Historic Wooden Architecture in Europe and Russia

Historic Wooden Architecture in Europe and Russia Evidence, study and

restoration

Edited by Evgeny Khodakovsky and Siri Skjold Lexau

6

I N T RO D U CT I O N

Wood in the Architecture of Europe and Russia: National Specifics and International Research / Evgeny Khodakovsky 18

1 . H I S TO R I C WO O D E N A RC H I T E CT U R E I N E U RO P E A N D R U S S I A : A RC H A E O LO G I CA L E V I D E N C E A N D S T U DY

20

Wooden Churches in Viking and Medieval Norway: Two Geometric and Static Strategies / Jørgen H. Jensenius

28

Timber Churches in Medieval England: A Preliminary Study /

42

The Archaeological Study of the Wooden Religious

Mark Gardiner Architecture of Medieval Novgorod / Marina A. Rodionova, Victor A. Popov, Nadezhda N. Tochilova 56

2 . W R I T T E N E V I D E N C E O N M E D I E VA L WO O D E N A RC H I T E CT U R E

58

Viking and Medieval Wooden Churches in Norway as

68

Written Sources for English Medieval Timber Architecture /

78

Archives and Historical Documents in Contemporary

Described in Contemporary Texts / Jørgen H. Jensenius Mark Gardiner Research of the Wooden Architecture of the Russian North / Evgeny Khodakovsky, Arina Noskova 92

3 . T H E A RC H I T E CT U R A L O B J E CT AS P R I M A RY H I S TO R I CA L E V I D E N C E

94

Stave Church Research and the Norwegian Stave Church

110

Traditional Structures in Russian Wooden Architecture:

Programme: New Findings - New Questions / Leif Anker Technical Aspects / Andrei Bode 122

Wooden Elements in the Stone Architecture of Medieval Novgorod / Ilya V. Antipov

134

Aspects of Bohemian and Swedish Wooden Bell Towers / Karel Kuča

144

4 . E M P I R I CA L S T U DY FO R P R E S E RVAT I O N A N D R E S TO R AT I O N : E X P E R I E N C E A N D P E RS P E CT I V E S

146

The Restoration of Wooden Architectural Monuments in Russia.

156

The Church of the Transfiguration at Kizhi Pogost –

Contemporary Methods and Approaches / Andrei Bode Some Reflections on the Building and its Restoration / Arnt M. Haugen 168

C O N C LU S I O N

Historic Wooden Architecture in Europe and Russia: Different Approaches to New Knowledge / Siri S. Lexau

178

AU T H O R I N FO R M AT I O N

182

APPENDIX / IMPRINT

8

I N T RO D U CT I O N WO O D I N T H E A RC H I T E CT U R E O F E U RO P E A N D R U S S I A : N AT I O N A L S P E C I F I C S A N D I N T E R N AT I O N A L R E S E A RC H

Evgeny Khodakovsky In the centuries-long history of art in Europe and Russia, wooden architecture occupies a special place. The extensive areas in southern Norway and the Russian North and those places in the Czech Republic, Poland and England where masterpieces of timber construction still survive are today regarded as unique architectural reserves. Some wooden churches, such as the Urnes Stave Church and the Church of the Transfiguration at Kizhi, are included in the UNESCO list of World Heritage Sites.

For a long time, however, against the background of a magnificent panorama of masonry architectural monuments, the role of wooden buildings was not rated very highly and their significance was not properly appreciated. To a large extent this bias was due to a deep-rooted perception going back to Antiquity of timber construction as an indicator of the low level of material and spiritual development of barbarian societies. Back in the first century AD, Tacitus, describing the world of the Germanic tribes that lay beyond the northern boundaries of Roman territory, observed that: ‘They are unacquainted with the use of mortar and tiles; and for every purpose employ rude unshapen timber, fashioned with no regard to pleasing the eye.’1: 16 (37) Nevertheless, the decline of masonry construction that followed the demise of the Western Roman Empire in the year 476 increased the significance of wood as a building material in the subsequent history of construction in Europe.

Against this background a certain symbolism would seem to attach to the mention of the earliest recorded wooden church in the Roman outpost of Quintanis (present-day Künzing) on the Danube. The early Christian Life of St. Severinus reports: ‘The inhabitants of this place had built outside the walls a wooden church which overhung the water, and was supported by posts driven into the riverbed and by forked props. In place of a flooring it had a slippery platform of boards, which were covered by the overflowing water whenever it rose above the banks.’2: XV (60) By its very location on the banks of the river forming the border, the church in Quintanis connected in a way the Roman and Germanic worlds, just as the period in which it is recorded – the 470s – connected departing Antiquity with the coming Middle Ages, when amid the marble ruins the role of wood in architecture grew considerably.

After the previous dominance of the ‘Mediterranean vector’ in the development of culture ceased to be unconditional, the forests of central Europe and the character of their usage became an important ‘factor in architecture’ because in the Middle Ages ‘relations between humans and the forest changed’. 3: 109 We would be to a large extent correct in picturing medieval construction as almost

9

Fig. 1: Urnes Stave Church in Sogn, Norway.

After 1130. Photo, 2012

Fig. 2: Kizhi Pogost (1694–1874), Republic of Karelia, Russia.

Photo, 2007

entirely wooden, although today that is hard to believe in view of the almost complete loss of structures made from that relatively short-lived material. A host of mentions of extensive construction in wood have come down to us from the Middle Ages. The collected data from various sources inform us of the existence across Europe and Russia of several hundred wooden churches created at various times. 4 And that is far from a complete picture.

10

Gradually, though, wooden structures were rebuilt in masonry, which – in an era of constant wars, raids, fires and destruction – was valued above all for its durability. Stone and brick edifices, orientated on some authoritative prototype, performed representational functions and then, as constructional methods and techniques advanced, they acquired ever greater aesthetic value. This process can already be observed in the Carolingian period. Adam of Bremen, for example, mentions that Bishop Willerich, who lived in the first third of the ninth century, ‘erected churches in appropriate places; three, indeed, in Bremen, the first of which, that is to say the Cathedral of Saint Peter, he made over from wood into stone’. 5: 24 The British Isles, which in the ninth and tenth centuries experienced incessant, devastating blows from the Scandinavians, also diplayed a strong upsurge in masonry construction in this period, which despite its very plain forms proved more in demand and more financially practical than inexpensive, but highly flammable wooden churches. Two Irish edifices can serve as examples – St. Columba’s House at Kells (circa 800) and ‘St. Kevin’s Kitchen’ at Glendalough (mid-ninth century).

Nevertheless, the seemingly irreversible process that saw the gradual replacement of dilapidated wooden churches by masonry buildings proceeded at slow pace in some parts of Europe and Russia: a number of regions in Scandinavia, the British Isles, the Novgorodian Republic, Bohemia and Silesia. The specific character of the development of these historical areas has determined the geographical boundaries of the questions addressed in this monograph, reflecting specialist research into the wooden architecture of Norway, England, Russia and the Czech Republic.

The great attention devoted in the book to Norwegian wooden architecture in particular is entirely explicable. Norwegian stave churches of the twelfth to fourteenth centuries are a group of very early wooden buildings that have survived down to the present. The exceptional nature of this phenomenon can be attributed to a wide range of factors. While the rest of the barbarian world (Gaul, Germany, Britain) had long ago been drawn into the sphere of influence of the Roman Empire and then the Church of Rome, Scandinavia – due to its geographical remoteness and peculiarities of social development – right up to the late eighth century took no active part in the formation of the cultural and political landscape of the early Middle Ages. It is only natural that in Norway – situated on the periphery of Europe and separated from the main territories of the continent by the North Sea and Skagerrak – art from the outset assumed a special local ‘endemic’ character. Isolation from the Roman legacy and the developed infrastructures of towns and roads associated with it, the complete absence of masonry construction with its elaborate techniques – the use of concrete, arches and vaults – also determined the distinctive character of the development of Norwegian architecture, which, drawing mainly on its own internal creative resources, preserved its uniqueness for centuries. This was also favoured by the natural conditions of northern Europe, where the abundance of forests preordained the use of timber as the main (and at times only) building material.

11

Fig. 3: Hopperstad Stave Church in Sogn, Norway.

Mid-12th century.

Photo, 2013

The preservation of wooden church architecture of the twelfth to fifteenth centuries in southern and central Norway, where the largest number of examples are located, is also due to the fact that until very recently Norway remained a country with a patriarchal agricultural way of life practically untouched by urbanisation and industrialisation. Norwegian geography – huge areas intersected by mountains, numerous fjords and passes that are closed in winter – favoured the conservation of many aspects of cultural life, the formation of local cultural traditions in different regions and the development of dialects of the language.

Another important circumstance is that as early as the late fourteenth century Fig. 4: Hopperstad Stave Church in Sogn, Norway.

Mid-12th century. Nave.

Photo, 2008

Norway lost its sovereignty and became dependent politically on Denmark and economically on the Hanseatic League. In the 1500s and 1600s, neighbouring Sweden became a wealthy state with imperial ambitions and financial capabilities that allowed it to embark on large-scale masonry construction, not only in the capital and the larger towns, but even in the countryside. As a rule, the construction of a new masonry church in place of an old wooden predecessor was the consequence of the parish acquiring sufficient funds, which testifies to the financial prosperity of the society. Ageing tumble-down wooden churches in Sweden and Denmark were gradually replaced with masonry ones, which were more reliable and functional, but with every passing generation possessed fewer and fewer of the features that characterised earlier stages in the evolution of medieval architecture. In

12

Norway, for centuries the situation was completely the opposite. The land remained as before on the periphery of the political and economic development of the realms in the Kalmar Union and had no such capabilities. But it was precisely this fact that facilitated the long retention of old churches, the majority of which remained wooden. As a result even today the overwhelming majority of timber dwellings, service and church buildings are in the provinces, while the appearance of the artistically most interesting cities in Norway – Bergen and Trondheim – is primarily formed by large-scale wooden construction.

Fig. 5: Eidsborg Stave Church in Telemark, Norway. Circa 1250– 1300. Photo, 2014

Works of wooden architecture, which easily catch fire and are more susceptible to atmospheric influences, are in a more vulnerable position compared with durable masonry structures. Despite active measures to preserve wooden churches in Norway, they now number some two and a half dozen objects, a mere three per cent of the total quantity of Norwegian churches in the Middle Ages.

The task of present-day researchers lies in devoting maximum attention to the surviving buildings as hands-on empirical material. It is essential to continue the research of them to expand the knowledge of this persisting building technology of the past, during which the current predominance of extant masonry edifices was not at all evident. The wooden architecture of Norway to some extent compensates for the lacunae created by the loss of so many timber buildings and substantially expands the history of construction in Europe in the twelfth to fourteenth centuries, its technologies and typological variety.

Fig. 6: Lom Stave Church, Oppland, Norway.

After 1158; extended in 1634 and 1663.

Photo, 2013

13

Fig. 7: Kvernes Stave Church, Møre og Romsdal, Norway. 14th– 17th century. Photo, 2014 Fig. 8: St. Andrew’s Church, Greensted, Essex, England. 11th century; extended 16th–19th centuries.

Photo, 2011

The flourishing of wooden architecture in Norway would be the logical result not only of the long and varied use of wood in structures of material and spiritual culture in pagan Scandinavia. It is inseparable from the history of the development of timber construction in the British Isles – a region that played an extremely significant role in the Christianisation of Norway. For that reason buildings in Britain are allotted a special place in the monograph. They are examined both in the chapter devoted to methods of archaeologically reading wooden constructions, alongside an interpretation of Norwegian and Novgorodian material, as well as in the section on written sources as one of the most important research tools. Within the extensive archaeological array revealed in the course of excavations in the British Isles, particular value is attached to objects that lend themselves to at least minimal reconstruction. Of especially great worth, though, against that background is St. Andrew’s Church in Greensted, Essex (not far from London).

St. Andrew's was constructed in the eleventh century and is today the oldest surviving wooden church in the world. Thus, through a multifaceted approach to the study of archaeological evidence, written sources and a surviving object, albeit the only one, that is in itself a precious source and evidence, the monograph reveals and examines the strong and long-lived tradition of wooden church construction in Britain.

The section on the traditional wooden architecture of the Czech lands, examined in the context of a possible link with the bell towers of Sweden, is not only innovative, but also deeply symbolic as the Czech Republic, situated in the heart of Europe, provides a more than merely visual geographical link between the main regions featured in the book – Britain in the west, Norway in the north and Russia in the east. A detailed study of the processes that took place in the architectural history of central Europe after the Reformation points to a possible interaction, albeit limited, between the Scandinavian and Slavic traditions. While in the architecture of church buildings such a consonance would seem impossible on the grounds of confessional differences and the need to express ethnic and cultural identity, in a more practical functional sphere, where the ideological aspect is reduced to a minimum, it is possible to find points of

16

Fig. 11: Church of the Presentation in the Temple, Zaostrov’e, Arkhangelsk region, Russia. 1688. Photo, 2014

it possible to draw conclusions about general patterns in the developmental history of wooden architecture in a particular period. Additionally a broad, comprehensive approach of this sort makes it possible to touch on adjacent topics, such as the use of wood in masonry architecture, or the specific character of the construction of bell-towers, which occupy an intermediate place between ecclesiastical and civil architecture.

The final chapter examines and compares various methods of restoration employed previously and currently in Norway and Russia. Emblematic in this respect is the active involvement of Norwegian specialists in the restoration of the Transfiguration Church in Kizhi. Despite the long-standing fundamental differences in the building traditions of the two countries, in this important

17

Fig. 12: The wooden Church of St. Demetrius of Salonica (1646; 1731) and the masonry Cathedral of St. George (1165). Staraia Ladoga (Old Ladoga), Leningrad region. Photo, 2005

project centuries of Russian and Norwegian experience is being brought together. At the same time a powerful motive force here is also the shared perception of wood as a living, organic material that not only possesses engineering and technical properties, but also reveals through its use in architecture a distinctive national perception of the world.

References 1 Tacitus, The Germany and the Agricola of Tacitus, Oxford translation revised with notes, with an introduction by Edward Brooks, Chicago: C. M. Barnes, 1897.

2 Eugippius, The Life of Saint Severinus, translated into English for the first time, with notes by George W. Robinson, Cambridge, MA: Harvard University Press, 1914.

3 Küster, H., Geschichte des Waldes: von der Urzeit bis zur Gegenwart, Munich: C. H. Beck, 2003.

4 Ahrens, C., Die frühen Holzkirchen Europas, Vol. I–II, Stuttgart: Konrad Theiss Verlag, 2001.

5 Adam of Bremen, History of the Archbishops of Hamburg-Bremen, translated with an introduction and notes by Francis J. Tschan; with a new introduction and bibliography by Timothy Reuter, New York: Columbia University Press, 2002.

6 Suslov, V., Putevye zametki o severe Rossii i Norvegii [Travel Notes on the Russian North and Norway], St. Petersburg: Tipografiia A. F. Marksa, 1888.

All illustrations are by the author.

1. Historic Wooden Architecture in Europe and Russia: Archaelogical Evidence and Study 1 . 1 WO O D E N C H U RC H E S I N V I K I N G A N D M E D I E VA L N O RWAY: T WO G E O M E T R I C A N D S TAT I C S T R AT E G I E S / Jørgen H. Jensenius 1 . 2 T I M B E R C H U RC H E S I N M E D I E VA L E N G L A N D : A P R E L I M I N A RY S T U DY / Mark Gardiner 1 . 3 T H E A RC H A E O LO G I CA L S T U DY O F T H E WO O D E N R E L I G I O U S A RC H I T E CT U R E O F M E D I E VA L N OVG O RO D / Marina A. Rodionova,

Victor A. Popov, Nadezhda N. Tochilova

20

WO O D E N C H U RC H E S I N V I K I N G A N D M E D I E VA L N O RWAY: T WO G E O M E T R I C A N D S TAT I C S T R AT E G I E S

Jørgen H. Jensenius The question of the ‘origin’ of medieval wooden churches has not yet been solved. It may be because the question is vague, because answers are thoughts which do not carry the weight of arguments, or that facts are seen isolated from contexts. In the available literature it has been claimed there was a sudden leap from posts set in the ground to staves placed on the ground, to prevent moisture damage to the posts/staves. To question the claimed relation of cause and effect, this paper describes two foundation strategies. Geometric and static consequences for the superstructures are discussed, and I ask how roof spans and loads affect the choice of footings.

Background

It has been assumed that Norwegian craftsmen – accompanying Viking wintering expeditions in central Europe – designed transportable and easily assembled buildings. 1 With the evangelisation of Norway, foreign customs and attitudes came to influence local ways of thinking. Travelling craftsmen brought with them knowledge of how to plan, design and prepare buildings; the clergy may have proposed what they were accustomed to. 2 The Church was protected by the king, who governed by force and fiat. One could become a church owner by inheritance, by purchasing property with a church standing on it, or by building a church oneself. Landholders ordered churches to be built on their demesnes and paid for the work in order to legitimise their rule and prove their loyalty and faith to the bishops and kings. 3 Gradually the administrative project of establishing bishoprics produced results. Estimates give a figure of more than 2,000 wooden churches in Viking and medieval times in Norway. The vernacular vocabulary of suitable wooden designs had to be adapted to ecclesias-

Fig. 1: Oseberg chamber, dated to 834 (University of Oslo: Museum of Cultural History)

tical specifications. The nineteenth-century art historian Lorentz Dietrichson named 322 wooden churches; for him all seemed to be ground-set buildings. 4

Church planning and design There is no specification in the New Testament for a sacred space. Descriptions of a sacred space in the Old Testament are reinterpreted by Paul as the assembly of true believers in Christ. 5 A thousand years later, there was no universal blueprint for building wooden churches – planning seems to have been a mixture of conformity to tradition and local diversity; design followed need, funding and the properties of the materials. The building process was not secret, difficult or mystical: If there was a ‘trade-secret’, it was the years of meticulous practice.

21

Guiding principles were pragmatism and simplicity, designs were shared by rules of thumb, and orally and visually through the apprentice system. General practice was known and the details may have been easy to learn and remember. 6 The building and its parts were shaped by measures and ratios, without resort to Euclidean theory. 7 Some ratios can be reconstructed, but theoretical assumptions and how the design was understood is probably lost in history and irretrievable.

The function of a church is liturgical practice, and a variety of local buildings may have been accepted as potentially suitable. Buildings made of wood or stone, large or small, earth-fast or ground-set, wattleand-daub or palisade construction were all consecrated in northern Europe. 8 Abundant and readily available in Norway, pine (ĂželliviĂ°i; Pinus sylvestris) could be locally sourced and delivered quickly. An embedded footing is termed earth-fast, an independent one is ground-set. In terms of durability, performance and cost, the earth-fast construction may have been seen as a viable concept of church building for hundreds of years.9 Many problems are associated with wooden building designs, relating to thermal breaks and leaks, Fig. 2: Map with earthfast churches

water penetration, moisture and freeze/thaw cycles. Direct or indirect traces of some twenty-five presumably earth-fast churches have been documented in Norway over the last 60 years; roughly dated as from the year 1060 and later.

All that may be left are pits with a bottom stone, infill and the lower stump of a post. In addition, negative forms of building remnants like post-holes, impressions of door posts, wall plates and flooring can be traced. Yet, even from meticulously documented foundation remains one cannot deduce the original superstructure of the buildings. Valid insights and formulated hypotheses by a building archaeologist notwithstanding, any description of missing parts of a church building will be nothing but speculation without tangible evidence.

Fig. 3: Drawing, pit with post

Earth-fast churches On the building site the oriented ordinates stipulating walls were marked on the ground. A typical building would have a series of roof-bearing posts along its perimeter, set into pits. If the ground was soft, pits could easily be excavated, but it could not support heavy earth-fast posts and so settling could occur. On the other hand, stony ground would prevent transmission of the load through the posts to a deeper soil stratum to avoid uneven frost heave or thaw weakening in spring. The inevitable spoils from the pits could easily damage markers, making it difficult to verify the accurate placing of posts, as may be deduced from excavation plans. 10

22

To prevent settling and avoid moisture and fungi from attacking the wood, it was necessary to ensure drainage for water run-off by putting sand and a stone slab at the bottom of the pits which would act as a moisture-resistant barrier and by charring the end of the post. Later, roof overhangs may also have been created for the same reason.

No precise levelling may have been done in the pits, so the top of the posts would be at slightly different elevations. To help keep the post plumb after erection, its lower end was fixed by bracing with heavier stones, before being packed tight with backfill of smaller stones and earth.

Fig. 4: Post-hole Høre

The use of 89 bracing stones has been recorded in 160 excavated post pits in Norway. A post with a diameter of 0.3 m and a length of 5.0 m weighs in excess of 280 kg. Even with considerable bracing a depth of 0.3 to 0.9 m will not give the posts lateral stability. When a church was disassembled, the posts might be lifted from the pit, leaving behind post-holes, negative impressions in the rammed infill. As I have shown elsewhere, preserved fragments of 66 posts from 17 excavated churches have diameters or width of 0.2 to 0.4 m. 11 The cross-section can be squared, rounded or rectangular, although the shape above ground may have differed. Empty post-holes in Kaupanger II, Eidskog, Høre I, Ringebu I and Bø I indicate that posts were lifted out when the building was disassembled.

Builders may have preferred to work with right angles, but, despite adjustments to make a better fit, the posts would often have been placed with undesirable deviations from the planned design. To nullify differential settlement and variation in centre distances and alignment, short, adjustable wall plates may have linked the posts. Therefore to make the wall a rugged construction, lateral stability in all directions was provided by inserting tie beams and diagonal braces

Fig. 5: Arson in Fantoft Stave Church 1992

at the upper ends of the posts. Even if excavations in Norway have not revealed wooden floors in earth-fast buildings, we cannot rule out suspended floor solutions. A tamped earthen floor is cheap, would require little transport and produced no waste which could rot. Substrates of pebbles and sand would act as a capillary break from moisture in the ground.

A floor may be flat even if tilted. A slightly sloping earthen floor was perhaps regarded as merely a minor inconvenience. An earthen floor is easy to fill in and smooth over after an indoor burial. Accumulation in floor level may gradually have covered the lower part of the sills. As time went by, thresholds had to be raised and doors trimmed. Local conditions of climate, maintenance, demography, economy, use and abuse and changing requirements made the durability of the embedded posts difficult to anticipate. A church had to be replaced when it was burned or swept away by inundation, landslide, avalanche or heavy wind.

23

Fig. 6: Post-hole and raft beams in Lom

Fig. 7: Uvdal Stave Church

Another reason to replace a church was when maintenance costs equalled the expense of rebuilding. This could explain why the earth-fast buildings at Urnes, Mære and Kaupanger were replaced by others with similar foundations and area.

The earth-fast design may still have been appreciated as a time-honoured viable church concept. Liturgical practices could change with little impact on the fabric of a building, except where nave and altar room had become too narrow for practical use. Earth-fast churches in Lom, Høre, Ringebu and Kaupanger were replaced by ground-set buildings in the later 1100s, at the same time as earthfast structures in Mære, Bø and Kinsarvik were superseded by stone buildings.

When the Archbishopric of Nidaros was established in the years 1152–53, the options available were small earth-fast, ground-set and stone churches, medium ground-set and stone churches, and large stone churches. None of these solutions is objectively better, though they may all have been appreciated as adequate constructions in their own way. In spite of foundation variants of old and new, buildings had a common base in design and craft, in social and economic practice. Earth-fast construction may have continued alongside ground-set for a long time, and both are still in use for certain purposes today.

Planning the ground-set church A tall roof was not prescribed liturgically nor was it a requirement for the practiceof faith. Grand loftiness in a church may have been seen as a cultural asset, a social want, a powerful political sign or have been inspired by the architecture in vogue.

34

couple. Close-set rafter couples or common rafters are typical of twelfth-century French buildings and there is some evidence of similar roof types in England at the same period and earlier. Table 1 provides a summary of the evidence for the spacing of rafter trusses from masonry buildings where this survives from an early date. These trusses are, if anything, even closer than the typical spacing of wall posts. How much earlier this type of roof construction was used remains unclear.

The earliest of the timber churches at Foxley shows many of the features of the English early medieval tradition of construction, while having a number of unusual aspects. It was built with vertical planks set in a wall trench, though they were exceptionally thin, measuring only 0.1 m wide. The planks on the south side were set close to the inner vertical edge of the wall trench; those on the north were set closer to the centre. The particularly notable aspect of the building was that the planks of the walls were contiguous set alternately into slots in the underlying limestone and resting on the surface of the limestone.

Post-holes set outside the wall-line may suggest that the walls were also supported by additional timbers. 31 The church at Thetford is, however, more typical of early medieval buildings.

Traces of individual posts were identified along parts of the wall trench, and we Table 1: Mean Spacing (Centre to Centre) of Rafter Trusses from Selected Early Masonry Buildings in England and France

English buildings

French buildings

must assume that there were similar close-set posts along the remainder. The slight evidence for the east wall is also reflected in the remains of the church at Norwich Castle, where both the east and west walls were marked only by a few post-holes. The weight of the roof was carried on the side walls and the end walls were formed using wattle and daub, of which little trace remains. The

SITE

DATE

SPACING OF RAFTER TRUSSES

Odda’s Chapel, Deerhurst, Glos

1056

0.67 m

Kempley Church, Glos

c. 1120

0.70 m

Fyfield Hall, Essex

1167–85d

0.57 m

Adel Church, West Yorks

c. 1160

0.54 m

Chapel, Harlowbury, Essex

c. 1180

0.60 m

Chapel of manor house, Harlowbury, Essex

1175–1200

0.56 m

Wistanstow Church, Shropshire

1200–1221d

0.86 m

Mesnil-Mauger, church of Notre-Dame de Sainte-Marie aux Anglais

1144d

0.73 m

18 Rue Saint-Romain, Rouen

1201–1216d

0.67 m or 0.82 m

Val-de-la-Haye, barn of the commandery, Sainte-Vaubourg

1216–1230d

0.62 m

Gisors, chapel of Saint-Laurent of the farm, Vaux

1224d

0.56–0.58 m

Note: dates with the suffix ‘d’ have been determined by dendrochronology

35

north wall of the Norwich church also clearly shows that the wall trench was divided into separate lengths, with the one-third to the west distinctly narrower than the remainder.

We should not imagine that these timber churches were insignificant buildings or were undecorated. Many Anglo-Saxon churches were richly painted and this decoration was not confined to masonry buildings alone. 32 Excavation of an early eleventh-century timber chapel at Colchester (Essex) has shown that the gaps between the posts were infilled with wattlework to which plaster had been applied and then painted. 33 Similar plaster panels set between vertical posts have been recovered at Eynsham (Oxfordshire) and dated to the tenth century.

The nature of the building to which they belonged is uncertain, although the site was a monastery in this period. The panels were about 200 mm wide and were separated by voids for the posts of similar width, suggesting that the timbers were set at centres 400 mm apart. 34 A third example of such panel infills has been found in the limited excavations of a timber church at Woodeaton (also Oxfordshire) dating to the period 1000–1080, but in this more modest building the material applied to the wattlework was daub. The panel widths were narrower in this building, measuring only 80 mm. The width of the timbers between which these daub panels were set is unknown. 35 A fairly clear picture emerges of the timber-built tenth- and eleventh-century churches, if we set the early building at Foxley aside. They were characterised by the use of close-set vertical timbers generally set in a building trench, but sometimes in individual post-holes. The timbers were carefully aligned and the evidence from the plaster and the daub panels suggests that the sides of the posts were grooved to take horizontal spars which held wattlework. Greensted church was different from this general pattern in that the posts were set contiguously and were rounded on the outside face rather than squared. The Greensted timbers were also grooved on their sides, but instead of holding spars, they carried fillets which sealed the gaps between the adjacent posts. The church buildings differ from most lay buildings of the period in the particularly close spacing of the posts, a feature which reaches an extreme at Greensted.

The persistence of timber construction in later medieval churches Twenty-nine later medieval churches still survive in England which were timberframed or had a large component of timber in their original construction. These were built from the early fourteenth to the first half of the sixteenth centuries and occur in four main areas – Cheshire, Essex, Herefordshire and Worcestershire, and Hampshire. 36 A number of the churches lie outside these historic counties in adjoining regions, but are considered under these headings for convenience. The fact that the churches are concentrated in discrete areas suggests that there were specific factors operating in those places which led to their construction, the most obvious is that there was a strong local tradition of timber building. A small number of other examples are known to have survived up to 1800, but have been demolished since. These simply fill out or extend the distribution of the groups of surviving churches. For example, the timber church

42

T H E A RC H A E O LO G I CA L S T U DY O F T H E WO O D E N R E L I G I O U S A RC H I T E CT U R E O F M E D I E VA L N OVG O RO D

Marina A. Rodionova, Victor A. Popov, Nadezhda N. Tochilova Velikii Novgorod – Novgorod the Great – is a unique archaeological and architectural phenomenon with a thousand-year history. Wood was the main building material in medieval Novgorod and the exceptionally favourable conditions for the preservation of wood in the city’s damp cultural layer makes it possible to study the actual remains of various structures. These include dwellings, service and public buildings, as well as defensive features such as the wooden elements of the rampart around the city’s Detinets or kremlin and the Okol’nyi Gorod outer line of fortifications, and engineering structures such as the bridge over the River Volkhov. Civic amenities, such as pavements and drainage installations, were also an inseparable part of the architectural look of early Novgorod.

The extent to which these different components in the wooden architecture of medieval Novgorod have been studied varies. The scale of the explorations of the city’s territory carried out systematically over the course of more than eighty years has produced a situation where the main archaeological source on Novgorodian wooden architecture is material on the building of houses first put into scholarly circulation by Piotr Zasurtsev. 1 As archaeological material accumulated, separate publications appeared containing an analysis of dwellings in medieval Novgorod. 2, 3, 4 In recent years Natalia Faradzheva has come to occupy a special place in this literature. The result of her researches in 2010 was a dissertation on the subject of ‘The Structures of the Liudin Konets District of Medieval Novgorod’ (on material from the Troitskii excavations I–XI). 5 The study of housing construction has a very important place in answering not only questions about early Russian architecture, but also more general questions on the history of early Russian cities, since the appearance of a dwelling is shaped by and depends on many factors. A characteristic feature of archaeological sources of this kind is that only the lower parts of the ancient dwellings have been preserved. In the absence of written or pictorial sources, researchers have to reconstruct the exterior of early Russian dwellings through extrapolation.

The extensive archaeological material has made it possible to identify the main types of homestead construction that existed in medieval Novgorod, the methods of creating buildings, the composition and layout of homestead complexes, and also to trace the formation and development of the urban environment.

Further archaeological studies will help to make these conclusions more specific.

The dominant architectural features of early Novgorod were the Detinets – the city’s topographical, administrative, religious, cultural and military-defensive centre – and the numerous churches located in various parts of the city.

Archaeological exploration of the Novgorod Detinets has been going on for more than a hundred years now, yet in that time only around three per cent of its total territory has been examined. The bulk of the digs were confined to

43

small areas and they were mainly connected to the restoration of architectural monuments or the installation of utilities. So, in comparison with the degree to which the territory of the wider city has been studied, the Detinets has not been subjected to systematic investigation. From the available architectural data it is possible to draw only two verifiable conclusions: First, the territory of the present-day kremlin was settled even before the construction of the earliest fortifications. Cultural deposits from the tenth to eleventh centuries have been discovered there, with surviving remnants of wooden dwellings dating from the initial phase in the development of this territory. Second, wood-and-earth elements of the rampart (oak cribs filled and covered with soil and clay) have survived from the earliest Detinets. All other questions concerning the structure of the Detinets fortifications and dating of them, as well as many questions about the historical topography and development of the layout of the city’s ancient core remain for the moment unresolved and require further study and evidence.

Columns of the oaken St. Sophia Cathedral (1040–80) When it comes to the wooden religious architecture of Novgorod, to which this article is devoted, there are not many prospects for study. Almost all Novgorodian masonry churches had a wooden predecessor, even the chief place of worship in the city – the St. Sophia Cathedral – was, according to the chronicles, originally built of wood. 6 Frequent fires obliged the Novgorodians to build these Fig. 1: The ‘columns of the oaken St. Sophia’ (1040–80) Novgorod State Museum-Preserve 7: 32

sacred edifices in stone and brick. Due to the standard practice in ecclesiastical construction of retaining the same location for the altar during reconstruction, masonry churches were erected on the site of the previous wooden ones. As the building of a masonry church required foundations, this completely excludes the possibility of finding any remains of their timber predecessors. While Novgorod’s masonry religious architecture has long been the object of study and its old churches are considered gems of early Russian architecture, practical studies of the medieval city’s wooden church architecture do not exist. It should be stated, though, that researchers have attempted to identify some archaeological finds as elements from wooden churches and even to relate them to the Scandinavian building tradition. The reference is to the so-called ‘columns of the oaken St. Sophia’ (1040–80) – two carved wooden architectural fragments that are kept in the Novgorod State Museum-Preserve. We should immediately state that the name employed here for this artefact, ‘the columns of the oaken St. Sophia’, is conventional and a tribute to tradition. There is no proof of a link between these fragments and the St. Sophia Cathedral that burnt down in 1049. The character of the carving at fi rst sight makes it possible

56

57

2. Written Evidence on Medieval Wooden Architecture 2 . 1 V I K I N G A N D M E D I E VA L WO O D E N C H U RC H E S I N N O RWAY AS D E S C R I B E D I N C O N T E M P O R A RY T E X TS /

Jørgen H. Jensenius 2 . 2 W R I T T E N S O U RC E S FO R E N G L I S H M E D I E VA L T I M B E R A RC H I T E CT U R E / Mark Gardiner 2 . 3 A RC H I V E S A N D H I S TO R I CA L D O C U M E N TS I N C O N T E M P O R A RY R E S E A RC H O F T H E WO O D E N A RC H I T E CT U R E O F T H E R U S S I A N N O RT H /

Evgeny Khodakovsky, Arina Noskova

58

V I K I N G A N D M E D I E VA L WO O D E N C H U RC H E S I N N O RWAY AS D E S C R I B E D I N C O N T E M P O R A RY T E X TS

Jørgen H. Jensenius In this text I will examine how written sources may throw light on wooden churches. Three groups of texts giving intentions and interpretations can be identified. Normative texts stipulate how a church ought to be or how a community should act. Secondly, attributive texts deal with symbolic meaning imputed to the building; a connection with a real church may be coincidental. Thirdly, descriptive texts relate to actual works, such as unique and memorable buildings, as viewed by the author. For translations there may have been a spectrum of practice, from strict verbatim to editing-cum-translation – a sort of rendition.

A combination of sources can provide a picture of the planning, design and assembly of wooden churches.

Evangelisation, church building There is no provision in the New Testament for a physical sacred space. Paul describes an assembly of believers in Christ meeting in private homes: ‘Aquila and Prisca with their house church send abundant greetings in the Lord.’1 By the second century, worship had become a stylised ritual in permanent assembly rooms. 2 Changing designs, experimental constructions and different materials accompanied the development of Christianity through the centuries, becoming consolidated as tradition. There was no such thing as ‘the definitive church’.

Some designs became so popular that they overwhelmed other possible approaches. No norms were formulated, however, regarding building material, the size or shape required for a building to qualify as a church: A small dedicated wooden structure in the countryside was just as much a church as a stone cathedral. Instead, Norwegian laws distinguished between churches of different status. In the eleventh century most of the evangelising bishops/priests in Norway were foreigners. Of the clergy, twenty-eight are known by name, three or four were Norwegians, one was an Icelander, one a Dane, another an Irishman, seven or eight Germans and eleven or twelve Englishmen. 3 They brought with them a mix of Mediterranean and central European customs, values, aesthetics, habits and language. Motives for translations may have been to teach basic skills in Latin in order to instruct the clergy, to defend property rights and to edify the laity. 4 Even placed on a ‘holy site’ or a so-called ‘sacred landscape’, the assembly building never became holy per se. To make it a functional church, relics of a saint had to be placed in the altar by the bishop in a consecration ritual. A relic is a mediator between the sacred and the mundane, a materialised myth.

Because the Divine Service did not centre on a cult object, it could be celebrated either on an immobile altar in a building or on a portable altar placed temporarily somewhere. 5 Thus king Olav Tryggvason had a tent (landtjald) put up, and

59

Fig. 1: Borgund Stave Church, nave, west elevation

let mass be sung on the beach at Moster in the year 995. 6 Bishops were responsible for increasing the number of assembly buildings by encouraging landholders to have them put up on their demesnes. 7 Gratian (about 1150) cites an early canon law stating that before any building starts, bishops are to determine if there are adequate funds for the future upkeep (custodia). 8 According to the Gulathing Law, the man or men paying for a church also had to keep it in repair and not ruin the site (alldrigin tuft eyða).9 Snorri Sturlason (c. 1178–1241) mentions wooden churches initiated by the monarchs, such as king Hacon the Good, 10 Olav Tryggvason 11 and Olav Haraldsson, 12 all of undisclosed construction.

The designation post church is quite modern, while stave church is post-medieval. The medieval texts refer only to churches of wood (viĂ°r) or stone (steinn).

Therefore, when Lorentz Dietrichson gave a list of 322 named wooden churches, he was unable to tell how their foundations were designed and his conclusions are therefore incomplete. 13

Site and orientation The choice of site would depend on ownership of property, quality and depth of soil, and access from a beach, river or road. It could also be connected with a place of remembrance. The building site may have been set out with sides long enough

92

93

3. The Architectural Object as Primary Historical Evidence 3 . 1 S TAV E C H U RC H R E S E A RC H A N D T H E N O RW E G I A N S TAV E C H U RC H P RO G R A M M E : NEW FINDINGS - NEW QUESTIONS / LEIF ANKER 3 . 2 T R A D I T I O N A L S T R U CT U R E S I N R U S S I A N WO O D E N A RC H I T E CT U R E : T E C H N I CA L AS P E CTS / A N D R E I B O D E 3 . 3 WO O D E N E L E M E N TS I N T H E S TO N E A RC H I T E CT U R E O F M E D I E VA L N OVG O RO D / I LYA V. A N T I P OV ) 3 . 4 AS P E CTS O F B O H E M I A N A N D SW E D I S H WO O D E N B E L L TOW E RS / K A R E L K U ČA

122

WO O D E N E L E M E N TS I N T H E S TO N E A RC H I T E CT U R E O F M E D I E VA L N OVG O RO D

Ilya V. Antipov Wooden elements were often used in the stone architecture of medieval Russia.

Unfortunately, the majority of them have been lost, while some structures have not yet been studied. As a rule, we can find information about the different wooden elements only in drawings; there are no descriptions or written analyses produced by restorers.

This chapter is devoted to the wooden elements of pre-Mongol Novgorodian architectural monuments (mid-eleventh to mid-thirteenth centuries). Wooden parts or traces of them have been found in sixteen buildings of that time (Table 1).

Wood was used for ties inside the walls and between the walls and pillars, for the construction of altar screens and ciboria, for the lintels of openings and the frames of doors and windows, for beams that supported galleries and the railings of those galleries, as well as for temporary builders’ structures (scaffolding, centerings and supports for centerings). The craftsmen also used wood in the lower parts of the foundations. It is possible to state that Novgorodian builders used wooden elements that were usual for pre-Mongol Russian architecture as a whole. 1: 133–144

Ties inside the walls and between the walls and the pillars We find sets of wooden ties inside the walls and between walls and pillars in all works of early Novgorodian architecture (Fig. 2). Usually an interlocking band of single round or square beams was inserted in the middle of the masonry of the walls. The beams were connected together with joints. As Sergei Lalazarov stated, these bands of wooden ties had two functions: Initially the rigid wooden frame would take the load of the wall’s mass until the lime mortar set; then, after the mortar had set, the wooden beams functioned as ties counteracting diagonal thrust. 2: 30 Also, if the pillars of the church were tall and thin, the ties could help to reduce movement in those elements. It is obvious that the levels of windows and, sometimes, door openings were deliberately made to correspond to the tiers of ties so that the builders could use the parts of the beams in the openings to attach window or door frames.

In works of early Russian architecture a lower tier of ties was often placed at the base of the walls, above the upper levels of the foundation. Among Novgorodian buildings of the pre-Mongol period we can find this feature only in St. Clement’s Church in Staraia Ladoga (although other Ladoga churches did not have it). The lowest ties of St. Clement’s are also unusual in that the builders laid two beams together that were connected by transverse beams. The outside ends of the beams lie directly beneath the exterior surface of the walls, so it is conceivable that they were used by the masons as a guide for laying the masonry above. 3: 115 The next tiers of ties were usually made in the areas where the arches or vaults were situated – in the abutments of the vaults which supported the upper galleries and in the abutments of the reinforcing arches. These two tiers of

123

ties were not only placed inside the walls but also connected the walls with the pillars or else connected the pillars alone. Also, we can often see ties in the base of the drum and in the abutment of the dome (Fig. 3). We do not have enough information to trace changes in the number of tiers of tie bands and alterations in their location during the eleventh to the first half of the thirteenth century.

It seems that the presence of five or more tiers of ties was usual for buildings in the early twelfth century, but in edifices built later, their number fell to three or four. Nevertheless, before a thorough investigation of the surviving examples is undertaken, this is only a hypothesis. The number of tiers in St. Nicholas’s Cathedral in Novgorod (1113) seems unusual: The restoration architects found seven tiers and postulated that an eighth also exists in the base of the walls. However, some of the beams visible in the window openings may be no more than holders for window frames.

The cross-section of the oak beams used for the Fig. 1: Works of medieval Novgorodian architecture: a) St. Sophia Cathedral. 1045–50; b) The Church of the Nativity of the Virgin in the St. Anthony Monastery. 1117–19; c) The Transfiguration Church on Nereditsa Hill. 1198.

ties was usually square, rectangular or trapezoidal, but round logs were sometimes used. It is not easy to determine the shape of a tie because we usually find only the imprints of beams in the mortar or empty channels. In the Transfiguration Church on Nereditsa Hill (1198), one particular tie beam was round at one end and square at the other. 4: 193 It seems that the main idea was to make the beams of the same thickness; the shape of the tie was not so important to the builders.

The ties were fastened together using halved joints with the ends of the beams projecting. Any extension to the length of ties was mostly made using various types of scarf joint (Fig. 4). Restorers have found iron spikes or nails hammered into the joints of the beams. When the builders had to make ties in an architec-

124

Fig. 2: The interior of the Church of the Nativity of the Virgin in the St. Anthony Monastery in Novgorod. 1117–19.

tural element of circular or semi-circular shape, they created a polygonal band of ties (normally they used three beams in the apses, eight beams in the drum) (Fig. 5). These beams were also connected with halved joints.

It seems very important to understand whether the joints between the beams were made in situ, actually on the walls, or the construction was first prepared on the ground and its parts were later reassembled up on the walls. Sergei Lalazarov believes that it would have been very hard to make joints on the walls because the mortar on which the beams were laid had not yet set; furthermore, no woodchips have been found in the mortar. It therefore seems more likely that the builders prepared the parts of the tie band on the ground and only put them together up on the walls. 5: 35

Wooden elements of altar screens and ciboria The altar screens (templa) and ciboria in pre-Mongol Novgorod were wooden.

Studies by Grigorii Shtender, Valentina Kovaleva, Vladimir Sarab’ianov and Tat’iana Chukova have provided us with a lot of information about the traces of altar screen and ciborium constructions in the pre-Mongol architectural monuments of Novgorod. 6: 88, 96; 7: 55–64; 8: 312–359; 9: 26–30, 35–44, 76–80 During archaeological fieldwork and restoration study of some churches, the traces of the main elements of altar screens (lower beam, partition, upper beam) have been found. We know from the traces of beams in the base of the altar screen that usually the beam was put in place before the floor was made. Also, we can mention the imprints of the wooden pillars of the partition in the plaster of eastern church pillars and traces of the fastening of the wooden pillars in the upper beam. The partition was possibly made from wood as well as the pillars, but fabric dividers could also have existed. The upper beams of the templon still exist in some churches, while elsewhere there are a lot of imprints or sockets for

125

them in pillars and walls. These upper beams were usually placed at the lower level of ties between the walls and pillars. Sometimes the upper beam was fastened to the ties (Church of the Nativity of the Virgin in the St. Anthony Monastery, 1117–19), or the wooden tie itself was used as the upper beam of the altar screen (St. George’s Church in Staraia Ladoga, 1160s; Annunciation Church on Miachino Lake, 1179). Sometimes the upper beam was not connected to the ties, although it was placed at the same level (Dormition Church in Staraia Ladoga, 1150–60s, Transfiguration Church on Nereditsa Hill, 1198 (Fig. 6)). The upper beam was set in the masonry during the laying of the walls and pillars; the lower beam of the altar screen and partition were installed after the walls and pillars had been made, before the flooring process started.

According to Vladimir Sarab’ianov, all known Novgorodian altar screens were of the open type without additional uprights between the lower and upper beams in the central part. 10: 324 We know two variants for the placement of altar screens. In the first type, the altar screen ran across the whole width of the church (the original screen of St. Nicholas Cathedral, 1113; the Fig. 3: St. Nicholas Cathedral in Novgorod.

1113. Ties in the central drum: a – tier in the middle of windows; b – tier in the base of the drum. Drawing

by Elena Skriptsova

screens of the Church of the Nativity of the Virgin in the St. Anthony Monastery, 1117–19; St. John’s Cathedral in Pskov, 1130s; and the Transfiguration Church on Nereditsa Hill, 1198 (Fig. 6)). In the second type, the screen was situated only in front of the main apse (the main altar screen of St. Sophia Cathedral in Novgorod, 1045–50, the second altar screen of St. Nicholas Cathedral, second half of the twelfth century; the altar screen of the Annunciation Church on Miachino Lake (Novgorod), 1179; and many others). According to the reconstruction by Grigorii Shtender, the altar screen of the St. John the Baptist chapel in St. Sophia Cathedral had an unusual shape – its central part was shifted to the west. 11: 13–15 The screen of the Nativity of the Virgin chapel in St. Sophia had two pillars in the central part. Instead of a lower beam they were inserted into stone slabs (Vladimir Sedov found the same kind of slabs in St. George’s Cathedral at the Iur’ev Monastery in 2014). This type of construction is more typical for a stone altar screen, but the screen of the Nativity of the Virgin chapel was originally wooden.

Considerable height was a distinctive feature of Novgorodian altar screens.

Their height ranges from 360 to 500 centimetres above floor level (most commonly 470–500 cm). Byzantine altar screens were usually lower (250–350 cm above floor level). 12: 325

128

Fig. 7: St. Sophia Cathedral in Novgorod. 1045– 50. Cross-section with a reconstruction of the ciborium. Drawing

by Grigorii Shtender

No pre-Mongol doors have survived. Grigorii Shtender thought that the doors of the Transfiguration Church on Nereditsa Hill were made from two or three vertical oak planks, connected by horizontal members. During the excavations Fig. 8: The Church of the Nativity of the Virgin in the St. Anthony Monastery, Novgorod 1117–19. Original window frame.

iron plates were found, so the architect thought that the doors might have been covered with such plates. 19: 201 This version seems probable, but we do not have much evidence to prove it. Doors of the same type were reconstructed by scholars studying the Transfiguration Cathedral of the Mirozhskii Monastery in Pskov. 20 It is worth mentioning that the western doors of this church were double. The doors were set in the frames using special spikes inserted in the upper and lower edges of the outermost plank.

Window frames The window frames of pre-Mongol churches have rarely survived. Sometimes window openings were filled in with masonry during repair work in later centuries and as a result the window frames or traces of them can be found. These finds can help us to understand the design and characteristics of these wooden elements. 21: 142–143 The window frames found in Novgorodian churches of the twelfth century were made from one or more oak or pine planks and had a thickness of 3–5 cm.

Some vertical planks were fastened by transverse planks placed on the inner side of the window frame (Fig. 8). In St. John’s Cathedral in Pskov, a frame made

129

Fig. 9: The Transfiguration Church on Nereditsa Hill in Novgorod. 1198.

The wooden platform of the upper gallery.

Drawing by Sergei Davydov

from a single plank with a thickness of 3–4 cm and an opening with a diameter of 10 cm was found in the round window of the central apse. 22: 77–78 The round openings in the planks were usually made by axe. Only in the frames of the Nereditsa church can we see cross-shaped and additional triangular openings between the round ones. 23: 191–192; 24: 143 The window frames were not set into the masonry. They were placed in the openings after the dismantling of the centering and then nailed to the parts of wooden ties in the aperture or to special wooden beams set horizontally into the masonry. Sometimes window frames were fastened to hooks that were hammered into those beams.

The outer parts of the window frames were covered with plaster. For better adhesion, cuts were made with an axe on the surface of the planks. The window frames and the church itself were rendered at the same time. 25: 84; 26: 26–26v, 52v

Supporting beams and barriers of the galleries The upper galleries (catechumena) in buildings of the eleventh century and early decades of the twelfth were built on vaults. The first time timber was used as a support for the galleries was in St. John’s Cathedral in Pskov. One of the tie beams which connect the western pillars carries the platform of the galleries, which was made from beams laid in a west–east direction. The upper beam of the railing was placed one metre higher than the ties. 27: 99 Nevertheless, in the architecture of the mid-twelfth to mid-thirteenth centuries, the corner chambers of the catechumena often had a vaulted base. For example, in St. George’s Church in Staraia Ladoga the corner chambers remained on vaults, while the wooden walkway in the western part was made from three large beams, laid close to each other between the doorways of the chambers. 28:102

140

Fig. 8: Gunnarsnås, Dalsland region, Sweden.

Photo, 2007

We can find it even in a distinctive and very popular group of bell towers in Jämtland and other mid-northern regions (Gästrikland, Hälsingland, Medelpad, Härjedalen) of Sweden (for example, the belltower at Håsjö). The Jämtland bell towers are, however, a very complex subject, and there are even connections to Norwegian stave churches, and there is not enough space here to explore the topic in detail.

One characteristic type of Bohemian bell tower has lost the tower shape and looks instead like a big box with the triangular structure hidden inside. This type was developed in the sixteenth century. They are usually of prismatic

141

shape: Stará Voda (Chlumec nad Cidlinou, perhaps ˇeský Brod, lower stone part around 1604), Vitice (C from the fifteenth century), Vyskytná (Pelhr ˇimov, dendro-dated 1612/13), Lomnice nad Popelkou (dendro-dated 1792/93), Hostín u Vojkovic (Kralupy nad Vltavou, dendro-dated 1712) and Hostivar ˇ (Prague).

There is also a younger group of very small bell towers of this type: Konc ˇice (Chlumec nad Cidlinou, before 1816), Údrnice (Libán ˇ), Drahoraz (Libán ˇ, 1893, copy of an older one), Svatý Jan t. Krsovice (Uhlír ˇské Janovice), Neume ˇtely (Hor ˇovice), Plazy (Mladá Boleslav, dendro-dating of the columnar cladding is 1743/44, triangular frames from 1748/49), Petrovice II (Uhlír ˇské Janovice, built in 1904 to replace a shorter one), Podolí (Prague) dendro-dating of the columnar cladding is 1642/43 but the triangular frames are ˇ epníky (Vysoké Mýto, probably 1780s), Kostelec from 1551/52; repaired 1865), R (Jic ˇín, built or else moved from another place in 1880), Sutom (Lovosice), Tr ˇeˇ itonice (Sobotka). On the benice (Lovosice), Polipsy (Uhlír ˇské Janovice) and R Fig. 9: Týn nad Rovenskem, Turnov district, Bohemia. Plan and cross-section (1983).

Photo, 2008

other hand, some of the large bell towers had a six-sided plan (only the bell tower in Sezemice near Pardubice has survived; it was probably built either in the second half of the sixteenth century or in the 1720s, but definitely before 1754).

We do not know of any bell towers of this type in Sweden yet, but theoretically they could exist because there are quite a few with a prismatic shape and their internal structure is very often not known. Viewing from the outside does not reveal the real structural type of these bell towers.

Fig. 10: Stará Voda, Chlumec nad Cidlinou district, Bohemia.

Plan and cross-section (1983). Photo, 2013

This collective monograph presents the current state of research regarding contemporary methods of dealing with historic timber structures in Scandinavia, the United Kingdom, Central Europe and Northwest Russia. The chapters are dedicated to the main aspects of the research and deal with archaeological evidence, written sources, the extant buildings themselves as evidence, as well as repair and maintenance. Researchers from four countries examine centuries-old timber structures that include churches, bell towers and dwellings in Europe and Russia. Based on the conclusions of these studies, they demonstrate various methods of archaeological, archival and empirical research and discuss appropriate measures of restoring and maintaining wooden structures. Historical and contemporary photographs along with new drawings richly document the buildings.

ISBN 978-3-0356-0566-2

www. birkhauser.com