of the intersection point Mi (Fig. 15). According to the descriptive geometry, the measured x-coordinate is proportional to real width. Hence the measured coordinates of point Mi can be expressed as:

The length of each grid is 5.41 units, so the measured x-coordinate of Qi4 is 5.41 Li1, and the measured xcoordinate of Qi3 is 5.41 Li2.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

180

Figure 15. Geometrical relationship to link street spatial configuration to street pattern. Source: Author

Since the perspective grid is an inhomogeneous grid, the values of y i1 and y i2 are not proportional to the real depth. Through look-up table methods, the real depths Di1 and Di2 corresponding to y i1 and y i2 are ascertained. To study the relationship between Li, Wi, and Di, four normals are developed, dividing the street spatial configuration in five areas. The forms of the normals are as follows:

Two groups of data are generated for statistical analysis. The first group consists of three subgroups whose data is generated from different normals, while the x-coordinates of the endpoints of skylines are the same: Li1=20, Li2=25. A scatter diagram analysis is used to find the relationship. The scatter diagram based on data from normal B shows that: Di1=5 Wi-0.05, Di2=4 Wi-0.02 (Fig. 16 (a)).The scatter diagram based on data from normal C shows that: Di1=5 Wi-0.08, Di2=4 Wi-0.08 (Fig. 16 (b)).The scatter diagram based on data from normal D shows that: Di1=5 Wi-0.05, Di2=4 Wi-0.06 (Fig. 16 (c)). In view of the data error, the subtrahends in the equations are negligible. This research assumes that if the x-coordinate of the endpoint of skyline remains unchanged, while the skyline moves, the relationship between real width Wi and the real depth Di maintains. The data of the second group is generated from normal B, and the x-coordinates of the endpoints of skylines range from 5 to 30. When Li1=5, the equation drawn from the scatter diagram is Di1=20Wi-0.1. When Li2=10, the equation drawn from the scatter diagram is Di2=10Wi+4Ă&#x2014;10-13. When Li3=15, the

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

181

equation drawn from the scatter diagram is Di3=20/3Wi-0.0352. When Li4=20, the equation drawn from the scatter diagram is Di4=5Wi-0.05. When Li5=25, the equation drawn from the scatter diagram is Di5=4Wi-0.02. When Li6=30, the equation drawn from the scatter diagram is Di6=10/3Wi-0.1 (Fig. 17). Likewise, the subtrahends and addends in the equations are negligible. This research assumes that if the x-coordinate of the endpoint of skyline changes, while the skyline remains motionless, the relationship between real width Wi and the real depth Di changes respectively. The relationship can be explained as:

Any point on a buildingâ&#x20AC;&#x2122;s skyline can be projected on plan, and the distance from the horizontal line of projected point to the horizontal line of the observation point Di can be calculated as 100 times the ratio of real width Wi to the x-coordinate of that point Li.

(a) (b) (c) Figure 16. (a) Scatter diagram based on data from normal B. (b) Scatter diagram based on data from normal C. (c) Scatter diagram based on data from normal D. Source: Author

Figure 17. Scatter diagram based on data from normal B. The Li varies from 5m to 30m. Source: Author

The standard grid-matrix can be applied to link street spatial configuration to street pattern by overlapping it on photographs capturing the view of street. If the built height H1 is known, the real width can be read on standard grid-matrix, and the building is located on a line which is W1m away from the street centre-line on the plan. The x-coordinates of the endpoints of skyline L1, L2 can also be read on standard grid-matrix. According to equation (11) the building is projected on the plan (Fig. 18). Applying the quantitative approach to the view perceived by the pedestrian in the street, the skylines can be projected on the plan (Fig. 19).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

182

Figure 18. Transformation from street spatial configuration to street pattern. Source: Author

(a) (b) Figure 19. Transformation from (a) Skylines perceived by pedestrian in the street to (b) street pattern. Source: Author

Street spatial configuration and street pattern within defined study range The normals have divided street spatial configuration into five areas. Though the normals remain motionless, the position of the normal projected on the plan changes according to the built height. The grey area illustrates the study range. On standard grid-matrix it presents the area of skylines visible within the study range, and this area is supposed to be the more active region of changing in spatial configuration. The relationship between the real built height Hi and the real widths between the normals located on plan and street centre-line can be explained as:

When the built height Hi is 3m, the widths between the normals located on plan and street centre-line are

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

183

as follows: WA=0.7, WB=1.3, WC=2.0, WD=4.0. The lower edge of the grey area on spatial configuration is located at the calibration of 4.3m (Fig. 20(a)). When the built height Hi is 6m, the widths between the normals located on plan and street centre-line are as follows: WA=2.2, WB=4.3, WC=6.5, WD=13.0. The lower edge of the grey area on spatial configuration is located at the calibration of 10.3m (Fig. 20(b)). When the built height Hi is 9m, the widths between the normals located on plan and street centre-line are as follows: WA=3.7, WB=7.3, WC=11.0, WD=22.0. The lower edge of the grey area on spatial configuration is located at the calibration of 16.3m (Fig. 20(c)). When the built height Hi is 12m, normal D exceeds the study frame, and the rest widths between the normals located on plan and street centre-line are as follows: WA=5.2, WB=10.3, WC=15.5. The lower edge of the grey area on spatial configuration is located at the calibration of 22.3m (Fig. 20(d)). When the built height Hi is 15m, normal D exceeds the study frame, and the widths between the rest normals located on plan and street centre-line are as follows: WA=6.7, WB=13.3, WC=20.0. The lower edge of the grey area on spatial configuration is located at the calibration of 28.3m (Fig. 20 (e)). When the built height Hi is 18m, normal D exceeds the study frame, and the widths between the rest normals located on plan and street centre-line are as follows: WA=8.2, WB=16.3, WC=24.5. The lower edge of the grey area on spatial configuration is located at the calibration of 34.3m (Fig. 20 (f)).

(a)

(b)

(c)

(d) (e) (f) Figure 20. Separated areas divided by normals corresponding to different built heights within study range. Source: Author

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

184

When the built height Hi is 21m, normal C and D exceed the study frame, and the skylines also go beyond the study frame. The calibration of the lower edge of the grey area on spatial configuration can be described as 2Hi-1.7. When Hi is larger than 10.9m, the normal D exceeds the study frame. When Hi is larger than 20.2m, the normal C and D exceed the study frame, and the skylines also go beyond the study frame.

Conclusion Former quantitative approaches have provided diversified methodologies for studying urban space from the human eye’s view, describing and evaluating the built environment. However, the connection between street spatial configuration and street pattern is not involved. The immediate feedback from perspective to plan and from plan to perspective is important for designers. In this paper a standard grid-matrix is developed to link street spatial configuration to street pattern, and an efficient connection between perspective and plan is achieved. Results observed suggest that for a given skyline the distance from the width between the street edge and street centre-line Wi and the horizontal line of projected point to the horizontal line of the observation point Di are proportional to the built height Hi. If Hi is larger than 20.2m, the skylines will exceed the study frame. The normals on spatial configuration divide the study range of street space into different areas which may lead to different levels of variability, and will be further investigated. The application to the perspective view manifests the feasibility of this quantitative approach as a practical tool to study urban space and assist design. However, this research developed the approach on the basis of ideal models in Rhino. Urban space in the real world is complex and many problems such as irregularity and discontinuity exist. Therefore, further work is needed to refine the grid-matrix. For example, the change tendency of spatial configuration on the frame while moving around a curve will be investigated and added to the grid-matrix. In addition, the study range will be further examined. This study is preliminary work for research on urban space using moving images, which is an effective tool to record and represent urban space. The effectiveness of this grid-matrix applied on moving images will be tested and verified. The standard grid-matrix can be the first step to develop a multifunctional tool to aid architecture and urban research as well as design.

Acknowledgement This thesis is supported by the National Natural Science Foundation of China (Program No. 51078177) and the Natural Science Foundation of Jiangsu (Program No. BK2009244), under the supervision of Professor Wowo Ding.

References Benedikt, M. L. (1979) ‘To take hold of space: isovists and isovist fields’, Environment and Planning B 6, 47-65. Batty, M. (2001) ‘Exploring isovist fields: space and shape in architectural and urban morphology’, Environment and Planning B: Planning and Design 28, 123-150. Cooper, J. and Oskrochi, R. (2008) ‘Fractal analysis of street vistas: a potential tool for assessing levels of visual variety in everyday street scenes’, Environment and Planning B: Planning and Design 35, 349-363. Ding, W. (2011) ‘Mapping Urban Spaces: Moving Image as a Research Tool’, Urban cinematics: understanding urban phenomena through the moving image, Edi. Penz, F. and Lu, A., Intelrlect Press, 315-355. Fisher-Gewirtzman, D. and Wagner, I. A. (2003) ‘Spatial openness as a practical metric for evaluating built-up environments’, Environment and Planning B: Planning and Design 30, 37-49. Fisher-Gewirtzman, D., Burt, M. and Tzamir, Y. (2003) ‘A 3-Dvisualmethod for comparative evaluation of dense builtup environments’, Environment and Planning B: Planning and Design 30, 575-587. Fisher-Gewirtzman, D., Pinsly, D. S., Wagner, I. A. and Burt, M. (2005) ‘View-oriented three-dimensional visual analysis models for the urban environment’, Urban Design International 10, 23-37.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

185

Kostof, S. (1991) The city shaped: urban pattern and meanings through history (A Bulfinch Press Book, Boston, Toronto, London). Lynch, K. (1960) The image of the city (The M.I.T. Press, Boston). Maertens, H. (1884) The Optical Scale in the Plastic Arts (Wasmuth, Berlin), cited by Moughtin C. in ‘Urban Design: Street and Square’, Architectural Press, Oxford, 2003. Moughtin, C. (2003) Urban Design: Street and Square (Architectural Press, Oxford). Porta, S. and Renne, J. L. (2005) ‘Linking urban design to sustainability: formal indicators of social urban sustainability field research in Perth, Western Australia’, URBAN DESIGN International 10, 52-64. Turner, A., Doxa, M., O'Sullivan, D. and Penn, A. (2001) ‘From isovists to visibility graphs: a methodology for the analysis of architectural space’, Environment and Planning B: Planning and Design 28, 103-121. Yang, P. P.-J., Putra, S. Y. and Chaerani, M. (2007) ‘Computing the sense of time in urban physical environment’, URBAN DESIGN International 12, 115-29.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

186

Shape grammar as a method for describing order of Chinese traditional urban pattern Guoxin Ke School of Architecture and Urban Planning, Nanjing University, Nanjing, China Email: ark89@qq.com

Abstract. Most studies of Chinese traditional urban fabric focus on either the description of street pattern and geometrical shape system or the type of courtyard houses. This paper investigates the morphological order of urban pattern through the use of shape grammar, which is particularly useful in unveiling urban morphological complexity, as demonstrated by many contemporary case studies. This paper focuses on the intrinsic mechanisms that synthesise the dimensions of urban fabric and spatial structure. Based on case studies of Beijing and Nanjing, representatives of traditional cities, this paper will summarise the shape grammar for the organisation of streets and the generative principles of the patterns of Chinese cities. It intends to develop a descriptive language for Chinese urban pattern and spatial structure, and to reveal the city as an organic whole with potential rules. Keywords: Shape grammar, urban pattern, urban fabric

Cities are a product and physical vector of human activities reflected in multiple aspects, such as political, social structure and economy. It might be said that the gamut of human civilisation is condensed within the urban (Karl, 2009, p. 105). As a physical form affected by different factors, urban patterns and spatial structures tend to be diverse and complex. The differences also make cities recognisable and distinguishable. The pattern of the city is the way different functions and elements of the settlement form are distributed and mixed together spatially (Lynch, 1981, p. 265). Different cities may perform variant physical space and image via different mechanisms and organisation of different urban patterns, while some others are not. Demolition and renewal are always taking place in cities likened to the metabolism of organisms. The city form and fabric are the sedimentary deposits of time. Historical cities had the capacity to absorb successive transformations without losing their essential structure (Serge, 2012, p. 56). According to Stephen (2005), Jacobs saw streets as the lifeblood of cities rather than mere traffic channels and Alexander saw streets as multi-functional urban â&#x20AC;&#x2DC;patternsâ&#x20AC;&#x2122;. This paper pays attention to the street pattern and the street layout of Chinese traditional cities.

Background Situation of Chinese historical urban districts Since the late Qing Dynasty, Western planning theories and thoughts have entered China and been applied to city planning. In the last decades, Chinese citiesâ&#x20AC;&#x2122; spatial structure and urban fabric have been widely changed, especially the traditional urban districts. Since the reform and opening-up policy in 1978, Chinese cities have undergone dramatically growth and transformation which are described as urbanisation. What has been created seems to be much different from the traditional urban patterns of China. Although presenting an opportunity to feed back to economy development and population growth, urbanisation has caused enormous controversy, unordered expansion of urban edge, ruin of historical urban fabric, unrecognisable street space, and urban heat island effect and so on.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

187

Now urban planning and design play an important role in controlling and guiding urban morphology. Urban planning and design propose code of urban, arrangement of function and quantities of building for the smooth running of city. As UN-Habitat has emphasised good urban planning and design should establish minimum common space, optimised street connectivity. However, it seems that modern planned urban environments are usually perceived to be worse than traditional ‘unplanned’ ones (Stephen, 2009). As a response to Modernism, movements such as sustainability and neo-traditional urbanism have replaced the rhetoric of the ‘motor age’. Compact, dense, mixed-use neighbourhoods are back in fashion, with a new breed of traditional-style buildings and street patterns to choose from (Stephen, 2005, p.9). At present, the protection and renaissance of historical districts have caused more and more attention. According to the reports, historic cities are abundant in cultural heritage and intangible cultural heritage in China (UN-Habitat Draft Working Paper, January 2011). However many urban areas in China are criticised for being monotonous and lacking locality. New city construction and protection of old city are at unbalanced state with different inherent spatial and physical features. Related research on cities The development of human civilisation are always accompanying with evolution of human settlements. Cities as a high-grade product have presented a complex system which naturally attracts broadly attentions compared with towns, villages. Investigations on urban form are carried out from various views and methods by scholars from different fields. Urban morphology was proposed by Schluter (1899). With the historical-geographical approach, town plan analysis is emphasised with the definition of constituent elements and evolution trough time (Conzen, 1969; Whitehand, 1992). At the same time, some researchers have focused on typology with the summarisation of building type and open space (Christopher, 1977; Aldo, 1982; Robert, 1979). With the rising of computer language, scholars have explored some new descriptive and generative languages to underlying the intrinsic mechanisms of cities (Bill, 1984; James, 1975; Alexandra, 2011).

Shape grammar as a method Since shape grammar was adopted as a descriptive method in 1970’s (Stiny and Gips, 1972); it has been widely used as a descriptive and generative method for architecture and urban study. A shape grammar is a formal set of rules applied to shapes to generate a language of design that allows the visualisation of the desired form and function of the rules (Thomas, 2007, pp. 79-102). Shape grammar has great potentials of operating shapes via designated orders or logics. Shape grammar contributes to the development of urban analysis on the basis of two main tendencies. First, shape grammar could be used as an analysis language for existing design. Li (2001, 2003) tried to analyse the organisation and structure of the traditional Chinese wood building via shape grammar. Consequently, a parametric shape grammar that generates the ground plans of famous architectures was developed (Stiny and Mitchell, 1978, pp. 5-18, 1980, p. 209-226). Second, shape grammar is also a design language. It has been used in visual areas: in painting and sculpture, in engineering and product design (Stiny and Gips, 1972; Agarwal and Cagan, 1998). Now western scholars apply shape grammar as a method for unveiling the order of European urban morphological complexity and as an approach to solve urban design problems. Urban grammar, originated from shape grammar, has been applied to encode the generation of urban form (Beirão, Duarte and Stouffs, 2009). The design generation involves the application of rules codifying typical urban design operations into shape grammar in such way as to match requirements specified in urban programs formulated using a description grammar (Beirão, Duarte and Stouffs, 2009). With the development of computer language, scholars have explored some new descriptive and generative languages that apply to the intrinsic mechanisms of cities. Cellular Automata (CA) is an effective method to simulate complex structure of self-organisation and evolutionary process with simple rules and programs. It tries to compute physical form in time and space with repeated applications of the same rules upon each cell. As with shape grammar, CA is a method to organise particular patterns and it

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

188

prefers to be carried out with a single rule recursively, while shape grammar usually consists of a set of rules. A major problem with the CA approach has been that discovering self-generating rules for a particular intent out of an almost infinite CA rule space is beyond human capability (Thomas, 2007). Space syntax is a quantitative description of human living space about architecture, settlements and urban provided by Hiller (1984). With the combination of social activities and space configuration, space syntax has been developed as a model and approach to describe and research different space orders of cities. The complex relations between different space elements of the whole are the basis of social and economic activities, not the independent space element. Shape grammar appears to be close to the notion of space syntax. However, syntactic generators are simpler than shape grammars. Moreover, they are shape free (Hiller, 1984).

Treatises and theories on traditional Chinese cities In dynastic China, the city planning and building norms were distinctively ruled by culture and politics. Traditional Chinese architectural form and urban setting were fundamentally influenced by Chinese cosmology and social hierarchy, which were embodied in the well-known Confucianism, Daoism and Fengshui (Fei Chen, 2008). Chinese building codes and planning reflect regional features. There are two main sets of ancient Chinese urban planning thoughts and works. One is Kaogongji, Rites of Zhou (Zhou Li) which described the fundamental principles for constructing a capital city and shaped most of Chinese cities, especially capital cities. Though there are few cities in China that fit Kaogongji exactly, its planning rules have been embodied in many cities, like the city Dadu, throughout Chinaâ&#x20AC;&#x2122;s long history. Guanzi is the other important work which advocates a native philosophy that city construction should rely on and take advantage of the natural conditions rather than the hierarchy of feudal society. Its principles are quite different from Kaogongji. Fengshui is an important factor for locating cities for the planners. The cities in the area of South China are affected by mountains and rivers. Usually the outline of these cities is irregular with non-orthogonal roads. Though not valued by the ancients, the principles of Guanzi have affected many cities directly or indirectly and deserve reference. Because of planning principles, methods and art configuration laws, traditional architecture and urban contexts could continuously develop. They eliminate the stale and bring forth the fresh, while retaining the independence and continuity of the Chinese architecture system (Fu Xinian, 2001; Wu Liangyong, 1994). In the housing system, streets and lanes system or urban structure system, it is possible to distinguish cities from each other while finding similarities between them.

A grammar description based on architecture Urban samples Both of Beijing and Nanjing are famous historical cities and capitals during various dynasties. In traditional districts, the urban form and fabric has been more or less retained. Beijing is a typical city of north China. The Beijing old city in Qing dynasty adopted the city pattern of Ming dynasty which was founded at the base of Yuan Dadu. The traditional district of Beijing consists of the Palace City, the Imperial City, the Capital and the outer. The Imperial City is located at the centre of Beijing. The size of the block was planned in modules, the blocks are organised by hutong, and the basic form of buildings in Beijing is the quadrangle courtyard. The main buildings are orientated to the south, and access is designed accordingly. Nanjing is a typical city of South China. The form of the city is greatly affected by the mountain and the river. Traditional residences are mainly amassed in the Nanjing Customs South. The streets generally accord with the river. The basic building form of Nanjing is courtyard house, distinctly different from Beijingâ&#x20AC;&#x2122;s quadrangle courtyard. This paper selects two samples from Beijing and Nanjing. The sample of Beijing is Nan Luoguxiang heritage area (Fig. 1). Nan Luoguxiang is one of the oldest heritage districts in Beijing. It was founded in Yuan dynasty 700 years ago. It is located at the north of Beijing old city and is one of the most important

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

189

residential districts in Beijing with many former residences of celebrities. With historical change, Nan Luoguxiang still keeps the structure of street called hutong and a large mass of traditional quadrangle courtyard. In this sample, the blocks are closer in size and shape. The transmeridional length of blocks is longer than the north-south ones. According to Yuan Shi (History of the Yuan Dynasty), a standard block size was 80mu (the unit used to measure area of land, 1mu=666m2). The street forms are reserved almost intact. The hierarchy of streets is comprised of three streets in this urban sample, including main streets, transverse streets and minor streets. The main streets are much wider than transverse streets and minor streets. The superblock is enclosed by the main streets without interference from traffic. The transverse streets are parallel and mostly north-south. The width of transverse streets is similar in the range of 4m to 8m. The distance between the centre lines of two adjoining and parallel transverse streets is about 75m. This data corresponds to the research that the distance between two adjoining hutong is 50bu (bu is also a unit used to measure length of land in ancient times, 1bu=1.54m) in Yuan dynasty (Zhao Zhengzhi, 1972). The minor streets are mostly dead end roads and orthogonal to transverse streets. The plots of the samples seem to be irregular and not of uniform size. The plots near the north-south transverse streets and corners are smaller and intensive. Usually the depth and width of plots are related to the organisation of the courtyard house. The Beijing quadrangle courtyard house is a typical house pattern which contains a courtyard at the centre with houses at two, three or four sides (Fig. 2).

Figure 1. Location of Samples. Left: Beijing. Right: Nanjing Source: Drawn by author

The sample of Nanjing is the Nanjing Customs South in the south of Nanjing old city (Fig. 1). The Nanjing Customs South has long historical standing, since Ming Dynasty as a residential and household economy and handicraft workshops district. The Qinhuai River runs through it. The structure of the Nanjing traditional district is made up of road network and river system. The shape and size of the blocks are diverse. The hierarchy of streets in Nanjing is comprised of four types of streets, including main streets, transverse streets, minor streets and mews lanes. It includes an additional street in the hierarchy compared to Beijing. The streets are not east-west or south-north orthogonal. The length and width ratio of plots in the sample is larger than that of Beijing. The traditional residence in Nanjing is mostly the courtyards with high walls and grand houses. The shape of patios is nearly a square but the specific layout is flexible and reasonable (Fig. 2).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

190

Block

Street

Plot

Building The maps of Beijing are redrawn according to the cad finished in 2003. The maps of Nanjing are redrawn according to the cadastre in 1936.

Figure 2. Left: Beijing. Right: Nanjing Source: Drawn by author

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

191

Comparison between samples of Beijing and Nanjing According to the sample of Beijing, the hierarchies of streets are clear. Streets meet at right angle. There are few crossroads between the main road and the minor road. While in Nanjing, the streets are laid out in various directions according to the natural factors. Streets at different levels in the hierarchy join together with T or X junctions. Abstract street pattern diagrams of Beijing and Nanjing are summarised in Fig. 3. Top Beijing, The hierarchy of the streets is clear. Streets meet at right angle. Mainly grid with crossroads and continuity of cross routes.

Bottom Nanjing, The streets go in all directions and mix T and Xjunctions. There is no access constraint between ‘major’ and ‘minor’ types.

Figure 3. An abstract diagram of the street patterns in Beijing and Nanjing

As we can see from the samples, relationships between the boundary of the blocks and the street hierarchy differ in Beijing and Nanjing (Fig. 4). The blocks of Beijing are defined by main streets and transverse streets, while the blocks of Nanjing are also defined by the minor street except for main streets and transverse streets. Blocks are subdivided into plots for the specific land use, and the relation between the plots and streets is important to bridge the gap between different dimensions of urban form. The access of a plot is a significant sign to reflect the relation between plots and streets. So, referring to a Soso map, the access of Beijing plots is mapped (Fig. 5). The plot access of the Nanjing sample is identified based on courtyard type, research by others and the cadastre of Nanjing in 1936 (Fig. 6). Based on the mapping of access of the samples, this paper uses a soft dot to represent a plot, and filled dots to represent access of the plot. Linking the soft dots and relevant filled dots, it is found that the are no more than two kinds of relationships between plot access and streets in the hierarchy. The streets of Beijing have a three-level hierarchy, while the streets of Nanjing have four-level hierarchy. People use different paths going through the main street into the plots. According to shape summary of streets with three levels and four levels, there are 10 kinds of paths in Beijing while 28 kinds of paths in Nanjing via permutation and combination. However, there are limited existing relation types in the samples of Beijing and Nanjing. According to the space relation map of the samples, the plots in Beijing usually do not have access to main streets and the mews lanes at the same time, and one plot usually does not connect with two mews lanes at the same time (Fig. 7).

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

192

Figure 4. The relation between blocks and different hierarchy streets. Top: The blocks of Beijing are defined by Main Streets and transverse streets. Bottom: The blocks of Nanjing are also defined by the minor street except for main streets and transverse streets. Source: Drawn by author

The courtyard is an indispensable element of open space in Chinese urban fabric. At the same time, it plays an important role in internal traffic and connecting the inside with outside. There are two cases about the connections between courtyards and streets. One case is that the courtyard is facing the street without any houses. The other case is that people need to go through a house to access the courtyard and other houses. It is the same with the connections between courtyard and courtyard in one plot. This paper tries to draw a permeability map of open space about Beijing and Nanjing samples. The triangle represents no direct connection between street and yard or yard and yard. Squareness represents direct connection between two open space elements (Fig. 8).

Figure 5. Urban sample of Beijing. The positions of plot entrance were affirmed in Soso Map Source: Drawn by author

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

Figure 6. Urban sample of Nanjing. The positions of plot entrance are affirmed according to courtyard house types and the cadastre of Nanjing in 1936. Source: Drawn by author

193

Figure 7. Shape summary of streets with different levels and the space relation map of the samples about plots, accesses, and streets. Top: Beijing. Bottom: Nanjing. Soft dot means a plot, and filled dot means the access Source: Drawn by author

Figure 8. The permeability map of open space about the samples of Beijing and Nanjing Source: Drawn by author

Rules summarisation According to analytical decoding on the samples, this paper tries to summarised a set of rules about the relationships and generations of Chinese traditional urban with two categories and four elements. These are rules for the blocks and street pattern and rules for the plot pattern and courtyards. With the different initial conditions, the process of applying a rule in shape grammar may be various. Each rule contains an initial type of shape with its functional size, label to control the operation, and opposite shape. The size and shape of blocks are closely related to the street pattern. Rule s(x) conducts the generation of streets pattern and block pattern.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

194

The rules of shape grammar can be shown as follows. Rule s4 is applying for generating parallel streets of the same hierarchy, while Rule s2, Rule s3 generates parallel streets of different hierarchies. Rule s1 and Rule s5 to Rule s10 are for intersecting streets. Because of the difference in urban pattern and spatial structure between Beijing and Nanjing, the suitable range of generation rules is different. The rules in the limits of the dotted line are only fit for Nanjing. During the analysis of the samples, the access is a valuable element to reflect the plot pattern. Rule p1 and rule p2 deal with the position of accesses of plots. With the rules for locating the access, the blocks are subdivided into primary plots (rule p3 - rule p6). Rule p7 to rule p11 deal with plot pattern. Usually a plot evolves over time. It may extend, contract, subdivide, and indent and so on. Rule p6 incises part of a plot into another. Rule p7, rule p8 and rule p11 fill the gap around the plot. Rule p9 create an open space in front of the plot. Rule p10 combine two little plot into a larger one. Rule c(x) is applied for the open space inside plots. Rule c1 to Rule c4 deal with the combination between the plot and street, while rule c5 to rule c9 for the combination between two plots (Fig. 9). Whole rules are not necessary for generation. Some rules may be omitted while some rules may be applied recursively. With different initial conditions and different sequences, the consequence of generation may be different (Fig. 10 and Fig. 11).

Conclusions With the summarisation of urban pattern and spatial structure, urban elements are described by a different range and order of rules. The different suitable range of rules reflects the differences under intrinsic mechanisms. Shape grammar as a method to describe different urban patterns of Chinese traditional cities is viable. With the analysis of samples from treaties and historical materials, the differences between the two samples are presented at different scales and levels. As an organic whole, the city is built following laws of nature and human will. This paper has analysed samples from four factors: block, street, plot and building/courtyard. These four factors reflect different urban characteristics and they are closely associated with each other. With comparisons of the maps of Beijing and Nanjing on the four factors, the similarities and differences between them are proposed. Urban morphological characteristics have their principles. As can be seen from the maps of samples, the layout of blocks has a strong relation with the organisation of street pattern. Beijing and Nanjing have different hierarchies on streets and different nodes of streets. Therefore the shape and size of blocks presented are various. The blocks of Beijing are similar, while Nanjingâ&#x20AC;&#x2122;s blocks are not. With certain street patterns and geometrical features of blocks, specific grammars are defined. The access of a plot is a significant sign reflecting the relation between plots and streets. It could both reflect the different paths people have taken to go through the streets into the plots and the spatial relations between two independent open spaces. The access is used to define plot pattern. But it also has limits in dimensions and fortuities. The rules presented still require further study. Some may be changed and finer rules may be added.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

195

Figure 9. Rules for the plot pattern and courtyards Source: Drawn by author

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

196

Figure 10. Derivation sequence for Beijing Source: Drawn by author

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

197

Figure 11. Derivation sequence for Nanjing Source: Drawn by author

Urban Form at the Edge: Proceedings of ISUF 2013, Volume 2

198

References Agarwal, M and Cagan (1998) ‘A blend of different tastes: the language of coffeemakers’. Environment and Planning B25, 205-226. Alexander, C. (1977) A Pattern Language (Oxford University Press, New York). Beirão, José and Duarte José(2005) ‘Urban Grammars: Towards Flexible Urban Design’. eCAADe 23, 491-500. Beirão, José, Duarte José and Stouffs Rudi (2009) ‘Grammars of designs and grammars for designing - grammar-based patterns for urban design’. Joining Languages, Cultures and Visions: CAAD Futures. Conzen, M.R.G (1969) Alanwick, Northumberland: a study in town-plan analysis (Institute of British Geographers, London). Fei Chen (2008) ‘Typo-morphology and the crisis of Chinese Cities’. Urban Morphology 12(2), 131-142. Fu Xinian (2001) The research on the methods of Chinese traditional urban planning, layout of cluster and architecture design (China Architecture & Building Press, Beijing). Gips , J(1975) Shape grammars and their uses – artificial perception, shape generation and computer aesthetics (Birkhauser Verlag, Basel und Stuttgart). He Yeju (1996) History of Chinese Ancient Urban Planning (China Architecture & Building Press, Beijing). Hillier, B. (1984) The Social Logic of Architecture (Oxford University Press, New York). Kehew, R. (2011) ‘Urban Patterns for Sustainable Development: Towards a Green Economy’, UN-Habitat Draft Working Paper 2011. Krier, R. (1979) Urban Space (Rizzoli International Publications, Michigan). Kropf, K. (2009) ‘Aspects of urban form’. Urban morphology 13(2), 105. Li, A.I. (2001) A shape grammar for teaching the architectural style of the Yingzao Fashi. PhD dissertation, Massachusetts Institute of Technology. Li, A.I. (2003) ‘The Yingzao Fashi in the information age’. The Beaux-Arts, Paul Philippe Cret, and Twentieth Century Architecture in China, 3-5. Lynch, K. (1981) A theory of good city form (The MIT Press, Cambridge). Marshall, S. (2005) Streets & Patterns (Spon Press, Oxon). Marshall, S. (2009) Cities design & Evolution (Routledge Chapman & Hall, New York). Paio. A. (2011) ‘An Urban Grammar Study A Geometric Method for Generating Plan metric Proportional and Symmetrical Systems’. Nexus Network Journal 13, 151-169. Rossi, A. (1982) The Architecture of the City (The MIT Press, Cambridge). Salat, S. (2012) City and Forms on Sustainable Urbanism (China Architecture & Building Press, Beijing). Salat, S. (2012) ‘Systemic resilience of complex urban systems’. TeMA Journal of Land Use, Mobility and Environment 5 (2), 56. Speller, T.H. (2007) ‘Using shape grammar to derive cellular automata rule patterns’. Complex systems 17, 79-102. Stiny, G. and Gips, J. (1972) ‘Shape grammars and the generative specification of painting and sculpture’. Best computer papers of 1971. Philadelphia: Auerbach, 125-135. Stiny, G. and Mitchell, W. J. (1978) ‘The Palladian grammar’. Environment and Planning B 5, 5-18. Stiny, G. and Mitchell, W. J. (1980) ‘The grammar of paradise: on the generation of Mughul Gardens’. Environment and Planning B 7, 209-226. Whitehand (1992) ‘Recent advances in urban morphology’. Urban Studies 29(3/4), 619-636. Wu Liangyong (1994) Beijing old city and Juer Hutong (China Architecture & Building Press, Beijing). Zhao Zhengzhi (1972) ‘the research on recovering Yuan Dadu planning. Essays of the Science History 1972-2, 14-27.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume II

199

Biographies Luisa Bravo Queensland University of Technology, Australia Luisa Bravo is an architectural engineer and an academic scholar, educated in Italy, UK and France. After completing her PhD, with a thesis on contemporary urbanism, at University of Bologna in Italy (2008), she has been researching, teaching and lecturing in several Universities, in Italy, UK, Spain, Portugal, Cyprus, Lebanon, USA, Hong Kong, Taiwan, Japan and Australia. She is currently Adjunct Associate Professor at the Queensland University of Technology in Australia and Adjunct Professor in Urban Planning and Design at University of Florence in Italy. Luisa is charter member and president of City Space Architecture, a non-profit organization based in Italy that performs as a multidisciplinary platform for scholars, professionals, artists and citizens engaged in architecture, public space, cities and urbanity. Monojit Chakma Queensland University of Technology, Australia Monojit (Mickey) Chakma is Assistant Professor at Primeasia University, Dhaka, Bangladesh. He has a Bachelor of Architecture from Bangladesh University of Eng. & Tech; a master of Urban and Regional Planning from the University of Sydney and a Master of Applied Science (Research) from the Queensland University of Technology. His research focuses on urban planning, architecture and in general urbanism, with particular interest in the urban form of traditional villages and settlements in Bangladesh. He is recipient of the Australian Leadership Award (2011) and Australian Development Scholarship (2007). Wowo Ding Nanjing University, China Professor Wowo Ding, Dean of School of Architecture and Urban Planning at Nanjing University, completed the master of Architecture from the Southeast University of China and gained her Doctor of Technical Science in ETH-z Switzerland. She has been academic and practicing architect for 30 years and has received many designing and research awards of urban projects. She is a Council Member of the Chinese Architects Association and Urban Design Association, as well as the Council Member of ISUF. Her current research interests are focused on urban morphology and urban policies, and the correlation between Urban Form and Microclimate. Professor Ding is the corresponding author of â&#x20AC;&#x2DC;Urban plot characteristics study: Casing Centre District in Nanjing, Chinaâ&#x20AC;&#x2122;. Yijia Dong Tongji University, China Yijia Dong is a PhD candidate at the College of Architecture and Urban Planning, Tongji University, China. Quanle Huang South China University of Technology, China Quanle Huang is a Researcher in School of Architecture, and Architect + urban designer in Modern Architecture Design and Research Center, both in South China University of Technology, Guangzhou, China. After 8 years of research and work in Paris, Quanle Huang received her PhD of Architecture from Doctoral School of City and Environment in University of Paris-8, France. Her interest is based mostly on the metamorphosis of urban form during rapid urbanisation process in Chinese metropolis. Apart from practical projects in the institute, she teaches urban design theory in School of Architecture in SCUT, and was involved in numerous urban studies collaboration with various foreign organisations like ITDP from New York and ENSAPB from Paris, etc. Pierre Gauthier Concordia University, Canada Dr Pierre Gauthier is Associate Professor in the department of Geography, Planning and Environment at Concordia University in Montreal, Canada. His research interests are related to the investigation of the built environments in the making, looking in particular at the urban tissues as a cultural form and at the

Urban Form at the Edge: Proceedings of ISUF 2013, Volume II

200

urban morphogenesis. Specific topics include: the morphogenesis of Quebec City and Montreal, the impact of spontaneous and purposeful building practices on urban form, and the impact of transportation infrastructures on urban and regional forms. Mirko Guaralda Queensland University of Technology, Australia Dr Mirko Guaralda is Senior Lecturer in Architecture at the Queensland University of Technology; his background includes experience in architectural design, landscape architecture and urban design. Before joining academia full time, Mirko has been working in industry and local government; he has been involved in a wide range of projects at different scales, from small dwellings and gardens, to new estates and urban strategic planning. Since 2001 he has been involved in research projects in Australia as well as in Italy, his home country; he is currently research associate with the Centre for Subtropical Design, the Urban Informatics Research Lab and the Children and Youth Research Centre at QUT. Maider Uriarte Idiazabal University of Basque Country, Spain Maider Uriarte Idiazabal is a lecturer with Universidad del PaĂs Vasco/Euskal Herriko Unibertsitatea and PhD researcher at University of Basque Country, Spain. She is a Member of the editorial board of Aldiri. Arkitektura eta Abar, and is a practicing architect. Guoxin Ke Nanjing University, China Guoxin Ke graduated from Nanjing University in 2014, and works in East China Architecture Design and Research Institute as an original architect. Tao Li Urban Regeneration Research Center/Zhubo Design Group, Guangzhou, China Tao Li is Principal of Urban Regeneration Research Center, and chief urban planner in Zhubo Design Group, Shenzhen, China. Tao Li graduated from School of Urban Planning and Landscape in Tongji University in Shanghai, and has abroad research experiences both in Italy and in France. His interest focuses mainly on citiesâ&#x20AC;&#x2122; historical studies and real projects during the radical city developments in China. With this interest, he has worked for various sorts of institutes of urban projects, from university to municipal urban planning organisation, from international firm to private enterprise in China, which, as a special way of sensing the design and research variety, has given to him special angles to observe the urban development process. Ye Li University of Auckland, New Zealand Ye Li is a PhD Candidate in the School of Architecture and Planning at The University of Auckland. Her research interests involve urban form and architectural typological studies in Asian cities. Current research focuses on the structure and evolution of urban landscape in Nanjing and Guangzhou, China. Nicola Marzot TU Delft, The Netherlands Dr Nicola Marzot is currently an Assistant Professor at TU Delft, The Netherlands, Department of Architecture, Chair of Public Building and Researcher at Ferrara University, Italy, Department of Architecture. He defended his first PhD in Building and Territorial Engineering in 2000, Faculty of Engineering, Bologna and his second PhD in Architectural Composition in 2014, Faculty of Architecture and the Built Environment, Delft, The Netherlands. His research activity focuses on Urban Regeneration Processes in theory and practice, developing interdisciplinary methods. He has been a member of ISUF since its foundation, and was Secretary General (2006-2009). He has authored more that 270 publications on scientific journals and books. He is vice-director of the international journal Paesaggio Urbano. He is partner of the architectural firm PERFORMA A+U, based in Bologna, Italy.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume II

201

Walter Peters University of the Free State/University of KwaZulu-Natal, South Africa Professor Walter Peters is an architect, editor and architectural historian with penchant for urban history and morphology. He is Professor of Architecture at the University of the Free State in Bloemfontein, and an Emeritus Professor of the University of KwaZulu-Natal in Durban, South Africa. Paul Sanders Queensland University of Technology, Australia Professor Paul Sanders lectures and researches in the School of Design at QUT. He has been an academic since 1998 and has been published in international journals in the fields of architecture and urban design. His research is in the field of urban morphology, and he completed a PhD in 2015 titled ‘Consonance in Urban Form; The Architectural Dimension of Urban Morphology’. He has been a practicing architect for 15 years, having extensive design experience through built work, as well as international design competitions. His areas of expertise include Architectural design, Architectural technology, Urban morphology and Regional modernism. Javeria Shaikh Hanyang University, South Korea Javeria Shaikh is Professor of International Forum of Spanish Archaeology, M.Phil in Building Repair and Diagnoses. She completed undergraduate studies in architecture and then Masters from UET Lahore followed by M.Phil from Spain, Italy and Poland, which is a integrated program of European Commission. She is multi-specialised scholar (her current projects includes the up-gradation and design of Alkaram Mill complex Karachi and Sir Syed Tower Karachi. She is completing her PhD in History, Theory and Criticism from Seoul, South Korea under HEC scholarship. Her focus is on Conservation of Sindh Heritage. Her specialisation is Healthcare, where she is designing Mother and Child hospital in Sukhar and Cardio Department Shiffa Karachi. She published 20 books on her work of Bachelors Masters and PhD, and presented 15 conference papers in Australia, USA, Korea and France while doing her PhD. She is working on research for the Mohenjodaro. She earned 6th Position in Worldwide competition of Erasmus Mundus. Todor Stojanovski KTH Royal Institute of Technology, Sweden Todor Stojanovski is an architect and researcher in the division for Urban and Regional Studies at KTH Royal Institute of Technology. He works on the linkage between urban form and transportation, on Transit-Oriented Development (TOD) as urban design/morphology for/of multimodal transportation, and integration of public transportation in cities. Zhenyu Li Tongji University, China Professor Zhenyu Li is Dean of the College of Architecture and Urban Planning, Tongji University, China Lina Zhang Nanjing University, China Lina Zhang is a PhD Candidate in School of Architecture and urban Planning, Nanjing University. Zhang obtained the Master degrees at School of Architecture in Nanjing University and completed the Masters of Advanced Studies at Department of Architecture in ETH Zurich. She has participated in several international activities and conference including 11th International Architecture Exhibition of the Venice Biennale, 2011 Shenzhen and Hong Kong Bi-City Biennale of Urbanism/Architecture, 2012 ISUF Conference. Her main research field is the urban morphological study in China especially on the historical investigation, formation analysis and quantitative description. Wu Zhouyan University of Tokyo, Japan Wu Zhouyan received her bachelor’s degree in Architecture from Zhejiang University, China in 2011. She received a master’s degree in Architecture Theory and Design from the School of Architecture and Urban

Urban Form at the Edge: Proceedings of ISUF 2013, Volume II

202

Planning, Nanjing University, China in 2014. She is currently a doctoral student and a member of Chiba Lab, Department of Architecture, Graduate School of Engineering, The University of Tokyo, Japan. Her research interests are related to transformation of urban form and cinematic approach in urban form study.

Urban Form at the Edge: Proceedings of ISUF 2013, Volume II

203

Urban Form at the Edge: Proceedings of ISUF 2013, Volume II

204