Historic Wooden Architecture in Europe and Russia : Evidence, Study and Restoration [1 ed.] 9783035605426, 9783035605662

Citation preview

Historic Wooden Architecture in Europe and Russia

RZ_wooden-architecture.indd 1

29.09.15 21:21

RZ_wooden-architecture.indd 2

29.09.15 21:21

Historic Wooden Architecture in Europe and Russia Evidence, study and restoration

Edited by Evgeny Khodakovsky and Siri Skjold Lexau

Birkhäuser Basel

RZ_wooden-architecture.indd 3

29.09.15 21:21

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Archives and Historical Documents in Contemporary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

134

Aspects of Bohemian and Swedish Wooden Bell Towers /

Novgorod / Ilya V. Antipov Karel Kuča

RZ_wooden-architecture.indd 4

02.10.15 16:32

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

RZ_wooden-architecture.indd 5

02.10.15 16:32

RZ_wooden-architecture.indd 6

29.09.15 21:21

Introduction 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

RZ_wooden-architecture.indd 7

29.09.15 21:21

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

RZ_wooden-architecture.indd 8

29.09.15 21:21

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.

RZ_wooden-architecture.indd 9

29.09.15 21:22

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.

RZ_wooden-architecture.indd 10

29.09.15 21:22

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

RZ_wooden-architecture.indd 11

29.09.15 21:22

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

RZ_wooden-architecture.indd 12

29.09.15 21:22

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

RZ_wooden-architecture.indd 13

29.09.15 21:22

14

similarity in the construction of Bohemian and Swedish bell-towers of the ‘triangular type’. Thus we have here a methodology that proposes an approach to an architectural construction itself as source and evidence. This makes it possible to take a fresh look at certain aspects of the history of construction in central and northern Europe in the early modern era, when in the course of the Thirty Years’ War (1618–48) there was intensive movement and mixing not only of European armies, but also of ideas and technologies. Russian wooden architecture is allotted a separate place in this book. In Russia, as in Norway, world-famous old timber churches still survive and both countries view wooden construction as one of the most important facets of their national culture. Besides, geographical proximity, similar climatic conditions and an abundance of constructional timber meant that Norwegian and Russian builders would inevitably turn first of all to timber, which became the foundation of a very ancient building tradition. Of importance here too are the historical paths of Fig. 9: Church of the Nativity of the Virgin, Gimreka, Lake Onego area, Leningrad region. 1695. Photo, 2013

Scandinavia and Rus’, which constantly intersected over the centuries. The intensive contacts between Slavs and Scandinavians as far back as the pre-Christian period are generally known and confirmed by numerous written sources and archaeological finds in Staraia Ladoga, Novgorod and along the whole route ‘from the Varangians to the Greeks’. The adoption and spread of Christianity in Norway and Kievan Rus’ took place practically simultaneously over the course of the eleventh century. In 1251 and 1326 Novgorod concluded historically important treaties with Norway. The subsequent process of economic and ethno-cultural interaction continued in the sixteenth to nineteenth centuries due to the active use by Pomors and Russian merchants of the circuitous sea route from the White Sea to western Europe past the Kola Peninsula and the Norwegian coast. The question of the mutual influence between the building traditions of the two countries would seem to raise itself, but long-standing efforts to find analogous buildings or at least evidence of indirect influences have so far not produced any convincingly proven results. The outstanding architect and restorer Vladimir Suslov undertook a major attempt to study this issue. As far back as 1886 he went off to Norway, ‘to acquaint myself with the ancient art of our neighbouring country and to obtain a clear idea of how far it was related to our own Northern art’. He concluded, however, that, ‘examining the ancient works of church architecture of the one country and the other, we will hardly see

RZ_wooden-architecture.indd 14

29.09.15 21:22

15

Fig. 10: Refectory Church of St. Nicholas in the Muezero Monastery, Republic of Karelia, Russia. 1602. Photo, 2013

even the slightest similarity’. 6: 6, 9 This is entirely explicable as Russia and Norway have their own distinctive traditions that developed initially from Slavic and Scandinavian house construction and then under the jurisdiction of the Roman Catholic and Graeco-Byzantine Churches respectively. As a result, the study of the wooden architecture of the Russian North and Scandinavia makes it possible to see how, while employing one and the same material and exploiting its properties, these adjacent cultures produced buildings that are completely different in appearance. Besides the geographical viewpoint that proposes viewing works of wooden architecture in several European countries and the northwestern region of Russia, the title of the book includes the key words ‘evidence, study and restoration’, implying the exposition of the main approaches to the general issues connected with wooden architecture. The traditional conception of archaeological evidence and the written (archival) document as primary sources for the history of lost buildings or individual parts of them is presented in the first three chapters. They formulate questions relating to the archaeological study of Norwegian, English and Early Russian (Novgorodian) material and give summarising descriptions of the written testimony about timber construction in medieval England, the Nordic countries (Norway and Iceland) and Russia. However, in view of the frequent insufficiency of archaeological and written data, as well as difficulties in interpreting them, the authors examine the surviving buildings themselves as ‘silent’ but highly informative pieces of evidence. The analysis of an architectural construction becomes analogous to the reading of a written document or the archaeological uncovering of material evidence. Study of structures, techniques and building methods carried out on location makes

RZ_wooden-architecture.indd 15

29.09.15 21:22

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

RZ_wooden-architecture.indd 16

29.09.15 21:22

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.

RZ_wooden-architecture.indd 17

29.09.15 21:22

RZ_wooden-architecture.indd 18

29.09.15 21:22

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

RZ_wooden-architecture.indd 19

29.09.15 21:22

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.

RZ_wooden-architecture.indd 20

29.09.15 21:22

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 (þelliviði; 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

RZ_wooden-architecture.indd 21

29.09.15 21:22

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.

RZ_wooden-architecture.indd 22

29.09.15 21:22

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.

RZ_wooden-architecture.indd 23

29.09.15 21:22

24

To prepare for a new edifice on a site, the builders had to refill older pits, then level and compress the ground. On a stony site, embedding posts evenly could be difficult; one solution was to reduce the number of pits. This, it was hoped, would prevent differential settlement of the building from heaving and thaw weakening in frost-susceptible soils during the spring. The roof is a sloped structure transferring loads to the walls below. The first measurements set out on the site would be the span of the central roof; the ratio of span to height in known churches gives the gable an angle of 55–58°. A heavy-duty base was required to support such a superstructure with substantial weight and a substantial potential overturning moment. The roofing was thus the cause and the ground-set footing an effect.

Fig. 8: Borgund Stave Church, roof

The ground-set footing The builder would set out coordinates for the pits. A pit was dug for the first bed stone in the nave, to provide adequate transfer and distribution of building loads to the underlying soil. Three more pits were dug and the other three bed stones were roughly placed and adjusted to the same level. The four bed stones would, in rugged terrain, provide fulcrums for a solid cut lumber floor, a levelled rectangle of four raft beams of equal thickness. Thus the raft beams had an independent bearing and did not hang on the sill. On the contrary, the beams extended out to carry the outer wall sills and thereby keep the walls together as well.

The floor Fig. 9: Bed stone and raft beams

To install the roof, a flat base surface has to be provided, upon which upright staves can be secured, spaced apart in rows. Simple spanning half-round timbers were laid flush with the beams, usually without floor joists. Underneath was the crawl space, an underfloor space that was not habitable. As a structural system, the floor would have an adequate capacity to resist lateral loads by constituting a horizontal diaphragm, being a spread footing. Prior art differs substantially from this solution. The floor would be the reference plane, the working platform, from which width and angular base dimensions were derived, both horizontally and vertically. All components had to be built to required standards; the floor was level, staves were plumb, joints were tight, and finished dimensions were as planned. The predictability of the position of joints made it possible to measure, cut and finish most elements of stave churches in advance, prior to the assembly of the parts and erection of the building. Critical connections are where the roof system connects to the supporting walls, where walls connect to each other, and where

RZ_wooden-architecture.indd 24

29.09.15 21:22

25

walls connect to the foundation. To assemble the centre volume, heel tenons in the staves were inserted into cut-outs in the surface of the raft beams. When erected, the staves would be kept in line at the same level, with even distances in two directions. Scribed lines, cut-outs and pegs would give permanent relative positions of elements and datum points for joints at any horizontal cross-section in the vertical plane, including the roof framing. Structures could then be located in reference to the floor and the nearest intersection of the arbitrary axes. It seems from Lomen Stave Church that the carpenters made use of a predetermined modular dimension. When initially marked in the floor, it would give the spatial relationships between the parts, and their absolute dimensions. Ratios thus defined may be either rational or irrational; dimensional accuracy would generally be highly self-consistent. In addition, sequences of dimensions, determined as multiples of an elementary modular unit, are explicitly incorporated following a prescribed formula. This method would give approximation of some irrational proportions, using a direct geometric construction as the initial design basis. This coherence is documented and discussed elsewhere, in an analysis of Lomen Stave Church. 12 In the centre of the nave only limited shear reinforcement was required to make the construction sideways resistant. Open bays between the columns were

Fig. 10: Borgund Stave Church, protecting gallery Fig. 11: Borgund Stave Church, floor

RZ_wooden-architecture.indd 25

29.09.15 21:22

26

created and internal adaptability was provided. To ensure that the incomplete structure was stable during assembly, the erection had to be carried out in the right sequence. Staves were kept in plumb with bracing elements like props, brackets, knee brackets and diagonal braces. In all four walls in the centre room, ‘St. Andrew’s Crosses’ of braces between the upper and lower wall plate function as lattice girders. The outer walls transferred lateral loads to the frame through in-plane shear stress, secured by braces. There are still 28 medieval ground-set wooden churches left in Norway. Some of them have been described, like Kaupanger, 13 Nes, 14 Lomen 15 and Urnes. 16

Conclusions

Rafter span may have been cause, and choice of footing the effect, for the design of the high-rise construction of the ground-set churches. The parallel traditions of footing constructions may have been looked upon as good solutions in their own right. The earth-fast footing was probably gradually replaced by the groundset approach in wooden churches after the year 1100.

325

324 164

355 356

16 3

501

502

Fig. 12: Elevation of Lomen Stave Church Fig. 13: Borgund Stave Church, nave, north-east corner

RZ_wooden-architecture.indd 26

29.09.15 21:22

27

Notes 1 Chouquer, G., F. Favory, Les arpenteurs romain. Théorie et pratique, Paris: Éditions Errance, 1992; Lucas, A.T., ‘The Plundering and Burning of Churches in Ireland, 7th to 16th Century’, in: Rynne, E. (ed.), Essays in Commemoration of Monsignor Michael Moloney, North Munster Studies, Limerick: Thomond Archaeological Society, 1967, 172–229; Keynes, S., ‘The Vikings in England, c.790–1016’, in Sawyer, P. (ed.), The Oxford Illustrated History of the Viking, Oxford: Oxford University Press, 1997, 48–82; Zimmermann, W. H., ‘Pfosten, Ständer und Schwelle und der Übergang vom Pfosten – zum Ständerbau – Eine Studie zu Innovation und Beharrung im Hausbau’, Probleme der Küstenforschung im südlichen Nordseegebiet, 1998, 25: 9–241. 2 Higham, R., P. Barker, Timber Castles, London: Batsford, 1992; Milne, G. (ed.), Timber Building Techniques in London c.900–1400, London & Middlesex Archaeological Society, Special Paper 15, London, 1992; Shirley, E. A. M., The Construction of the Roman Legionary Fortress at Inchtuthil, British Archaeological Reports, British Series 298, London, 2000; Jungmann, J. A., The Mass of the Roman Rite, I–II, Dublin: Four Courts Press, 1950/1986. 3 Wood, S., The Proprietary Church in the Medieval West, Oxford/New York: Oxford University Press, 2006, 918. 4 Dietrichson, L., De norske stavkirker, Christiania: Alb. Cammermeyer, 1892, 96–97. 5 I Corinthians 3:16. 6 Coyne, R., A. Snodgrass, ‘Is Designing Mysterious? Challenging the Dual Knowledge Thesis’, Design Studies 12, 1991, 3: 124–131 (129ff). 7 Shelby, L. R., ‘The Geometrical Knowledge of Medieval Master Masons’, Speculum, 1972, 47: 395–421; Evans, G. R., ‘The “Sub-Euclidean” Geometry of the Earlier Middle Ages, up to the Mid Twelfth Century’, Archive for History of Exact Sciences, 1976/77, 16: 105–118; McCade, J., ‘Problem Solving: Much More Than Just Design’, Journal of Technology Education 2, No. 1, 1990, 28–42. 8 Meeson, R. A., C. M. Welch, ‘Earthfast Posts: The Persistence of Alternative Building Techniques’, Vernacular Architecture, 1993, 24: 1–17; Zimmermann, W. H., 1998. 9 Ahrens, C., Die frühen Holzkirchen Europas, Vol. I–II, Stuttgart: Konrad Theiss, 2001. 10 Jensenius, J. H., Trekirkene før stavkirkene. En undersøkelse av planlegging og design av kirker før ca. år 1100, Con-Text 6, Oslo: Arkitekthøgskolen i Oslo, 2001, 91–100. 11 Ibid., 121–171. 12 Jensenius, J. H., Lomen stavkirke. En matematisk analyse, Oslo: Riksantikvarens Skrifter 5, 1988, 25 ff. 13 Bjerknes, K., H.-E. Liden, ‘The Stave Churches of Kaupanger’, Norwegian Antiquarian Bulletin, No. 1, The Central Office of Historic Monuments, Oslo, 1975. 14 Christie, H., Nes Stavkirke. The Stave Church of Nes, Riksantikvarens Skrifter nr. 3, Riksantikvariatet, Oslo, 1979. 15 Jensenius, J. H., 1988. 16 Christie, H., Urnes stavkirke: den nåværende kirken på Urnes, Oslo: Pax, 2009; Krogh, K. J., Urnesstilens kirke. Forgængeren for den nuværende kirke på Urnes, Oslo: Pax, 2012. All illustrations are by the author except where noted.

RZ_wooden-architecture.indd 27

29.09.15 21:22

28

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 Our understanding of timber churches in medieval England has been based upon two assumptions. The first is that many, perhaps most, masonry churches were preceded by a timber structure. Many parish church histories claim without any evidence at all that the present stone buildings occupy the site of earlier wooden places of worship. Architectural historians have adopted a similar perspective and in particular have identified the years between 1050 and 1150 as the period of the ‘Great Rebuilding’ in England, when timber churches were swept away and replaced with masonry. 1 Excavation has provided some evidence for that view. Three out of four excavations below stone churches in Essex appear to have found some evidence for timber predecessors. However, we must treat these results with caution. Excavations beneath churches elsewhere have found no signs of earlier timber structures and Essex may not be representative of wider practice in England, for, as John Blair has noted, the county had a very strong tradition of timber church building which persisted into the later Middle Ages. 2 So while there was undoubtedly a period in the eleventh and twelfth centuries when many churches were constructed, it is far from certain that this was a period of a Great Rebuilding. 3 The second assumption is that masonry is technologically superior to timber construction, and that just as iron tools replaced bronze implements in prehistory, stone superseded wood as a building material. It follows that, where communities continued to build in the inferior medium of wood, it was because they lacked the resources to build in stone. 4 This belief may seem to have some support from a recent survey of surviving timber churches which showed that the great majority of these were constructed as chapels, and therefore, it could be argued the builders lacked the resources available to those founding parish churches. 5 This line of thought is examined further below, but it is sufficient here to note that it does not provide a complete explanation. These two assumptions have coloured almost everything that has been written about timber churches. For example, the late thirteenth- or early fourteenthcentury church (1293–1318) at Rushton Spencer (Staffordshire) is interpreted as a rare survivor of a once-widespread early medieval practice of plank-walled construction. The argument is that if this church was being constructed at a relatively late date in this way in timber, it must be due to the persistence of earlier forms of building. 6 In this paper, an alternative approach is used. Instead of regarding building materials solely in terms of technological progress, it is argued that timber was chosen as a construction material in certain circumstances for its symbolic character. Some preliminary steps have already been taken in this direction. It has been suggested that stone buildings in early medieval England carried an association with Roman practices. Roman ruins could still be seen in the English countryside, and the church sought to create a connection with that past and so through

RZ_wooden-architecture.indd 28

29.09.15 21:22

29

the classical past with the papacy in Rome. The longevity of stone churches within the landscape provided a contrast with the impermanent character of early medieval timber houses and farm buildings, which were prone to rot and decay and had a relatively short life. Church buildings were therefore symbolic of the undying spiritual world represented by the Roman church. 7 However, if that were simply true, why were any churches built in timber? In fact, the association of churches and stone construction was strengthened in the eleventh century as masonry construction became more widespread. The gradual disappearance of such timber was part of the wider replacement in the mideleventh century of the wooden elements of religious buildings. Timber fonts and altar tables were superseded by those of stone. 8 Stone became established as the material de rigueur for the construction of churches in a way it had not been in earlier centuries. The present study takes a broad look at the use of timber in church construction in the Middle Ages in England and considers the evidence in two phases. The first phase looks at those timber buildings of the eleventh century or earlier which are known mainly from excavation. The only exception is the church at Greensted by Ongar (Essex), which remains as a standing building from a period when timber construction was widespread. The second phase includes those buildings from the thirteenth to sixteenth centuries. They are known from standing remains or, if no longer surviving, from topographical drawings. We need to first examine the evidence for the presence of timber churches in the Middle Ages, and the assumption that these typically were constructed before masonry churches. Second, we need to consider the character of timber churches and how they relate to wider practices of timber construction. Third, we need to examine the reasons why timber continued to be used for churches after the eleventh century.

The evidence for early medieval timber churches The evidence for English timber churches of early medieval date is scarce, and much of it is circumstantial. Because it is assumed that many stone churches had timber predecessors, almost any post-hole or slot found below the earliest phase of masonry has been interpreted as part of a timber building. The argument has developed a circular character in which slight evidence is interpreted as traces of timber churches and this reinforces the impression that such buildings commonly predated masonry structures.9 The evidence from Essex, already mentioned, illustrates this most clearly. A timber phase preceding the twelfthcentury stone church was identified at Little Oakley on the basis of two postholes and a grave; at Cressing two separate buildings were suggested by six unaligned post-holes; the evidence at West Bergholt was three post-holes suggesting a corner of a structure. 10 At Asheldham in the same county, a timber church was identified on the basis of a single wall trench in 1975–76, but when further excavations were possible some years later, no trace of such a building was found, casting doubt on the original conclusions. 11

RZ_wooden-architecture.indd 29

29.09.15 21:22

30

Such slight remains cannot be confidently identified as ecclesiastical. Even when excavations have uncovered more substantial building remains, it is not always possible to conclude that the recorded structures were in fact churches rather than secular buildings. The two buildings at Nazeingbury (Essex) lay adjacent to a cemetery with burials of the seventh to ninth centuries, overwhelmingly of females. They were identified by the excavator as two successive timber churches serving a nunnery, but that is by no means a necessary conclusion. There is nothing distinctive which allows the buildings to be identified as churches and indeed the presence of a hearth inside one of the buildings makes the interpretation seem rather improbable. 12 Timber structures found beneath the masonry footings of All Saints, Oxford were almost certainly not from an earlier church. The latest of these has been identified as belonging to a cellar of a commercial building. 13 The difficulties of identification presented by the features beneath the earliest stone church at Wharram Percy (East Riding, Yorkshire) are rather different. The post-holes present such an incoherent plan that it is impossible to make any plausible interpretation. It is indeed uncertain whether they represent a single structure. The excavation report and a number of subsequent writers have been rather cautious in describing these remains as a timber church. 14 There are similar problems with the recognition of a plan at Beckery Chapel (Somerset) and St. Bertelin’s Church, Stafford where no structure could be identified, and it is uncertain whether the post-holes belonged to any single building. 15 More substantial evidence for timber buildings is limited to a few sites. Although known mainly from aerial photographs and geophysical survey, and very limited excavation, there seems little doubt that the building with an apsidal east end at Foxley (Dorset) is a church. The only buildings with this form are churches and the presence of human bone in the surrounding ditch tends to support the identification. A seventh- to ninth-century date is probable. 16 Two successive churches have been identified at Brandon (Suffolk) dating to the early- to mid-eighth century and mid-eighth to early-ninth century respectively. Although the plans resemble contemporary domestic buildings, the presence of a ritual horse burial in the earlier church, the absence of domestic waste, their position outside the main enclosure and, most of all, the adjoining cemetery, allows them to be recognised as ecclesiastical buildings. 17 Very little evidence survives for the early tenth-century timber church at Burnham (Lincolnshire), since traces were largely removed by the construction of the subsequent masonry church. This must have always been a minor ecclesiastical building, never achieving parochial status and finally being abandoned in the mid-sixteenth century. The remainder of the church lies beneath the masonry building which replaced the timber structure in the late tenth century. It is thought to have been a single-cell timber structure measuring only 5.5 m by 4.1 m. 18 Similar, somewhat vestigial evidence remains for the church of St. Martin at Palace in Norwich. The first phase of posts set in trenches is not considered here as too little was excavated to allow meaningful comment. The second phase, dating to the mid-eleventh century, comprised six post-holes set east–west and a

RZ_wooden-architecture.indd 30

29.09.15 21:22

31

single post-hole marking the west return. These

a

were regularly placed about 1.5 m apart (measuring centre to centre). The timber posts which had a square or slightly rectangular cross-section were centrally set within the post-holes. 19 soakaway

The two buildings found at Thetford (Norfolk) and in the bailey at Norwich Castle (Norfolk) can be considered together because of their similarities 0

5 m

of form. Both are small, two-celled buildings aligned east–west. Almost the whole plan of the church was recovered at Thetford and evidence of individual

b

timber ‘ghosts’ was also recorded (Fig. 1a). The east font base & soakaway

end was marked by a single post, but similar absence baby burials

of archaeological evidence for the ends is found in many contemporary buildings. This makes the number of timbers in the west wall of the nave more notable, since they served no structural purpose, but were apparently intended to provide an impressive facade adjoining the entrance. Although little evidence of the timbers of the side walls re-

c

mained, to judge by the two in the north wall of the baby burials soakaway

nave, these were also close-set, as were those in the chancel. The Thetford building was superseded by a somewhat large stone building of similar pattern and the presence of burials allows no doubt that it was a church. 20 The series of buildings found around Norwich

Fig. 1: Three excavated 11th-century churches – a. Timber church at Bury Road, Thetford (after Dallas). b. Timber church at Norwich Castle (after Ayers). c. Stone church at Raunds Furnells (after Boddington). Redrawn by Libby Mulqueeny

Castle are probably attributable to the tenth and eleventh centuries, although the excavators were somewhat cautious about ascribing a date. The remains of the first phase (Building A) were so vestigial that no useful discussion is possible. The second phase was formed from four pairs of post-holes representing a structure 5.5 m long and 3.2 m wide. This was superseded by a third building which was certainly a church with separate nave and chancel (Fig. 1b). The side walls were marked by trenches and the end walls by sparse post-holes. The position of the doorway on the south side is evident from two post-holes which were inset from the line of the wall, and that of the font is marked by a soakaway at the west end. Two foetuses were buried at the junction of the nave and chancel. The excavators note the existence of an infant grave in a very similar position in the contemporary stone church at Raunds (Northamptonshire), suggesting that these burials may be examples of a wider ritual (Fig. 1c). 21 The oddly shaped building at Potterne (Wiltshire) marked by wall trenches was identified as a church by the excavator, but it has been difficult to interpret until recently since it has a form which is apparently unlike any other ecclesiastical building, or indeed any secular structure. However, Rodwell has interpreted the remains as a turriform (tower-like), otherwise called a tower-nave

RZ_wooden-architecture.indd 31

29.09.15 21:22

32

church built of timber. Almost thirty other examples of this type are known, but Potterne remains the only timber tower-nave church. Rodwell has outlined one possible scheme of development, but this remains somewhat speculative. 22 The final building we need to consider in this group of small early churches is the standing building at Greensted by Ongar (Essex) (Fig. 2). This building has provoked an immense amount of speculation and discussion in the past, but some clarity is beginning to emerge. The building has now been dated by dendrochronology to the later eleventh Fig. 2: The timber church at Greensted by Ongar

century, which lays to rest improbable claims of its antiquity. Excavation established that the close-set vertical timbers which form the walls had originally been set in a trench. Sometime before 1845 the lower parts of the timbers were set on a sole-plate which stood on a dwarf wall of flint. This had become rotten and about 150 mm of the vertical wall timbers were cut away, set in a new soleplate and a brick wall set beneath. 23 The method of construction using contiguous posts or staves is unusual, but it has been suggested that the hall at Goltho (Lincolnshire) and that at West Cotton (Raunds, Northamptonshire) were built using a similar technique. 24 The timber churches considered here, with the possible exception of St. Bertelin’s, Stafford, were minor structures. Yet timber was also used for major buildings. The abbey church of Bury St. Edmund’s (Suffolk) was constructed of wood until replaced by a masonry building in the early, or according to other sources, in the later eleventh century. 25 The cathedral church at Elmham (Norfolk) in the eleventh century was also said to have been built of timber. 26 The church of St. Alkmund in Derby is called Wythechirche, a spelling which suggests it means ‘withy church’. 27 There is archaeological evidence that the great timber church St. Martin’s, (Dover, Kent) began as a timber structure and was rebuilt on four occasions in wood before being replaced by a stone structure in c. 1070. Its location in the Roman Saxon Shore fort must have meant that there was abundant spolia (reusable masonry) for a stone structure, if one had been desired. A floor of opus signinum was also laid down during the third phase of timber building, which suggests that there was no lack of knowledge of the use of mortar. Persistence in the use of timber therefore seems like a deliberate choice. It is unfortunate then that both the interpretation and the dating of this important church are so problematic. According to the excavator’s interpretation, the first-phase building remained standing for a considerable period and the later phases were erected around it until finally the whole church burnt down at the end of what was defined as ‘Period 3’. The preservation of the earliest phase structure suggests that it had a symbolic importance, and the continuing use of timber may have been consciously conservative in its symbolism.

RZ_wooden-architecture.indd 32

29.09.15 21:22

33

The Period 4 building which succeeded the fire needs to be discussed at greater length, since it had an unusual form of construction (Fig. 3). The east wall was not excavated, but the remainder of the walls were formed of alternating planks set a short distance apart. The planks were marked by different coloured soil, and it was not possible to identify the position of all of these. It was not apparent how the gaps between the vertical planks were filled; 0

5

10  m

it may have been with wattlework or horizontal boards. The south wall was clearly constructed in a

Fig. 3: The Period 4 timber church of St. Martin’s, Dover (after Philp). Redrawn by Libby Mulqueeny

series of short lengths about 3 m long which were presumably joined to form a continuous side. It is possible that the north wall was built in a similar manner, but too little survived to indicate this. The church was unusually wide with an internal breadth of 8.65 m. 28 The evidence of these buildings suggests that timber was widely used for church construction before the eleventh century, but at present we lack any way of knowing what proportion were built in wood and in stone. The standing buildings of this period, with the exception of Greensted, are all stone churches, but we cannot draw any conclusions from that, since it is a reflection of differential survival of materials. 29 It is apparent that builders had a choice of materials and while many opted to use stone, others used timber even for grand buildings. The significance of this eludes us, but the ideas that timber was chosen for the sake of economy and that stone had associations with the permanence of the Roman church are both inadequate without qualification.

Timber churches and the wider practices of medieval building It is impossible to interpret the character of early medieval timber buildings without drawing from a wider body of evidence which includes domestic houses. The construction techniques used in churches belong broadly to the wider vernacular tradition of timber building, particularly in the use of trenches to seat a series of vertical posts. English timber buildings from the fifth century onwards had been constructed with ground-fast timbers set either in individual post-holes or in wall trenches. 30 There is some indication that by the eleventh century vertical timbers were beginning to be set on the ground surface in grooved sole-plates. The posts in early medieval buildings were comparatively close-set, certainly by comparison with later medieval buildings. They were carefully aligned and were linked at their head by a wall plate, or horizontal timber. Walls were probably formed from a number of lengths, each a separate panel with its own wall plates. It remains unclear how these were joined, and particularly how the end walls were linked to the side walls. The use of diagonal ties (dragon ties) is suggested by the fact that adjoining walls were often separated by a small distance, and the wall trenches do not always meet. It is common for each post in the side wall to be matched by a corresponding post on the opposite wall, and it is possible that each pair of posts supported a rafter

RZ_wooden-architecture.indd 33

29.09.15 21:22

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

RZ_wooden-architecture.indd 34

29.09.15 21:22

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

RZ_wooden-architecture.indd 35

29.09.15 21:22

36

Fig. 4: The north side of St. Peter’s Church, Melverley (Shropshire), showing the use of close studding even on the least significant face of the church

at Frimley (Surrey), demolished in the 1820s, and that at Plaistow (Sussex),

Fig. 5: Assembly marks with the numbers ‘iii’ and ‘iiii’ on the mid-rail and studs of the north side of St. Peter’s Church, Melverley (Shropshire). The studs in each bay were numbered from the west to the east with corresponding figures inscribed on the mid-rail

a parish church that received the tithes of the congregation and had the rights

replaced in the 1850s, extend the distribution of the Hampshire group further eastwards, but this does not materially alter the picture. Perhaps the most striking aspect of later medieval timber buildings is that they were overwhelmingly constructed as chapels rather than parish churches. From the late eleventh century onwards, a sharp distinction was made between of baptism and burial, and a chapel which generally had none of these but which was founded for the convenience of a lord or local people who were distant from the main church. 37 There were some exceptions to this rule. Some ancient chapels had the right to baptise, hear confessions and bury parishioners, but these were a minority. 38 It is rare after this date to find a parish church built in timber, but some chapels were so constructed. The usual explanation is that the worshippers in chapels lack sufficient wealth to build in masonry. A chapel received no income from tithes and had to be erected and maintained either by a patron or by local people. Many chapels in due course were raised to be parish churches in their own right, but this was often many centuries after their foundation. However, there is no evidence that communities saw their timber chapels merely as a temporary structure, a step on the way towards the construction of a stone church. The quality of construction of timber chapels was notable. The use of close-set studs in Melverley church (Shropshire), probably built in the early sixteenth century, is not a mark of economy, but of display. 39 The studs were unusually broad, indeed equal in width to the weight-bearing posts, so that the walls presented an appearance of alternating timbers and wattle-anddaub panels (Figs. 4, 5). Equally, the chapel built at Denton (Cheshire) in the 1530s was constructed in the vernacular style of the north-west of England with close-set studs and diagonal braces in the panels. With these buildings in mind, we can suggest that the church at Rushton Spencer, which had walls formed from planks, was not so much a late survival of an archaic practice of construction, but part of a recurring pattern of church-building which made conspicuous and even extravagant use of timber. The communities that erected all these buildings were not seeking to economise, but to build elaborate structures (Fig. 6).

RZ_wooden-architecture.indd 36

29.09.15 21:22

37

2

6

metres

feet

2

6 0

groove for boarding

N

Fig. 6: The timber church of Rushton Spencer (with permission of Bob Meeson)

If economy was not a clear consideration in the decision to build in timber, what might have been the factors? The use of timber for ecclesiastical buildings was unusual by the thirteenth century. Construction in timber, in effect, acknowledged that a chapel was different in status to the parish church and provided a distinction between the place at which the parishioners worshipped and the superior church to which they paid their tithes. This contrast is evident also in a small number of chapels founded to provide a continuing ritual presence at a former religious site. The chapel of Warburton (Cheshire) originated in a cell of Premonstratensian monks which had been founded in the 1190s and lasted only a few decades. A timber chapel was maintained at the site there so that mass could continue to be said for the soul of the donor of land. 40 Similarly, the Knights Hospitallers’ preceptory of Halston (Shropshire) had been founded in the mid-twelfth century, but by at least the early fifteenth century was at lease. The present timber-framed chapel dates from about that time, although there had been a chapel before then. 41 There were, however, further factors influencing the way timber was used. Church buildings were constructed for the glory of God and work of the highest quality was expected. The employment of abundant timber reflected this imperative and served to differentiate churches from domestic buildings. At least half of the surviving timber chapels were built in the fifteenth or early sixteenth centuries. This was a period of contracting population when a number of churches were abandoned, and was not an obvious time to be constructing new chapels. The explanation is that few of these chapels were new at this period; many were simply reconstructions of earlier buildings. Rotherwick (Hampshire) has a chancel of the late thirteenth century, but the nave is a timber-built structure of the fifteenth century. It seems to have originated in

RZ_wooden-architecture.indd 37

29.09.15 21:22

38

a chapel built for the manor house, but must have been more widely used, as in the later fourteenth century the parishioners were instructed by the bishop that they should attend the parish church at Odiham. 42 Similarly, the timberframed chapel at nearby Mattingley was constructed c. 1500, but there had been a building on the same site since at least 1387. 43 These chapels replaced earlier structures and had probably been built because the preceding timber buildings had rotted and had to be replaced.

Conclusions

Two main conclusions have emerged from this study of timber churches in England. The first is that timber was not just a second-rate material adopted for reasons of economy. Indeed, we have seen that in the early Middle Ages very substantial churches might be built in wood until the eleventh century, when there was a decisive shift in favour of masonry, and thereafter most new parish churches were built in stone. The timber buildings which persisted or were built anew in the later Middle Ages were largely chapels and seem to have been constructed in that material to distinguish them from parish churches. There was a recognition of hierarchy of places of worship and in some parts of the country, though not everywhere, it was felt appropriate to acknowledge this in the building materials. The second feature which has been touched upon is that timber churches frequently used wood in a different manner to secular buildings. From the earliest churches onwards, the walls seem to have been constructed with close-set or even contiguous timbers. The tradition of using wood in this way appears to have persisted and spans the gap between the early medieval examples, largely found in excavation, to the late medieval examples known from standing structures. Close-set studs (wall supports) are often used on the front of secular buildings as a form of display, but are rarely used on the less visible portions where economies could be made. However, in timber churches no such compromise was made. At Rushton Spencer, for example, both the more visible south wall and the north wall had continuous planking along their length. Indeed, the recorder found it difficult to find parallels for the use of vertical planks for the walls of the building. 44 Close studding can be found on every face of the churches at Melverley and at Denton. There was similar extravagant use of timber in the standing building at Greensted where the walls unusually were constructed of adjoining staves, and in the excavated church at St. Martin’s, Dover, where the timbers were placed alternately. We might question how far this was a persistent tradition of construction, or whether it reflected the continuing very high quality of work expected in all churches, however minor. It has been possible here to provide only an overview of some of the issues related to timber churches in medieval England. The subject has been largely overlooked by architectural historians and archaeologists alike, but this paper has sketched the outlines of a problem which deserves considerably more attention.

RZ_wooden-architecture.indd 38

29.09.15 21:22

39

Acknowledgements

I am grateful to Michael Shapland for discussing with me the meaning of timber in the construction of churches, to Bob Meeson for allowing the reproduction of his drawing of Rushton Spencer and to Libby Mulqueeny for preparing the other drawings.

References 1 The phrase was not originally used in this way by its originator – Gem, R., ‘The English Parish Church in the 11th and Early 12th Centuries: A Great Rebuilding?’, in: J. Blair (ed), Minsters and Parish Churches: The Local Church in Transition 950–1200, Oxford: Oxford University Committee for Archaeology, 1988, 21–30. However, later writers have adopted the phrase to refer to the replacement of timber by stone: Morris, R., Churches in the Landscape, London: J. M. Dent, 1989, 165–166; Blair, J., The Church in Anglo-Saxon Society, Oxford: Oxford University Press, 2005, 411–417. 2 The careful excavations at Barton-upon-Humber suggested that the stone church was the first such building on the site and was added to an existing graveyard, Rodwell, W. and C. Atkins, St. Peter’s, Barton-upon-Humber, Lincolnshire: A Parish Church and its Community: History, Archaeology and Architecture, Vol. 1, Oxford: Oxbow Books, 2011, 169–173; Blair, J., ‘Churches in the Early English Landscape: Social and Cultural Contexts’, in: J. Blair, C. Pyrah (eds): Church Archaeology: Research Directions for the Future, York: Council for British Archaeology, 1996, 6–18. 3 Morris, R., ‘Local Churches in the Anglo-Saxon Countryside’, in: H. Hamerow, D. Hinton, S. Crawford (eds), The Oxford Handbook of Anglo-Saxon Archaeology, Oxford: Oxford University Press, 2011, 172–197. On the rhythms of ecclesiastical building, see Gem, R., ‘ABC: How Should We Periodize Anglo-Saxon Architecture?’, in: L. A. S. Butler, R. K. Morris (eds), The Anglo-Saxon Church: Papers on History, Architecture and Archaeology in Honour of Dr H. M. Taylor, London: Council for British Archaeology, 1986, Fig. 96. 4 Mercer, E., English Architecture to 1900: The Shropshire Experience, Almeley: Logaston Press, 2003, 288–289. 5 Nevell, M., I. Hradil, St. Lawrence’s Church and the Archaeology of the Medieval TimberFramed Churches of England and Wales, Ashton under Lyme: Tameside Metropolitan Borough Council, 2005, 30. 6 Meeson, B., ‘Plank-Walled Building Techniques and the Church of St. Lawrence, Rushton Spencer’, Vernacular Architecture, 1983, 14: 29–35. Dates from Vernacular Architecture 29, 1998, 107. 7 Shapland, M., ‘Timber and Stone in Anglo-Saxon Building Practice’, in: M. Bintley, M. Shapland (eds), Trees and Timber in the Anglo-Saxon World, Oxford: Oxford University Press, 2013, 21–44; Gardiner, M., ‘Timber Buildings Without Earth-Fast Footings in Viking-Age Britain’, in: J. Hines, A. Lane, M. Redknap (eds), Land, Sea, and Home: Proceedings of a Conference on Viking Age Settlement in the British Isles, London: Society for Medieval Archaeology, 2004, 351. For much later arguments about the symbolism of wood and stone, see Scott, W., ‘On Wooden Churches’, The Ecclesiologist, 1849, 9: 14–27. 8 Ibid., 37; Blair, J., ‘The Prehistory of English Fonts’, in: M. Henig, N. Ramsay (eds), Intersections: The Archaeology and History of Christianity in England, 400–1200: Papers in Honour of Martin Biddle and Birthe Kjølbye-Biddle, Oxford: Archaeopress, 2010, 149–177.

RZ_wooden-architecture.indd 39

29.09.15 21:22

40

9 A number of the wooden churches listed by Ahrens are somewhat dubious: Ahrens, C., Die frühen Holzkirchen Europas, Stuttgart: Konrad Theiss Verlag, 2001, 163–191. 10 Essex County Council, Four Church Excavations in Essex, Chelmsford: Essex County Council, 1984, 21, 35, 52. 11 Drury, P., W. Rodwell, ‘Investigations at Asheldham, Essex: An Interim Report on the Church and the Historic Landscape’, Antiquaries Journal, 1978, 58: 133–151; Andrews, D., M. Smoothy, ‘Asheldham Church Revisited’, Essex Archaeology and History, 1990, 21: 146–151. 12 Huggins, P., ‘Excavation of a Belgic and Romano-British Farm with Middle Saxon Cemetery and Churches at Nazeingbury, Essex, 1975–6’, Essex Archaeology and History, 1978, 10: 30–117. 13 Dodd, A., Oxford Before the University, Oxford: Oxford Archaeology, 2003, 210–220, 233–236. 14 Bell, R. D., M. W. Beresford, Wharram: A Study of Settlement on the Yorkshire Wolds III – Wharram Percy: The Church of St Martin, London: Society for Medieval Archaeology, 1987, 55–57; Wrathmell, S., Wharram: A Study of Settlement on the Yorkshire Wolds XIII – A History of Wharram Percy and its Neighbours, York: York University Archaeological Publications, 2012, 208; cf. Blair, J., 2005, Fig. 48. 15 Rahtz, P., S. Hirst, Beckery Chapel, Glastonbury, 1967–8, Glastonbury: Glastonbury Antiquarian Society, 1974; Oswald, A., The Church of St. Bertelin at Stafford and its Cross: Excavation Report, Birmingham: City of Birmingham Museum and Art Gallery, 1955. 16 Hinchliffe, J., ‘An Early Medieval Settlement at Cowage Farm, Foxley, near Malmesbury’, Archaeological Journal, 1986, 143: 240–259. 17 Tester, A., S. Anderson, I. Riddler, R. Carr, Staunch Meadow, Brandon, Suffolk: A High Status Middle Saxon Settlement on the Fen Edge (East Anglian Archaeology 151), Bury St. Edmunds: East Anglian Archaeology, 2014, 48–51, 63–66. 18 Coppack, G., ‘St. Lawrence Church, Burnham, South Humberside: The Excavation of a Parochial Chapel’, Lincolnshire History and Archaeology, 1986, 21: 39–60. 19 Beazley, O., B. Ayers, Two Medieval Churches in Norfolk (East Anglian Archaeology 96), Dereham: East Anglian Archaeology, 2001, 5, 55. 20 Dallas, C., Excavations in Thetford by B. K. Davison Between 1964 and 1970 (East Anglian Archaeology 62), Gressenhall: East Anglian Archaeology, 1993, 84–87. 21 Ayers, B., Excavations Within the North-East Bailey of Norwich Castle, 1979 (East Anglian Archaeology 28), Gressenhall: East Anglian Archaeology, 1985, 24; Boddington, A., Raunds Furnells: The Anglo-Saxon Church and Churchyard, London: English Heritage, 1996, 21. 22 Davey, N., ‘A Pre-Conquest Church and Baptistery at Potterne’, Wiltshire Archaeological and Natural History Magazine, 1964, 59: 116–23; Rodwell, Atkins, St. Peter’s, Barton-uponHumber, 307–320. 23 Christie, H., O. Olsen, H. M. Taylor, ‘The Wooden Church of St. Andrew at Greensted, Essex’, Antiquaries Journal, 1979, 59: 92–112; The Builder 7, 27 January 1849, 45. 24 Beresford, G., Goltho: The Development of an Early Medieval Manor c. 850–1150, London: Historic Buildings and Monuments Commission for England, 1987, 38–49; Chapman, A., West Cotton, Raunds: A Study of Medieval Settlement Dynamics AD 450–1450, Oxford: Oxbow Books, 2010, 58–60. 25 Arnold, T., Memorials of St Edmund’s Abbey (Rolls Series 96), London: HMSO, 1890, 1, 19, 84–85. Though the Gesta Sacristarum says that the timber church survived until the later

RZ_wooden-architecture.indd 40

29.09.15 21:22

41

eleventh century, Arnold, Memorials of St. Edmund’s Abbey 2, 289. 26 Luard, H., Historia Anglicana Bartholomaei de Cotton (Rolls Series 16), London: HMSO, 1859, 389. 27 Blair, 2005, 387, n. 72. 28 Philp, B., The Discovery and Excavation of Anglo-Saxon Dover, Dover: Kent Archaeological Rescue Unit, 2003, 58–72, 125–131. 29 Stone churches of this period are listed in Taylor, H., J. Taylor, Anglo-Saxon Architecture 1–2, Cambridge: Cambridge University Press, 1965. 30 Gardiner, M., ‘An Early Medieval Tradition of Building in Britain’ (‘La tradición constructiva de la alta Edad Media en Gran Bretaña’), Arqueología de la Arquitectura, 2012, 9: 231–246. 31 Hinchliffe, J., ‘An Early Medieval Settlement at Cowage Farm’, 243. 32 Rodwell, W., ‘Appearances Can Be Deceptive: Building and Decorating Anglo-Saxon Churches’, Journal of the British Archaeological Association, 2012, 165: 22–60, especially 47–49. 33 Drury, P., ‘Anglo-Saxon Painted Plaster Excavated at Colchester Castle, Essex’, in: S. Cather, D. Park, P. Williamson (eds), Early Medieval Wall Painting and Painted Sculpture in England, Oxford: British Archaeological Reports, 1990, 111–122; Drury, P., ‘Aspects of the Origins and Development of Colchester Castle’, Archaeological Journal, 1982, 139: 302–419. 34 Blair, J., H. Hamerow, ‘Anglo-Saxon Plaster-Infilled Timber Wall’, in: Hardy, A., A. Dodd, G. D. Keevil (eds): Ælfric’s Abbey: Excavations at Eynsham Abbey, Oxfordshire, 1989–92, Oxford: Oxford Archaeology, 2003, 207–209. 35 Blair, J., ‘Archaeological Discoveries at Woodeaton Church’, Oxoniensia, 1999, 63: 221–237. 36 Nevell and Hradil, 2005. 37 Blair, 2005, 384–385; Lennard, R., Rural England 1086–1135: A Study of Social and Agrarian Conditions, Oxford: Clarendon Press, 1959, 297–303. 38 Orme, N., ‘The Other Parish Churches: Chapels in Late-Medieval England’, in: C. Burgess, E. Duffy (eds), The Parish in Late Medieval England: Proceedings of the 2002 Harlaxton Symposium, Donington: Shaun Tyas, 2006, 78–94, especially 83. 39 Mercer, E., English Architecture to 1900: The Shropshire Experience, Almeley: Logaston Press, 2003, 286–289. 40 Baggs, A. P., A. J. Kettle, S. J. Lander, A. T. Thacker, D. Wardle, ‘House of Premonstratensian Canons: The Priory of Warburton’, in: C. R. Elrington and B. E. Harris (eds), A History of the County of Chester, 3, London: Victoria County History, 1980, 171. 41 Angold, M. J., G. C. Baugh, M. M. Chibnall, D. C. Cox, D. T. W. Price, M. Tomlinson, B. S. Trinder, ‘House of Knights Hospitallers: The Preceptory of Halston’, in: A. T. Gaydon, R. B. Pugh (eds), A History of the County of Shropshire, 2, London: Victoria County History, 1973, 87–88. 42 Moger, O. L., F. Kennedy, K. M. Upcott, ‘Odiham Hundred’, in: W. Page (ed), The Victoria County History of Hampshire and the Isle of Wight, 1911, 4: 99–101. 43 Brough, F., ‘Holdshot Hundred’, in: W. Page (ed), The Victoria County History of Hampshire and the Isle of Wight, 1911, 4: 48–51. 44 Meeson, 1983, 29–35. All illustrations are by the author except where noted.

RZ_wooden-architecture.indd 41

29.09.15 21:22

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

RZ_wooden-architecture.indd 42

29.09.15 21:22

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 first sight makes it possible

RZ_wooden-architecture.indd 43

29.09.15 21:22

44

to relate these pieces to the established tradition of monumental woodcarving in northern Europe, known from works of the eleventh to thirteenth centuries found in Norway, Sweden (the island of Gotland) and Denmark. In Rus’, however, in contrast to Scandinavia, extremely few examples of this form of art have survived. The fact that the ‘columns of the oaken St. Sophia and the works of Scandinavian carving belong to the same time period and their common monumental character would Fig. 2: Reliefs on St. George’s Cathedral in Iur’ev-Pol’skii. 1234 11 (Figs. 11; 21)

seem to suggest that the architectural fragments from Novgorod should be viewed as evidence of the influence of Scandinavian tradition on the Novgorodian lands in the first half to the middle of the eleventh century, the era that saw perhaps the most intensive political and trading links between Rus’ and Norway in the entire history of the two countries’ relations. However, the issues connected with the ‘columns of the oaken St. Sophia’ are far more complex. Study of this artefact comes down to the question of attribution and interpretation. An analysis of the ornamental design on the ‘columns’ is the first direction in the history of the study of them; a second line of research interest is determining the specific purpose of these architectural details. Among other things, it has been suggested that the pieces were supports for a gallery, 8 the central pillars of some secular edifice, or parts of the wooden St. Sophia Cathedral that was constructed immediately after the conversion of Novgorod to Christianity in the late tenth century and burnt down in 1049.9 One of the first researchers to address this artefact was Artemii Artsikhovskii. 10 He suggested that the surviving fragments were parts of columns split in two. An initial analysis of the character of the ornament led Artsikhovskii to the conclusion that the gryphon and centaur have analogies among the reliefs on St. George’s Cathedral in Iur’ev-Pol’skii (1234). Nevertheless, the following work devoted to this find, written by Ivan Kudriavtsev, put forward a bold assertion, comparing the ‘columns’ not to Russian, but to Norwegian monumental-decorative carving. 12 The author takes as analogies some of the latest examples of carving in the Norwegian wooden churches of Lardal, Dal and Øyfjell. Kudriavtsev’s approach of seeking Norwegian analogies suggested that the church itself may have been constructed in the frame technique that was not traditional for Novgorod. The northern European tradition of religious architecture developed types of churches characterised by the use of massive pillars decorated with carving at the corners of the building, as, for example, in the famous church at Urnes. The floor in such buildings was not structurally connected to the walls and effectively lay on the ground. Kudriavtsev’s paper had a major influence on the course taken in the study of this artefact – an assumption of its Scandinavian character. The interpretation put forward by Vladimir Konetskii and Kirill Samoilov continues and develops the theme of the Scandinavian building tradition having a presence on the territory of early Novgorod. 15 The building described in

RZ_wooden-architecture.indd 44

29.09.15 21:22

45

Fig. 3: The ‘columns of the oaken St. Sophia’ and the portals from Dal and Lardal 13: 174; 14: 5

their work may have had a structure similar to the church in Hemse. So the study of the ‘columns’ in Russian historiography presents a sequence by which, on the basis of the archaeological data, attempts are made to demonstrate the presence of a Scandinavian constructional and artistic culture in Rus’, although the very style of the carving raises questions, since it is iconographically in no way similar to works of Scandinavian monumental carving from the period. Besides, a Norwegian religious building constructed in Novgorod for visiting merchants and guests should have had obvious elements of Scandinavian decoration in its facades or architectural details. The Scandinavian hypothesis was criticised by Erla Bergendahl Hohler, who applied an approach standard for the study of Scandinavian carving, in which Fig. 4: The church in Hemse, Gotland. Reconstruction by E. Ekhoff 16: 12

woodcarving was formed under the influence of stone-carving. 17 In describing the design, she observes that the style employed for the depiction of mythological beings, and indeed the beings themselves, was not typical for the Scandinavian countries at that time. Analysing the character of the interlace, Hohler states that such a loopless pattern of double strips is often encountered among Novgorodian finds from the twelfth century, something borne out by items of jewellery. 18 She points out that the use of a double line and mythical creatures enclosed in medallions was widespread in Europe, but appeared in Scandinavia only in the thirteenth century (a portal from Mæl). Of particular interest for this researcher are the reliefs on churches in Vladimir from the twelfth and thirteenth centuries, in which she detects a number of elements very closely akin to the pattern on the ‘columns’. Like Artsikhovskii, Hohler adheres to the

RZ_wooden-architecture.indd 45

29.09.15 21:22

46

once popular viewpoint that the carving on the ‘columns’ has much in common with the reliefs on the edifices of Vladimir and Suzdal and surmises that the Novgorodian finds were created under their influence. But is such an approach possible? The main tool in the study of Scandinavian portal carving has been stylistic analysis, which is used not only to establish a typology, but also to determine dates. In the present case, though, we are dealing with an archaeological artefact, the dating of which to the mid-eleventh century is indisputable due to the clearly legible stratification. According to Hohler, wooden sculpture undoubtedly existed in Early Rus’, but practically no fragments have survived from which to compile a definitive picture of it. Such motifs as the gryphon and centaur most probably came from Byzantium, as did the elaborate chain with a double outline. Similar motifs are widespread in the monumental decorative sculpture of western Europe and are found in the reliefs of Vladimir-Suzdalian churches of the eleventh century that took shape under a Germano-Lombardic influence. Taking into account the fact that those reliefs and the carving on the Novgorodian columns have much in common, Hohler dates them to the same time. Thus, in her opinion, the Novgorodian columns are a variant in wood of the Vladimir-Suzdalian carving. One more analogy, not mentioned in Hohler’s work, is the capitals of columns in the Cathedral of Saints Boris and Gleb in Chernigov (first third of the twelfth century), which have an interlaced pattern very similar to the carving on the ‘columns’. 19 At this point approaches to the research diverge, as the methodology of Russian and Scandinavian scholars is different. The study of works of Norwegian wooden portal carving is conducted mainly using stylistic analysis, while, due to their scantness, archaeological data play a subordinate role. Russian researchers view the ‘columns’ as a part of the archaeological picture, one that has a definite place and date, and consider the character of the ornament a secondary matter. The two approaches are equally interesting, but in the present case they result in contradictory opinions. On the one hand, the ‘columns’ are supposedly the peculiar result of the presence of a Scandinavian tradition of monumental decorative woodcarving in early Novgorod. On the other hand, this artefact is not only totally unconnected to Scandinavia, but dates from a different time and was created under the influence of stone sculpture. Both sides have weighty arguments in favour of their point of view. Hohler, not without good reason, insists that the carving on these fragments has little in common with eleventh-century Scandinavian patterns, thus calling into question the generally accepted date. However, in Novgorod there is a system of absolute dating of all strata and finds, which refutes this hypothesis. Even without an exact date for the find, it is impossible to deny the fact that the design on the ‘columns’ really does not match a single style in the Scandinavian art of the period. As a result we have a pair of finds dated to the eleventh century but with a fairly non-standard ornament, especially if viewed in the context of the art of northern Europe. Therefore, despite the fact that a considerable amount has been said about this artefact, there is still no certainty on the two main questions: purpose and decoration.

RZ_wooden-architecture.indd 46

29.09.15 21:22

47

The state of preservation of the fragments makes it possible to express a number of theories suggesting a kinship between these finds and eleventh-century works of Scandinavian monumental carving. The idea that a building put up using Scandinavian construction techniques existed in early Novgorod is feasible, but the putative carved decoration of that building bore no relation to the art of Scandinavia in that period. In the present case we should not pay much attention to details such as the gryphon and centaur, since those elements were very common in works of decorative and applied art, fabrics and miniatures, so finding a true analogy is not a realistic possibility. Of particular interest are the Fig. 5: Reconstruction of a dwelling according to Spegal’skii 20

elaborate plaits with unusual braiding that do indeed differ strongly from eleventh-century Norwegian and Swedish examples and do have analogies in twelfth-century jewellery. Still, this observation, based on a stylistic analysis, cannot compete with the dating system based on stratification. The structural role of these artefacts is also still an open question. It was assumed that these fragments were part of a wall or served to adorn the entrance to a building as in the Hemse church, the decorative treatment is thoroughly original and might give rise to a large number of hypotheses regarding the reconstruction of early Russian pre-Mongol wooden architecture. But questions relating to a supposed structural role, the stylistic peculiarities of the ornament and the figurative motifs bring researchers up against the difficult issue of the attribution of the artefact. Here the version put forward by Iurii Spegal’skii seems fairly convincing. According to his reconstruction, the fragments were part of a dwelling, presumably a porch, gallery or entrance hall. Spegal’skii noted the smooth band at the upper end of the fragments that probably separated the pillars from their termination, and also a vertical strip framing the edge of the ornament that apparently testifies to the presence of a vertical groove that served to attach the walls of galleries consisting of boards or split logs. An analysis of structural features, which include a groove to take a tongue and the heads of small nails driven into the upper part of the surviving belt, made it possible to produce a more realistic interpretation that made the ‘columns’ part of a set of gates or the inner porch of a residential building. 21 Despite the greater likelihood of this suggestion, the attribution of the ‘columns’ to a dwelling is also open to question on account of their uniqueness. In view of the fragmentary nature of these finds, any reconstruction will remain no more than a hypothesis until additional proof for a particular theory is found. Whatever the opinions about the interpretation and attribution of the ‘columns’, most important and, unarguably, absolutely precious, is the fact that these fragments are the earliest known works of monumental woodcarving from pre-Mongol Rus’.

RZ_wooden-architecture.indd 47

29.09.15 21:22

48

It remains highly questionable whether these pieces belonged directly to the wooden church constructed on the territory of the kremlin that burnt down in the eleventh century, especially as the ‘columns’ were found a considerable distance away from the historical location of that church and had been ‘recycled’ as flooring. Attribution of the artefact is substantially complicated by the poor state of preservation of the finds, which hampers a precise recreation of the structural role that these fragments originally performed. All of this puts a primary emphasis on the ornament, the correct interpretation of which would make it possible to understand the development of old Russian art at its earliest stages. Up until now, however, any hypothesis regarding the place of the carved fragments in question remains extremely uncertain. Nevertheless, an important result has now been achieved in the ‘demythologisation’ of these artefacts, which had been viewed by scholars from different countries from a particular angle and with firm a priori conceptions.

The question of the decagon in Detinets No less complex a challenge for scholarly interpretation is the archaeological material connected with pre-Mongol wooden construction in Novgorod that was uncovered by excavations carried out in 2006–11 in connection with the restoration of the Faceted (or Archbishop’s) Chamber. With their scope and significance, these studies seriously expanded the body of archaeological sources in the early period in the history of the Detinets and the eparch’s court and may possibly serve as proof of the existence of a wooden church on the territory of the Novgorodian kremlin. Fig. 6: Plan of the northwestern part of the Novgorod kremlin (Detinets) showing the excavations of 2008–10. After M. Rodionova

The area in which these excavations were carried out is located in the northwest part of the present-day kremlin. The site is of exceptional importance from an archaeological point of view. The majority of researchers regard it as the most ancient core of the Detinets, where the first fortifications appeared as well as the city’s religious centre – the Bishop’s (later Archbishop’s) Court, located alongside the chief place of Christian wor-

c

ship, St. Sophia Cathedral. The kremlin excavations in 2008 uncovered part Archbishop's Palace

of a srub (framework of logs) with two obtuse angles. Only the lowest tier of logs had survived and the eastern facet of the srub was fully exposed. The northern and southern walls extended beyond the limits of the excavation. The logs were 24–30 centimetres thick and joined at the corners ‘v oblo’ (with

St. Sophia Cathedral

the ends protruding). Small boulders, fragmented pieces of flagstones and broken bricks were placed beneath the northeast corner of the srub; beneath the southeast corner there was an underpinning of short logs bearing indications of reuse. The splen-

0 10 20 30 40 50 M

RZ_wooden-architecture.indd 48

did state of preservation of the wood made it possible to take three samples that upon examination

29.09.15 21:22

49

in the laboratory headed by Olga Tarabardina gave a date of 1287. The shape .

of the srub, the first of its kind uncovered in Novgorod in many years of excavations, gave some grounds for theorising that the structure may have been the remnants of a wooden church. There are insufficient data for a complete reconstruction. It is only possible to seek broad analogies and employ the retrospective method. Extant Russian wooden churches date mainly from the 1600s–1700s, with only rare examples going back to the 1400s–1500s. The hypothetical reconstruction focuses, first and foremost, on the ground plan of the excavated structure and is based on the most general principles of the development of Russian wooden religious architecture. Here the question of its roots immediately arose. Undoubtedly the main impulse for wooden church construction was the adoption Fig. 7: The 2008-I excavation in the Archbishop’s Court of the Novgorod kremlin (Detinets). Srub with obtuse angles, viewed from the north. Photograph by M. Rodionova

of Christianity and the canonical shapes of places of worship adopted from Byzantium. This is the viewpoint maintained by supporters of Byzantine sources, but there is another hypothesis that postulates Slavic origins for wooden religious architecture. As Viacheslav Orfinskii justly observed, ‘by virtue of its dual Byzantine-Slavic sources Russian culture was characterised by a capacity for dialogue’. 22 What is undisputed is that timber church construction in Rus’ developed along its own lines, acquiring features of striking individuality, in which the main principles adopted at one time from Byzantium were undoubtedly preserved. Proceeding from these premises, two possible versions for the reconstruction of the excavated building are under consideration.

Version 1 . A decagonal church (a ‘round decagon’) The length of the eastern side of the uncovered srub was 4.5 metres, while the entire log was 5.6 metres long. A graphic recreation of the plan on the basis of the length of this facet and the angles of the two uncovered corners produces a decagon. The surviving corners are 138º and 150º, giving an average of 144º, exactly the internal angle of a regular decagon. The ratio of length of the side of the decagon (a) to the length of the side of the triangle (b) is 1:1.618, effectively 1:φ – the irrational number that is the basis of the golden section. Proceeding from this, the side of the triangle (which is the radius of the circumscribed circle) is equal to 7.28 metres and the diameter of the decagon 14.56 metres. The dimension of the sides of the constructed decagon fits well with the traditional measures of length employed in Early Rus’, for example the common sazhen, which is known from the eleventh century and was equal to 1.51 metres. The name sazhen comes from the verb stiazhat, ‘to reach’, and meant the span of the outstretched arms. The 4.5-metre side of the excavated srub is equal to three common sazhens, while the radius of the circumscribed circle, at 7.28 metres, is a little less than five common sazhens. Early builders used ‘sazhen cords’ and folding wooden skladeni for measurement. When the plan of the srub was laid out in practice they used a radius (b) of five common sazhens and divided the circle up with ten cords (a) of three sazhens. The discrepancy was ‘spread around’, leading to small differences in the size of the angles between the sides (as in this case) and in the length of the sides.

RZ_wooden-architecture.indd 49

29.09.15 21:22

50

a = 4,5 m 72°

Version 2. A church with a five-walled (‘round’) sanctuary

144°

In this version the marking out of the sanctuary

b=

could have been done in the same manner as de-

7,28

138°

scribed above. In that event the body of the church

m

could have measured no less than ten by ten com36°

mon sazhens. The most vulnerable part of this ver150°

sion is the size of the main volume of the building, which would make for a very large church. A wall over 15 metres across could be achieved only by joining logs lengthwise, which is a fairly dubious, although possible, method of making joints in ancient timber constructions. The splicing of logs does occur in archaeologically explored buildings from medieval Novgorod, however. Nonetheless

0

1

2

3

4

the dimensions of such a church would be simply

5 M

too big. And thus, the first version – a decagonal or ‘round’ church – remains the more acceptable. One

Fig. 8: Version 1. The geometric structure of the plan of a decagonal church (‘round decagon’), made by the architect-restorer V. A. Popov

more detail in favour of that shape is the fact that the srub uncovered by the excavation shows no traces at all of any sanctuary extension (prirub) cut into the main part of the building. This interpretation of the structure refers us to the theme of centrally planned rotundas and ‘round’ churches in early Russian architecture. The centrally planned church type based upon the rotunda and its derivatives is known to have been widespread in the architecture of the early Byzantine period. The functional canonicity of such places of worship was determined by a specific prototype that had the status of a prime sanctuary for the whole of Christendom – the rotunda Church of the Holy Sepulchre in Jerusalem. In

Fig. 9: Version 2. Reconstruction of the plan of a church with a five-walled (‘round’) sanctuary, made by the architect-restorer V. A. Popov

the process of Christianisation, the centrally planned rotunda type of church spread in western Europe. Early Rus’ was not excluded from this process either, although there are only isolated instances of rotundas in its architecture. Today we know of only seven rotunda churches dating from the pre-Mongol period: one each in Kiev and Smolensk, with the rest being in Galich (Western Ukraine). ‘After the Mongol invasion, in the second half of the thirteenth and fourteenth centuries,

10 c

instances of the construction of rotundas become 3c

even rarer. In north-eastern Rus’ (Vladimir, Rostov, Moscow and Tver) not a single rotunda from

5c

3c

10 c

this period has been discovered.’ 23 In Galicia, there are two known cases of wooden, centrally planned churches being built – the Oleshkov and Voskresensk rotundas. The Oleshkov rotunda dates from the twelfth century. The one in Voskresensk falls outside the limits of the pre-Mongol period, that is to say, beyond the middle of the thirteenth century.

RZ_wooden-architecture.indd 50

29.09.15 21:22

51

The wood in these rotundas has not survived. The shape of the churches has been reconstructed from the traces of shallow stone foundations, capable of

The Kremlin excavation site

supporting a light superstructure, and post-holes. If we examine the architecture of early Rus’ and medieval northern Europe, it is possible to see that there were no purely service structures of a rotun-

Archbishop's Palace

da type. As a rule the tower-like ‘trading churches’ of Scandinavia and northern Germany were multitiered and, besides being places of worship, had defensive and commercial functions. 24 It is absolutely

0

5

10  m

M 1:200

obvious from all that has been said that the centrally planned rotunda church was a rare phenomenon for early Russian architecture.

Fig. 10: Reconstruction of the plan of the wooden church (both versions)

However, if we turn to the wooden religious architecture about early Rus’, we find that the Ustiug Chronicle contains interesting information on that city’s Dormition Cathedral. The first mention of it dates from 1290 and concerns the construction of a wooden church. Over the course of 200 years the building burnt down repeatedly and was constructed anew. When, in 1491, after another fire, craftsmen again started to build in timber, but with a cross-shaped plan, the people of Ustiug demanded ‘a round one as of yore’. 25 The main value of this document lies in the fact that it gives a characterisation of the type of the Ustiug edifice that in the late fifteenth century was already regarded as fairly ancient. As is well known, the main building material in northern Rus’ was wood. With the adoption of Christianity, Russian carpenters mastered a new field of construction – timber churches. Using their accustomed material, accumulated practical skills and artistic devices, they rethought religious buildings in their own way and created them in wood. The reasons are entirely understandable: first, the difficulty of creating in wood the canonical forms of Christian masonry churches; second, wooden churches were built by local carpenters, who had great experience in their craft, previously developed construction techniques and artistic conceptions. These factors lie behind the multiplicity of forms of wooden churches. Researchers into early Russian religious architecture have still not arrived at a single common opinion about the courses of its evolution, nor indeed about its sources. The origins of the ‘round’ type of church may well have been a consequence of the limited useful length of a log that restricted the size of a klet’ – the rectangular framework of logs that was the usual basic unit of timber architecture. Its sides might be 10 metres at best, as it was difficult to work with longer pieces of timber. Early Russian carpenters were therefore unable to simply scale up a building to increase the space inside, but they found a very simple solution through increasing the number of walls. If you increased the number of walls by, say, a factor of two (eight instead of four), the floor space inside grew by a factor of over two and a half!

RZ_wooden-architecture.indd 51

29.09.15 21:22

52

It is customarily assumed that churches described as ‘round with twenty walls’ consisted of a central octagon with four protruding, three-sided priruby, bringing the total number of walls to twenty. Geometrically, however, this sort of plan is similar to a Greek cross and better fits those churches described as kreshchataia – ‘crossed’. An interpretation of the church partially uncovered in the Detinets as ‘round with ten walls’, combined with a study of the rotunda buildings that existed in early Rus’, makes it possible to take a somewhat different view of the ground plans of ancient wooden churches. The excavated framework of logs joined at obtuse angles may be the remnants of a centrally planned ‘round’ church without a prirub sanctuary. We do not know of any service buildings from either archaeological or ethnographic sources relating to wooden architecture that might fit this pattern. There are only eight-sided windmills, but they would appear much later, during the time of Peter the Great. It is, moreover, hard to imagine a mill, or indeed any other service building of that size, inside the Detinets. It is impossible to equate this structure with a wooden bell tower either, because ‘bells came into more or less widespread use with us in the fifteenth century’. 26 The existence of a guard or watch tower inside the Detinets and in the immediate proximity to St. Sophia Cathedral does not seem practical either. An analysis of the typology of timber churches in the Russian North shows that ‘round’ buildings were fairly common. In Russian wooden architecture of the sixteenth and seventeenth centuries, ‘round’ or pillar-like churches were popular in specific regions and at a certain time. In all probability this distribution and persistence of pillar-like or ‘round’ churches in the late Middle Ages was a consequence precisely of the ancient and traditional character of the type, which undoubtedly had variations regarding the number of faces surrounding the main volume. We note that churches based upon an octagonal srub with four priruby and usually described as ‘round with twenty walls’, date in the main from the eighteenth century and only occasionally from the second half of the seventeenth. Octagons with two priruby, meanwhile, were common from around 1600. The Church of St. Nicholas in the village of Liavlia, for example, dates from 1584. It follows that the number of priruby added to a ‘round’ srub increased over time and not the reverse. By way of analogies to that which was found at the Detinets we can look at ‘round’ churches from early Rus’. Among those still extant is one in the village of Vazhiny, Podporozh’e district, Leningrad region – the Church of the Resurrection with a decagon at its base. It dates back to 1630 (cladding reconstructed in 1830) and has two priruby. ‘Round’ churches could have even more walls, such as the twelve-faceted central srub of the church in the Cossack village of Kletskaia on the River Don (built in 1730, not extant). The building in the Novgorod kremlin may have had prototypes in the masonry rotundas that existed in the medieval architecture of northern Europe and Scandinavia, with which the city had close trading links at that time. 27 The building excavated in the Detinets dates from the 1280s, when Archbishop Kliment was in office. Under him church construction resumed in Novgorod,

RZ_wooden-architecture.indd 52

29.09.15 21:22

53

after a long hiatus caused by the Mongol invasion. In 1292 the Church of St. Nicholas on Lipno Island was built to the Archbishop’s commission. 28 Researchers into early Russian architecture note this church’s peculiar character, which results from a combination of native traditions and the Romano-Gothic style. This is a possibility, considering that a ten-sided design would have come very close to the rotunda shape. Decagons are very rare in masonry architecture too, although we can identify two such edifices dating from the early Middle Ages – the Mausoleum of Theodoric in Ravenna and St. Gereon’s Basilica in Cologne. It is entirely possible that a wooden church was also constructed on the territory of the Archbishop’s Court that had a masonry rotunda as a prototype. 29 At a time when innovative tendencies were increasingly manifesting themselves in Novgorodian architecture, we cannot exclude a variety of borrowings. Less than a century and a half later, another archbishop had constructed in his court what would be the only secular building in Rus’ in the Gothic style. Summing up, we can state that the character of the structure uncovered during the kremlin excavations does not exclude the possibility that it was a church with a centrally planned or ‘round’ shape, and in that case, being ascribed to the last quarter of the thirteenth century, it has the earliest presently known dating for any surviving Russian wooden church. And in conclusion a few words regarding the attribution of the excavated structure. Relating to the early stage in the replanning of the Archbishop’s Court, we only have at our disposal chronicle reports about a Church of the Presentation in the Temple of 1191, which the Archbishop had built in his court, and a Church of the Nativity dating from 1290. 30 There are, however, insufficient grounds to identify the latter with the excavated structure. It is not possible to determine the precise location of the church from the brief evidence in the chronicle, which only states that it was situated ‘in the Archbishop’s Court, by the rise’. In 1362 a masonry Church of the Nativity was constructed. 31 It is possible to state with caution that the excavated wooden church stood for around ten years before being dismantled for some reason (the logs were definitely taken apart; there are no noticeable signs of a fire) and removed to another site (not a difficult process with wooden churches). Possibly this was connected to the replanning of the Archbishop’s Court and radical reconstructions of the Detinets that took place in the early 1300s. On the other hand, it could be a completely different church that for some reason was never completed and therefore gets no mention in the chronicles. We could obtain answers to some of these questions by examining the structure in its entirety. And thus we have examined the two archaeological finds that can hypothetically be regarded as the only presently known evidence for the existence of wooden churches in medieval Novgorod. The first takes the form of isolated architectural details; the second is part of a log-built structure. The former has an impressive bibliography, but still no unambiguous interpretation; the interpretation of the latter is as yet unchallenged. The only indisputable facts that link the two finds are precise stratigraphic dating – in the first case to the first half of the eleventh century, in the second (backed up by dendrochronology)

RZ_wooden-architecture.indd 53

29.09.15 21:22

54

to the 1280s. The remaining evidence varies depending on which system of indicators a researcher chooses to apply to them. In the context of the suggested hypotheses, both objects are regarded as parts or elements of structures equated with church buildings. In both instances, the possibility of their belonging to one or another building tradition (Slavic or Scandinavian) has been entertained. If the two hypotheses are ever confirmed by archaeology, the history of wooden religious architecture in medieval Novgorod will begin with these two artefacts. In the meantime, we can only hope that, despite the depressing forecasts regarding the preservation of wooden churches, further archaeological exploration in Novgorod will produce new finds and new evidence.

References 1 Zasurtsev P. I., G. P. Chistiakov, G. V. Borisevich, V. P. Tiurin, ‘Arkhitekturnaia rekonstruktsiia novgorodskoi usad’by serediny XIII v.’ [An Architectural Reconstruction of a Novgorodian Homestead of the Mid-13th Century], in: Novgorodskii istoricheskii sbornik, 10, 1961, 207–216. 2 Kolchin, B. A., P. A. Rappoport, G. V. Borisevich, Zhilishche v drevnei Rusi [The Dwelling in Early Rus’], Moscow: Nauka, 1995. 3 Sorokin, A. N., Blagoustroistvo drevnogo Novgoroda [The Amenities of Early Novgorod], Moscow: Obshchestvo istorikov arkhitektury, 1995. 4 Khoroshev, A. S., ‘Novgorodskie usad’by X–XV vv.’ [Novgorodian Homesteads of the 10th– 15th Centuries], Trudy VI Mezhdunarodnogo kongressa slavianskoi arkheologii [Transactions of the 6th Congress of Slavic Archaeology], Vol. 2: Slavianskii srednevekovyi gorod [The Medieval Slavic Town], Moscow: Institut Arkheologii RAN, 1997, 136–148. 5 Faradzheva, N. N., Postroiki Liudina kontsa srednevekovogo Novgoroda [The Structures of the Liudin Konets District of Medieval Novgorod], Moscow, 2010. 6 ‘Novgorodskaia tret’ia letopis’’ [The Third Novgorodian Chronicle], in: Polnoe sobranie russkikh letopisei [The Complete Collection of Russian Chronicles], Vol. 3, St. Petersburg: Eduard Prats, 1841, 208; ‘Voskresenskaia letopis’’ [The Resurrection Chronicle], in: Polnoe sobranie russkikh letopisei [The Complete Collection of Russian Chronicles] Vol. 7, St. Petersburg: Eduard Prats, 1856, 155. 7 Artsikhovskii, A. V., ‘Kolonna iz novgorodskikh raskopok’ [A Column from the Novgorodian Excavations], in: Drevnosti Vostochnoi Evropy [Antiquities of Eastern Europe] 169, Moscow: Nauka, 1969, 7–43. 8 Bocharov, G. N., ‘Rez’ba po kosti v Novgorode (X–XV vv.)’ [Ivory Carving in Novgorod (10th–15th centuries)], in: Drevnii Novgorod. Istoriia. Iskusstvo. Arkheologiia. Novye Issledovaniia [Early Novgorod. History. Art. Archaeology. New Researches] Moscow, 1983, 175f. 9 Zasurtsev, P. I., Novgorod, otkrytyi arkheologami [The Novgorod Discovered by Archaeologists], Moscow: Nauka, 1967, 98f. 10 Artsikhovskii, A. V., 1969. 11 Vagner, G. K., Mastera drevnerusskoi skul’ptury. Rel’efy Iur’eva-Pol’skogo [Masters of Early Russian Sculpture. The Reliefs of Iur’ev-Pol’skii], Moscow: Iskusstvo, 1966, 1–66. 12 Kudriavtsev, I. N., ‘Dereviannye polukolonny iz Novgoroda i ikh severoevropeiskie analogy’ [The Wooden Half-Columns from Novgorod and their North European Analogies], in: Slavianorusskie drevnosti. Drevniaia Rus’: novye issledovaniia [Slavic-Russian Antiquities:

RZ_wooden-architecture.indd 54

29.09.15 21:22

55

Early Rus’ New Researches], 1995, 2, 211–218. 13 Hohler, E. B., Norwegian Stave Church Sculpture, Vol. I, II, Oslo: Scandinavian University Press, 1999. 14 Nicolaysen N., Norske bygninger fra fortiden. Norwegian buildings from former times. Vol. 3. Kristiania: Carl C. Werner & CO’s Bogtrykkeri, 1877-1880. Pl. XVIII. 15 Konetskii, V., K. G. Samoilov, ‘O nekotorykh aspektakh kul’tury srednevekovogo Novgoroda’ [On some Aspects of the Culture of Medieval Novgorod], Novgorodskii istoricheskii sbornik [Novgorodian Historical Anthology], Issue 7 (17), St. Petersburg, 1999. 16 Ibid. 17 Hohler, E. B., ‘Two Wooden Posts Found in Novgorod: A Note on their Date and Stylistic Connection’, Collegium Medievale, 16, 2003, 37–50. 18 Sedova, M. V., Iuvelirnye izdeliia Drevnego Novgoroda (X–XV vv.) [Works of Jewellery from Early Novgorod (10th–15th Centuries)], Moscow: Nauka, 1981, 116–118. 19 Arkhipova, E. I., ‘Romanskaia arkhitekturnaia rez’ba Borisoglebskogo sobora v Chernigova’ (The Romanesque Architectural Carving in the Ss. Boris and Gleb Cathedral in Chernigov), in: Material’na ta dukhovna kul’tura pivdennoi Rusi [The Material and Spiritual Culture of Southern Rus’], Kiev-Chernigov, 2012, 25–35. 20 Spegal’skii, I., Zhilishche Severo-Zapadnoi Rusi IX–XIII vv. [The Dwelling in the NorthWest 9th–13th centuries], Leningrad: Nauka, 1972, 172. 21. Ibid. 22 Orfinskii, V. P., I. E. Grishina, Tipologiia dereviannogo kul’tovogo zodchestva russkogo severa [A Typology of the Wooden Religious Architecture of the Russian North] Petrozavodsk, 2004, 17. 23 Ioannisian, O. M., ‘Khramy-rotondy v Drevnei Rusi’ [Rotunda Churches in Early Rus’], in: A. L. Batalov, A. M. Lidov (eds), Ierusalim v russkoi kul’ture [Jerusalem in Russian Culture], Moscow: Nauka, 1994, 100–142. 24 Ioannisian, O. M., ‘Arkhitektura Drevnei Rusi i srednevekovoi Skandinavii. Ikh vzaimosviazi’ [The Architecture of Early Rus’ and Scandinavia. Their Interconnections], in: Izuchenie i restavratssiia pamiatnikov drevnerusskoi arkhitektury i monumental’nogo iskusstva [The Study and Restoration of Works of Early Russian Architecture and Monumental Art], Transactions of the State Hermitage, St. Petersburg, XXXIV, 2007, 99–135. 25 Ustiuzhskii letopisnyi svod [The Ustiug Chronicle Digest], Moscow-Leningrad: USSR Academy of Sciences, 1950, 98. 26 Krasovskii, M. V., Entsiklopediia russkoi arkhitektury. Dereviannoe zodchestvo [An Encyclopaedia of Russian Architecture. Wooden Construction], St. Petersburg: Satis’, 2010, 143. 27 Khodakovsky, E., Wooden Church Architecture of the Russian North: Regional Schools and Traditions, London-New York: Routledge, 2016, 10. 28 Novgorodskaia pervaia letopis’ starshego i mladshego izvodov [The Novgorodian Primary Chronicle Older and Younger Recensions], Moscow-Leningrad: USSR Academy of Sciences, 1950, 327. 29 Khodakovsky, E., 2016, 89. 30 Novgorodskaia pervaia letopis’ starshego i mladshego izvodov [The Novgorodian Primary Chronicle Older and Younger Recensions], Moscow-Leningrad: USSR Academy of Sciences, 1950, 326. 31 Ibid., 368.

RZ_wooden-architecture.indd 55

29.09.15 21:22

56

RZ_wooden-architecture.indd 56

29.09.15 21:22

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

RZ_wooden-architecture.indd 57

29.09.15 21:22

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

RZ_wooden-architecture.indd 58

29.09.15 21:22

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 (viðr) 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

RZ_wooden-architecture.indd 59

29.09.15 21:22

60

that timber could be stored and processed, and pre-cut materials assembled conveniently. Where the terrain necessitated extensive foundation walls, some deviation from a true east alignment may have been accepted. The Benedictine Walahfrid Strabo, writing about 840, says that churches could be placed in the direction which suited the site best. 14 In Norway, compass orientation of about 200 medieval wooden and stone churches is known. As shown elsewhere, the tendency was for an oriented axis where this was possible topographically, and preferably on an altitude contour. 15 Following the tradition, the churchyard would be given an oriented axis, as would the altar room (þat er upphaf laga varra at ver scolom luta austr). 16

Planning and design Over the course of history, architectural ambitions in the Church have been both grandiose and ascetic. The Augustinian Alexander Neckam (1157–1217) said that costly buildings showed an empty pride. 17 Petrus Cantor (died 1197) commented that to build in stone was absurd when the Old Testament tells us that people lived in tents. 18 The monastic orders like Dominicans and Franciscans constructed their churches with a demonstrative simplicity, without columns, vaults or steeples. 19 St. Bernard of Clairvaux, talking as a monk to monks, derided the improper heights, lengths and widths of new churches. 20 Abbot Suger therefore put forward pious and symbolic arguments in support of his plans for Saint-Denis. 21 This may have been the background when around 1200 bishop Jon had to remind builders that churches were to be erected for God and men, not out of a thirst for power or sovereignty. 22 Next to nothing is known about the relationship between patrons and builders. Contracts may have been concluded orally. In Torpo Church, erected after 1163, some names of craftsmen may be preserved, however. Above the door to the alFig. 2: Kaupanger Stave Church

tar room (altarisgolf), runes are carved on an arcade. The master builder is called Thorolv, his workmen are called Asgrim, Haakon, Erling, Paal, Eindride and Sjone, while Olav was the name of the carver. Likewise, a preserved deal with painted runes from the demolished Ål church, not far from Torpo, gives the name of the master builder Thorolv. His workmen were Geirstein, Gunnar, Vidar, Eivind, Eirik, Gunnar and Alf. All these craftsmen may have been local or regional. 23 The Icelandic Bishops’ sagas (Biskupa sögur, 1200– 1250) recall how bishop Jón Ögmundsson (1052–1121) of Skálholt symbolically marked out the extent of a new church (markaði sjállfr grundvöll undir kirkjuna). 24 Around 1273, archbishop Jon of Nidaros in the spirit of Gratian explains how bishops are to mark out the foundations on the surface, erect a cross on the place of the altar, and carry the first stone to the site (skall byscup marka tuft. oc ræisa kros

RZ_wooden-architecture.indd 60

29.09.15 21:22

61

i. eða (oc) fyrsta stein til kirkiu (bera)). 25 It is doubtful whether a busy bishop would carry out this ritual for most churches, and it is not known if or how it was actually performed 300 years earlier in Norway proper.

Assembling There is no such thing as the perfect plan, only the best compromise between competing priorities. Many texts distinguish only between wood and stone, buildings small and large, between plain and heavily decorated surfaces. The type of fabric and size or costliness of buildings was up to the builder to decide. Any demands for ecclesiastical conformity may have been met with arguments for local diversity. Craftsmen, brought up with the use of tools and working procedures, may have been used to oral mnemonics. Their pragmatic techniques were based on set procedures, which need not even have been theoretically correct, as they were judged solely by the results produced. 26 Avoiding going into details about design, some European texts just state that church buildings were done ‘the right way’ (Ut mos est). 27 How this ‘right way’ may have been, the Laxdæla Saga informs us: Thorkel Eyjolfsson (was) a leader of prominence in Iceland. In the summer he had his ship launched. They made land in the north of Norway, and Thorkel made his way to Nidaros to king Olaf. That same winter King Olaf had a church built of wood in the town. It was large and very impressive, and care was taken with its every aspect. (…) Timber which the king had given Thorkel was loaded aboard ship. The timber was both of fine quality and in great quantity, for Thorkel spared no pains with its selection. One morning when the king had risen early and was accompanied by only a few men, he saw a man up on the church which was then under construction. He was very surprised at this, for the morning had not yet advanced to the time when his carpenters were accustomed to rise. The king recognized the man: it was Thorkel Eyjolfsson, who was measuring all the largest beams, the cross-ties, joists and supports. The king went over to him immediately and said, ‘What are you up to, Thorkel? Do you plan to cut timber for your church in Iceland on this model?’ Thorkel answered, ‘Right you are, my lord’. Then King Olaf spoke: ‘Chop two ells off the length of each beam and your church will still be the greatest in Iceland’. Thorkel answered, ‘Keep your timber then, if you fear you have given of it too generously, or regret making the offer, but I’ll not chop as much as an ell’s length off it. I lack neither the energy nor the means to obtain my timber elsewhere’. The king answered that this timber would not be a church. And so it happened, the next year Thorkel drowned in Iceland and the timber was washed ashore on islands all around. 28 Though it may be fiction, the story shows that people from Iceland, ‘Austmen’, came to Norway around 1026 to collect timber for their houses and churches back home. Also, it may suggest that this royal church was presumed to have special ratios between parts of the building, or elements of special size. There is no description, however, of how the footings and foundation were designed.

RZ_wooden-architecture.indd 61

29.09.15 21:22

62

Fig. 3: Høre Stave Church, nave raft beams, archaeological excavation 1979

Húngrvaka describes how bishop Klængr Þorsteinsson ordered a church built around 1151: In two ships came out great trees, which Klæng the bishop let hew in Norway for that church which he had built at Skálholt, which was in all things choice far above any other house that was built in Iceland, both in wood and workmanship (…) These were the head artificers at the church at Skálholt; Arni, who was called the architect, and Bjorn Thorvald’s son, the skilful; Illugi Leif’s son shaped and cut wood. 29 (þessir váru höfuðsmiðir at kirkjunni í Skálaholti: Árni er kallaðr var höfutsmiðr, ok Björn hinn hagi þorvaldsson; Illugi Leifsson teldi ok viði). 30

Churchyard, yard The early Gulathing Law states how a church and churchyard may have been established and how the community would construct and maintain the site and building. 31 For churches with burial rights, the wall around the yard was to be erected by the owners within twelve months of the erection of the church. Until that was done, the community could not ask for consecration, which was performed in return for food and board for the bishop and his men. 32 The churchyard could only gradually be viewed as blessed, following the sacramental blessing of each burial. In some early Norwegian laws though, the notions of blessing, dedication and consecration seem to be muddled up by the author.

Dedication The reasons for dedication were first for cleansing, exorcising and blessing the building spot. Second, to set the altar and building apart from secular use, to reserve it for liturgical acts and the practice of faith for eternity. The dedication

RZ_wooden-architecture.indd 62

29.09.15 21:22

63

of the altar was done to God by a saint whose relic was symbolically buried in the mensa. 33 The dedication ritual was written in an ordo, a functional document describing the actions and choreography of the bishop, the priests and deacon. No such dedication ordo has survived in Norway. A surviving Irish example, however, may have points of resemblance. This can be found in the Leabhar Breac collection of Gaelic transcriptions from the fourteenth century. It may have been adapted into Irish tradition after about 900 CE. 34 The ordo starts with the bishop and the priests singing outside the western entrance, before crossing the threshold. They all gather by the altar where the bishop consecrates water and salt. Then he writes the alphabet in Greek and Latin diagonally on the floor with his staff to sanctify the foundation and the floor of the building. He then marks crosses with his knife in the four corners of the altar and three crosses along the east–west axis of the table. He washes and wipes the mensa, kindles incense with a vessel and anoints the seven crosses marked with consecrated oil. After anointing the baptismal font, bishop and priests walk around the church, one way or the other, to the middle post in the east wall of the altar room. Here the bishop marks a cross in the post with his knife. Afterwards, back inside the church, the bishop goes from the altar to the four corner posts, marks a cross on each with his knife and sprinkles them with holy water. 35 The ordo is not a design description or a building programme, but more like a functional equivalent. The reason for dedication was to confirm publicly the proprietorship and the specific use of the site and building. Description of the actual church in the text is vague. As I have shown elsewhere, the nave and altar room are described as one, and in addition there may be an entrance hall in the west part; corner posts and a middle post in the east wall of the altar room indicate a wooden building. 36 Until the prohibition at the synod of Dublin in 1186, the altar table and base could be of wood in that area, 37 but as late as 1320 Archbishop Eilif assumes tables were made both of wood and stone in Norway. 38

Symbolic explanations With the development of the rite of dedication came works explaining its purpose, like homilies for the feast day. In these texts the vocabulary pointed to ideas beyond the form, a sort of ‘theological architecturology’. 39 One example is the quasi-realistic description of a wooden church in the twelfth-century Icelandic/Norwegian ‘Stave Church Homily’. 40 This homily on the day of dedication of a wooden church (kirkjudagsmál), explains faith through multi-level allegorical parallels, open to several divergent interpretations. 41 The altar room signifies both angels in heaven and an expression of prayers; the four corner posts are interpreted both as the four Gospels, and as the four cardinal virtues. The building described cannot be placed topographically or typologically, nor can it be reconstructed; it never existed except in the homilist’s imagination. 42 On the other hand, this description in the vernacular helped to confirm the terms for several building elements in the wooden churches.

RZ_wooden-architecture.indd 63

29.09.15 21:22

64

Such a sermon shows that, while there is a connection between a liturgy and the building itself, theology has no physical equivalent in materials and structures; exegeses were operating with intellectual constructs far removed from the craftsman’s reality. And liturgical practices may also be changed with little impact upon the fabric of a building. 43

Visitation, maintenance and laws Visitation is the bishop’s general supervision of clergy, people and building. In 906 Bishop Regino of Prüm wrote a widely used questionnaire to be filled out by the inspector. 44 On his visit to the main congregational churches the bishop was to ask about the daily life of the faithful, the income and the condition of the building proper, as stipulated in Norwegian law. 45 People may have wanted a long-lasting church, but the roof was vulnerable and the earth-fast wooden posts could become weak while the upper part was still durable. It had to be determined how much subsidence of the ground and leaking of the roof could be tolerated. Replacement of a dilapidated building was a question of economy and expediency. Eldre Borgartings kristenrett (the Borgarþing law, 1274) states that the dedication of a church lasts as long as the bed stone which carries the corner stave is in situ (Nu hældr vighingh a kirkiuæ meðan horn steinar standa). 46 Nyere Gulatings kristenrett (the Gulaþing law), states that if the church collapses and the corner staves are broken (En ef thær kirkiur falla niðr oc brotna hornstafr), fresh timber has to be delivered to the building yard within twelve months. 47 As I have discussed elsewhere, it is unclear if the paragraphs indicate that the condition of decay is beyond repair, actually falling down, or has already become a ruin. 48 Dereliction may have been a longterm process, like when incorrect detailing has led to moisture ingress and decay; the texts are kept constructionally neutral. The laws cited seem to describe Fig. 4: Ringebu Stave Church 1980, north-east corner stave

a time when the nave and altar room are finally seen as unfit for liturgical use, and archbishop Jon says that the dedication sits in the walls (þui at i væggium vighizt kirkia). 49 To be more precise canonically, St. Anselm, Archbishop of Canterbury (1093–1109) wrote that the church exists because of the altar; a church can no longer be considered a church when its altar is destroyed. Prior to the Council of Trent (1545–63), destroyed churches could not be converted to profane or secular use. 50 A capitulary from the reign of Charlemagne says that wooden material from a former church was only to be used in the construction of the new building (IV. Ligna ecclesiae non debent ad aliud opus iungi nisi ad ecclesiam aliam). 51 Orders like this confirm the existence of the conditions. The erection and deconstruction of the churches was a practical and mundane affair. When a church closes, there has never been an officially approved rite. Closing may have been done by singing a final

RZ_wooden-architecture.indd 64

29.09.15 21:22

65

mass and removing the relics from the reliquary. The re-consecration that would follow the enlarging or rebuilding of a church gave scope for a change of dedication. 52 To sum up, the dedication of a church was independent of the material of the construction and dedicated material could not be put to a secular use.

Conclusions

The question pertained to was what written sources can tell us about medieval wooden churches in Norway proper. Technical rhetoric is the skill of subduing language so that it most accurately conveys reality. For a building archaeologist, the sources are frustratingly vague and indeterminate: They convey few answers to our constructive questions. Therefore, the relevant information has to be obtained from dry site evidence, from the fabric of the parish church itself and from examination of the twenty-eight preserved medieval wooden churches in Norway.

References 1 I Corinthians 16:19. 2 White, L. M., Building God’s House in the Roman World, London: Johns Hopkins University Press, 1990, 60ff. 3 Andersen, P. S., B. E. Crawford (eds.), Northern Isles Connections: Essays from Orkney and Shetland Presented to Per Sveaas Andersen, Kirkwall: Orkney Press, 1995, 333. 4 Adam of Bremen, History of the Archbishops of Hamburg-Bremen (Records of Western Civilization Series), New York: Columbia University Press, 2002; Chouquer, G., F. Favory, Les arpenteurs romain. Théorie et pratique, Paris: Éditions Errance, 1992. 5 Welsh, T. J., The Use of the Portable Altar: A Historical Synopsis and a Commentary, Washington: Catholic University of America Press, 1950. 6 Snorre Sturlason, Heimskringla: or, The Lives of the Norse Kings, New York: Dover, 1990. 7 Wood, S., The Proprietary Church in the Medieval West, Oxford: Oxford University Press, 2006, 918. 8 Friedberg, Ae. (ed.), Corpus Iuris Canonici, 1993, v. I–II (Leipzig 1879–81), Graz; I: 1296; 1303. 9 Keyser, R., P. A. Munch, G. Storm, E. Hertzberg, Norges gamle Love indtil 1387 (1846–95), I–V, Christiania; I (1846): 8. 10 Snorre Sturlason, Heimskringla or The Lives of the Norse Kings, New York: Dover Publications, 1990. 11 Sturlason, 1990; Chapter 47. 12 Sturlason, 1990; Chapter 53. 13 Dietrichson, L., De norske stavkirker, Christiania: Alb. Cammermeyer, 1892: 96–97. Dietrichson, L., Die Holzbaukunst Norwegens in Vergangenheit und Gegenwart, Dresden: Verlag Gerhard Kühtmann, 1893. 14 Strabo, W., Walahfrid Strabo’s Libellus de Exordiis et Incrementis Quarundam in Observationibus Ecclesiasticis Rerum, A. L. Harting-Correa (transl.), Leiden: Brill, 1996, 1–37. 15 Vogel, C., ‘Versus ad Orientem. L’orientation dans les Ordines Romani du haut moyen âge’, in: Studi Medievali, Vol. 1, 1960, 447–469; Eide, O. E., ‘Om kirkers orientering’, Arkeologiske skrifter fra Historisk Museum, no. 3, Bergen: Historisk Museum, 1986, 119–130: 129–130; Jensenius, J. H., Trekirkene før stavkirkene. En undersøkelse av planlegging og design av

RZ_wooden-architecture.indd 65

29.09.15 21:22

66

kirker for ca. år 1100, ConText 6, Oslo: Arkitekthøgskolen i Oslo, 2001, 84. 16 Keyser et al. (eds), 1846, I, 3. 17 Frisch, T. G., Gothic Art 1140–c1450: Sources and Documents, New Jersey: Prentice Hall, 1987, 31. 18 Migne, J. P., Patrologia cursus completus ... series latina, prima. Vol. 1–221, Paris (1844–65), 205:255. 19 Onians, J., ‘The Last Judgement of Renaissance Architecture’, Journal of the Royal Society of Arts, Vol. 128, 1980, 701–720. 20 Bernard, St. Bernard’s Apologia to Abbot William, Kalamazoo: Cistercian Publications, 1970, 63; Rudolph, C., Artistic Change at St.-Denis, Princeton: Princeton University Press, 1990. 21 Panofsky, E., Gothic Architecture and Scholasticism, Pennsylvania: Latrobe, 1951, 90–91. 22 Keyser et al. (eds), 1848, II, 346. 23 Schirmer, H. M. (1899), ‘Vore middelalderlige arkitekter og bygningsfolk’, Foreningen til Norske Fortidsmindesmerkers Bevaring, Aarsberetning, Kristiania, 1900, 100; Blindheim, M., Graffiti in Norwegian Stave Churches c.1150–c.1350. Oslo: Universitetsforlaget, 1985, 45. 24 Jóns saga helga, 17; Biskopu Sögur, Copenhagen: S. L. Möllers, 1858, I,171; Stories of the Bishops of Iceland, London: J. Masters and Co, 1895. 25 Keyser et al. (eds) 1846-95. 26 Shelby, L. R., ‘The Geometrical Knowledge of Medieval Master Masons’, Speculum, 1972, 47, 395–421, 401. 27 Binding, G., Der Früh- und Hochmittelalterliche Bauherr als Sapiens Architectus, Darmstadt: Wissenschaftliche Buchgesellschaft, 1996, 288. 28 The Sagas of Icelanders, London: Penguin, 2000, 415. 29 Stories of the Bishops of Iceland. 30 Biskopu Sögur, 60–61, 81–82. 31 Keyser, R. et al. (eds.), 1846-95, I, 8–9. 32 Gejvall, N.-G., Westerhus. Medieval Population and Church in the Light of Skeletal Remains, Kungl. Vitterhets Historie och Antikvitets Akademien: Lund, 1960, 120–127. 33 Repsher, B., The Rite of Church Dedication In The Early Medieval Era, Lampeter: The Edwin Mellen Press, 1998, 17ff. 34 Ó Cuindlis, Murchadh, Leabhar Breac, Facsimile Royal Irish Academy, I–II, Dublin, 1876; Olden, T., On an Early Irish Tract in the Leabhar Breac Describing the Mode of Consecrating a Church, Cambridge: Camden Society Transaction, 1900, 4: 98–104,177–180. 35 Stokes, W., The Lebar Brecc Tractate on the Consecration of a Church. Miscellanea linguistica in onore de Graziadio Ascoli, Turin: Ascoli, 1901, 363–387. 36 Jensenius, J. H., Innvielsesrituale for trekirke, Århus: Hikuin, 1997, 24: 83–98. 37 Braun, J., Der christliche Altar in seine geschichtliche Entwicklung, München I, 87. 38 Keyser et al. (eds), 1849, III, 258. 39 Markschies, C., Gibt es eine ‘Theologie der gotischen Katedrale’?, Heidelberg: Winter, 1995, 60ff. 40 Indrebø, G., Gamal Norsk Homiliebok, COD, AM 619 4°, Oslo: Dybwad, 1931. 41 Hjelde, O., Norsk preken i det 12.århundre. Studier i Gammel Norsk Homiliebok, Oslo: O. Hjelde, 1970, 290ff. 42 Bekker-Nielsen, H., ‘The Old Norse Dedication Homily’, Festschrift für Konstantin

RZ_wooden-architecture.indd 66

29.09.15 21:22

67

Reichardt, Bern: Francke, 1969, 127–134. 43 Markschies, C. (1995) Gibt es eine ‘Theologie der gotischen Katedrale’?, 40; Jungmann, J. A., The Mass of the Roman Rite, I–II, Dublin: Four Courts Press, 1950/1986. 44 Hellinger, W., ‘Die Pfarrvisitationen nach Regino von Prüm’, Zeitschrift der SavignyStiftung für Rechtsgeschichte, 1962–1963, I: 1–116; II: 76–137. 45 Keyser et al. (eds), 1846, I, 7. 46 Keyser et al. (eds), 1846, I, 345. 47 Keyser et al. (eds), 1848, II, 328. 48 Jensenius, J. H., Staver som brytes og kirker som faller. Vedlikehold og rivning av kirker i middelalderen, Kirke og Kultur, 2007, 5, 418–429, 423. 49 Keyser et al. (eds), 1848, II, 346. 50 Gulczynski, J. T., The Desecration and Violation of Churches, Washington DC: the Catholic University of America, 1942, 16, 27. 51 Vykoukal, E., ‘Ein unbekanntes liturgisches Kapitulare, aus der Karolingerzeit’, in: Mélanges d’histoire offerts à Charles Moeller, I, 1914, 337–349, 339. 52 Arnold-Forster, F., Studies in Church Dedications or England’s Patron Saints, London: Skeffington & Son, 1899, I, 12. All illustrations are by the author.

RZ_wooden-architecture.indd 67

29.09.15 21:22

68

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 English medieval historical records survive in greater quantities than almost anywhere else in Europe. In spite of this wealth of material, historians have made only limited use of documents about buildings and methods of construction. The potential of written records was initially indicated by Turner in a pioneering work on English architecture, developed further in Louis Salzman’s study of Building in England down to AD 1540 and then in a joint examination of royal building works published a few years later. 1 Though these are important studies, the possibilities for considering timber architecture in particular remain incompletely exploited. If there is a reason for this, it is that extracting the information useful to architectural historians is a painstaking and often unrewarding task. Medieval written records rarely provide detailed descriptions of buildings. Too often the references to construction work are brief, or if they are more extended, as in the description of individual – usually masonry – buildings, they are too idealised to be useful. The present paper seeks to outline first the written evidence for early timber architecture in England and the light it casts on buildings in the period before 1200. In the second part, the sources of evidence for timber church buildings, in particular, are considered for the period before the Reformation.

Written sources for timber architecture before 1200 Saints’ lives, charters and chronicles provide a useful record of some of the building work which took place on churches and monastic buildings in the period before 1200. The possibilities of combining this scattered information into a coherent account of patterns of ecclesiastical construction have been explored by Richard Gem in a largely unpublished study. However, the key results of that work now available show that there were peaks in building work in the 980s and 1050s, separated by a recession in the early eleventh century, probably the consequence in part of the large sums of money being withdrawn from the English economy to pay off Viking raiders. The data for the period before the mid-tenth century are insufficient to identify similar patterns of work. 2 It is unfortunate for our purposes that the majority of large building projects recorded by chroniclers were masonry not timber structures. Stone buildings were deemed particularly worthy of note and timber buildings were often recorded only when they were replaced by masonry structures, or to show the modest nature of earlier churches. For example, the saintly Bishop Dunstan is said to have built a timber church at the remote vill of Mayfield (Sussex). 3 One of the earliest detailed records of a timber building is a pre-Conquest agreement noted in Domesday Book (1086). 4 Peterborough Abbey undertook to provide the materials to construct a house (domus) 60 feet (18.3m) long in Huntingdonshire and to maintain it in return for fishing rights in the Fens. The building was surrounded by an enclosure (curia), which suggests that this was

RZ_wooden-architecture.indd 68

29.09.15 21:22

69

LENGTH INCLUDING END AISLES

BREADTH INCLUDING AISLES WHERE PRESENT

60 feet 40 feet

16 feet, plus aisles 15 feet

HEIGHT TO RIDGE

County Durham Hall Peasant house Huntingdonshire Hall

60 feet

Kensworth Hall

35 feet

30 feet

22 feet

Building between the hall and chamber

22? feet

17 feet

17 feet

Chamber

22 feet

16 feet

18 feet

Sources: Co. Durham: Austin, D., Boldon Book, 1982, 14–15, 36–37. Huntingdonshire: Domesday Book, i, 205a. Kensworth (1152): Hale, W. H., Domesday of St. Paul’s, 1858, 128­–29.

LENGTH

BREADTH INCLUDING AISLES WHERE PRESENT

HEIGHT TO RIDGE

Fyfield Hall (1167–85d)

38 feet

35 feet

c. 25 feet

Temple Balsall (1176–1221d)

33 feet

34 feet

28 feet

Bishop’s Palace, Hereford (1179d)

110 feet

55 feet

Westwick Cottage, Hertfordshire (1184–1219d)

23+ feet

23 feet

c. 19 feet

Newbury Farmhouse, Kent (1187–1227d)

31 feet

34 feet

c. 25 feet

Source: Walker, J., ‘Late-Twelfth and Early-Thirteenth-Century Aisled Buildings: A Comparison’, Vernacular Architecture, 1999, 30: 21–53.

Table 1: Documented Dimensions of Houses in the Eleventh and Twelfth Century Table 2: Dimensions of Houses of TwelfthCentury Standing Halls

no mere peasant building. The dimensions of the house were comparable to those later constructed by the tenants of the Bishop of Durham recorded in a survey of 1183. It noted that the unfree tenants in Aucklandshire had to build a hall and chapel for the bishop in the Forest of Weardale. The hall was to be 60 feet (18.3 m) in length and 16 feet (4.9 m) wide ‘within the posts’, a phrase which suggests that it was aisled. Two service rooms, chamber and a privy are also mentioned, as was a timber chapel which was to measure 40 feet (12.2 m) long by 15 feet (4.6 m) wide. 5 The records of leases of estates granted by St. Paul’s Church in London provide quite exceptional detail of the sizes of a series of late twelfth-century buildings, including some indication of their structural detail (Table 1). The leases were particularly concerned with the size of barns because of the implications for the volume of crops which were supplied to the tenant at the start of the lease. However, the dimensions of domestic buildings were also provided for the demesne farmstead at Kensworth (Hertfordshire) in a lease of 1152 (Fig. 1). The hall is said to have measured 35 feet (10.7 m) long and 30 feet (9.1 m) wide,

RZ_wooden-architecture.indd 69

29.09.15 21:22

70

which means almost certainly that it was aisled.

Kensworth

Close to it was a chamber (bedroom) measuring 22 feet (6.7 m) long and 16 feet (4.9 m) wide, and be-

Chamber

tween those two buildings was a third structure of uncertain function, measuring ‘12 feet’ long (presumably an error for 22 feet (6.7 m) with the initial x omitted) by 17 feet (5.2 m) wide. We can place these figures in context by comparing them to surviving twelfth-century houses Building between

(Table 2). The standing buildings are all aisled. Of the documented buildings, the Bishop of Durham’s hall had aisles and we can assume that the hall at

Hall

0 0

10 m 25 ft

Kensworth, with a width of 9.1 m, was of similar form. We have no breadth for the hall built by Peterborough Abbey, so the character of that build-

Fig. 1: Diagrammatic reconstruction of the buildings at Kensworth, recorded in a lease of 1152 of St. Paul’s, London

ing is uncertain. The lengths of the documented buildings are notably longer than the standing structures, but that is because only the surviving elements are included in the dimensions of the latter. We can assume the buildings were originally longer. The two sets of figures are broadly comparable, but the dimensions of the building at Kensworth require some comment. This was an unusually low building for its breadth. If we compare it with the very modest hall at Westwick Cottage, we can begin to appreciate this. Westwick was not a manorial hall and was relatively low, measuring only 11 feet (3.4 m) below the arcade plate and with a width for the central body of the hall of 4.3 m. We are told that the height of the hall at Kensworth below the beam (and arcade plate) was similar, with another 11 feet (3.4 m) above the beam to the ridge. If we assume that the roof had a pitch of 45°, which was typical of this period, then the central body of the hall was much broader than Westwick, measuring 6.8 m wide with two aisles of only 1.2 m breadth. The twelfth-century St. Paul’s leases provide a great deal of information about the barns, and these have been examined in detail elsewhere. 6 The majority of barns described in these leases had two longitudinal aisles and two return (or end) aisles. The evidence for the barns in the St. Paul’s leases can be summarised diagrammatically to indicate the range of sizes and the presence of side and return aisles (Fig. 2). The profile of the buildings is generally uncertain. It has been assumed, unless there is evidence to the contrary, that the roof line was continuous from the nave to the aisles, something which was certainly the case in Wickham barn 1, for which the measurements of the height of the walls in the side aisles are provided. The St Paul’s leases note a number of other agricultural buildings which were to be found in the manorial farmsteads or curiae. There was, for example, a cattle shed, sheep house and lambing shed at Kensworth, while at Thorpe there was a brewhouse, malthouse, dairy, cattle shed and three small hen houses. Sandon (1155) had also a wash house and pigsty, while at Ardleigh (1141) there was a granary, kitchen, hayhouse and stable. 7 An extent of the royal manors of

RZ_wooden-architecture.indd 70

29.09.15 21:22

71

Leighton Buzzard, Aylesbury and Brill was compiled in c. 1155 and provides a similar record of agricultural buildings. It was drawn up in the period after the civil war in England between Stephen and Matilda to provide an assessment of the buildings, harvested crops and livestock present on the manors, and a determination of the repairs and deficiencies which needed to be made good before they could be leased. It appears to have been compiled using a demesne lease of the St. Paul’s type, since it lists not only the livestock, but also the size of the barns and the crops which should be present within them. 8 A number of the documents provide a valuation of buildings, although it is not clear what is being valued. It may be that the figures relate to the value of materials in the standing building, or it could be that they are the costs of replacing the existing structure with another of similar size. If it is the latter, the cost of replacing a building itself will depend upon whether the timber had to be purchased or could be obtained from the estate, and whether labour services might be obtained without charge from customary tenants. The figures often seem remarkably modest. Barns on the manor of Ingham of Bury St. Edmund’s Abbey were among the smaller recorded, but none of them was worth more than a mark (13s. 4d.). Even the hall and the adjoining buildings together were worth only 20s. These may be compared with the value of eight cows and one heifer in the same document, which were worth 34s., which is presumably their Wickham barn 1 Wickham barn 2

Wickham barn 3

Navestock barn

Fig. 2: Diagrammatic reconstruction of barns at Wickham and Navestock based on the leases of St. Paul’s, London

RZ_wooden-architecture.indd 71

0 0

10 m 25 ft

29.09.15 21:22

72

purchase price.9 The Ingham buildings bear comparison with others on the St. Paul’s estate, such as those at Navestock which have even lower valuations. Estimates for repairing barns at Brill help to set these figures in perspective. The two barns there were said to require 30s. expenditure and a cattle shed another 5s. 10 This can be compared with the standing, and presumably not ruinous, cattle shed at Navestock, which was valued itself at only 5s. 11 Early deeds contain useful, if somewhat enigmatic descriptions of urban buildings. A grant of 1184×1198 by Oseney Abbey of a tenement in Oxford explains that it was a building with two solars (upper rooms) which were constructed with oak timbers and roofed with tile. The rooms were set above undercrofts (cellars), of which one was vaulted in stone and the other with a plank floor. 12 That building seems to have been rather similar to another in Oxford also leased by Oseney, which is described as an oak building with a stone tile roof with a solar and undercroft. The undercroft was evidently quite large because it contained a mill driven by a horse. 13 A particularly early reference to this arrangement of solars and undercrofts is a description of 1098×1108 of a house (mansionaria) in London with an underground building (domus subterranea), evidently an undercroft. 14 A rather different sort of building is indicated in the Winchester survey of c. 1110, Winton Domesday. It refers to the king’s balchus where thieves were imprisoned. This was evidently a building constructed of balks or beams strong enough to prevent escape and can be compared to the Balkerne mentioned in Colchester in the mid-twelfth century. The use of the word ‘balk’ suggests a structure with horizontally laid timbers. 15 Most of these buildings were constructed of timber, but by the end of the twelfth century the situation was changing. Regulations drawn up in or before 1212 by Henry Fitzailwin, Mayor of London, noted that in former times the great majority of buildings had been constructed of timber and were roofed with straw thatch, but after an extensive fire during the reign of Stephen (1135–54) more buildings had been constructed of stone with tiled roofs. 16

Documentary evidence for timber churches A number of the documentary references to timber churches have been collected by Zimmermann and Ahrens, but most provide little more information than noting that a particular church is made of timber. 17 As we have already noted, the main sources for such references are saints’ lives, charters and chronicles. The reliability of saints’ lives is always questionable, and although they may contain faithfully recorded traditions associated with their subject, the details of buildings mentioned are less secure. Timber churches were increasingly uncommon by the twelfth century and by then were treated as symbols for humility and modesty. By the early thirteenth century, the Franciscans could adopt wooden buildings as part of their espousal of poverty. 18 Furthermore, the memory that timber churches had once been more common may have made them appropriate details to give veracity to accounts of the past. For these reasons, we must treat with caution records of timber churches in later records of saints’ lives and monastic foundations.

RZ_wooden-architecture.indd 72

29.09.15 21:22

73

The modesty of timber churches was surpassed in medieval accounts only by reference to buildings made of wattle. The first resting place of the body of St. Cuthbert at Durham at the end of the tenth century was said to be a church made of wattles (de virgis). Eadmer’s account of the Miracles of St Oswald (written 1113×1116) recounts how the remains of the saint prevented the spread of fire through a house made of woven branches (virgas intexterat) standing in Worcester. The Lanercost Cartulary records that a chapel was built of rods (de virgis) at Trierman (Walton, Cumberland) during the episcopate of Aethelwulf (1133–57). 19 A further wattle chapel in the early twelfth century is mentioned in Wales, but this was perhaps a temporary structure erected during a military campaign. 20 However, we should be a little cautious about entirely dismissing such buildings as mere literary tropes. There was a tradition of wattle construction in the north and west of Britain, which makes the record of such buildings at least plausible. 21 Once we move into the thirteenth century and beyond, the most important source for our knowledge of churches are episcopal visitations. These were supposed to take place every three years to inspect, among other matters, the state of the church and its furnishings, and check that it was kept in good repair. The rate of survival of such records is extremely poor and as Forrest has recently noted, the visitations which do survive were kept as evidence in disputes over jurisdiction and may not be representative of the whole. 22 How far that should affect our use of such records as a guide to the materials of churches is uncertain. Certainly, some visitations showed a greater interest in that matter than others. Zimmermann drew attention to an early surviving record of visitations conducted by the Bishop of Salisbury in the 1220s. 23 The bishop showed an unusual interest in the details of church buildings, noting wooden chapels at Ruscombe, Arborfield and Earley (St Bartholomew) in the parish of Sonning (Berkshire), at Knook in the parish of Heytesbury (Wiltshire) and at Hurtmore in Godalming parish (Surrey). The visitation to the chapel of St. Bartholomew at Earley notes the presence of piles of stones which had been brought together to replace the timber building. All of these timber buildings were chapels; all of the churches were built of stone. 24 Regrettably, most other visitations do not show the preoccupations of the Bishop of Salisbury and record buildings only if they required repair. For example, a visitation of the diocese of Hereford held in 1397 noted that the church at Vowchurch was unroofed and that the parishioners had made an agreement with a carpenter for its repair and had the money for his payment. This is particularly notable because the surviving church building, though encased in stone, has a roof supported on timber posts, suggesting it was originally an entirely wooden structure. The same visitation made no mention of the timber bell tower at Pembridge, presumably because it was in an adequate state. 25 However, a visitation of 1297 of churches belonging to St Paul’s Cathedral did find it appropriate to note that the tower at Aldebury was of timber. 26 A visitation in 1293–95 in the diocese of Canterbury recorded that the church at Fairfield (Kent) had not been dedicated because it was built of wood and daub.

RZ_wooden-architecture.indd 73

29.09.15 21:22

74

The altars of the church were made of wood and lower parts of the walls of the chancel had holes. 27 The wooden structure of this church still survives, though it was reconstructed in the fifteenth century, and a visitation of 1511–12 dismissed it briefly with the comment omnia bene (all is well). The sixteenth-century archbishop clearly did not deem the use of timber to be worthy of note. 28 The duty of maintaining the nave of the church fell to the parishioners; by the end of the thirteenth century the maintenance of the chancel was recognised as the responsibility of the rector. In order to manage the works on the nave, the parishioners appointed churchwardens who were answerable both to their fellow churchgoers and to the bishop making his visitation. 29 The churchwardens were required to record their income and expenditure while they served in office, generally a period of two years. The numbers of surviving accounts produced by the churchwardens is not large, with fewer than twenty for the fourteenth century and a hundred or so for the fifteenth century. It is not until the sixteenth century that they begin to survive in greater numbers. These accounts are of particular interest for the study of the building history of churches since they record all expenditures on the maintenance, repairs and reconstruction of the nave, down to the purchase of nails and timber. 30 It is unfortunate that none of the surviving accounts apparently concerns timber churches, because they would provide a remarkable insight into the costs of maintaining such a building. Whether chapels had similar accounts to those kept by the churchwardens, or their accounts were included in those of the parish, is uncertain. The final sources we need to consider for medieval timber church buildings are antiquarian records and drawings. Timber buildings caught the attention of those interested in antiquities because they were notably different from the normal run of churches. They were able to record these surviving medieval structures before they were finally swept away. William Stukeley, for example, noted and illustrated the twelfth-century stone front of the chapel at Toft (Lincolnshire), then used as a blacksmith’s shop, but recorded that the body of the church was built of wood. 31 William White’s gazetteer of the same county mentions that there was a chapel of wood at Whaplode Drove which had been replaced in 1820 by one of brick. 32 These and similar records suggest that timber churches were once more common than the surviving number of buildings might suggest.

Conclusions

Given the wealth of documentary sources which survive for so many aspects of life in medieval England, it is rather disappointing that so few records have survived for timber churches. While further detailed searches would undoubtedly produce a greater harvest, even Salzman – who spent a considerable part of his life perusing documents in the Public Record Office (now The National Archives) – was not able to find many references to timber churches. 33 Timber ecclesiastical buildings were largely chapels and their records have survived even less well than those of parish churches. It is ironic that the church which had a central role in educating clerks actually generated so few records at the level of the parish, and moreover preserved so few of these. Although written

RZ_wooden-architecture.indd 74

29.09.15 21:22

75

records are important for understanding the context in which timber churches were constructed and reconstructed, they provide somewhat meagre returns for those studying the buildings themselves.

References 1 Turner, T. H., Some Account of Domestic Architecture in England. from the Conquest to the End of the Thirteenth Century, London: John Henry Parker, 1851; Salzman, L. F., Building in England down to 1540: A Documentary History, Oxford: Clarendon Press, 1952; Allen Brown, R., H. M. Colvin, A. J. Taylor, The History of the King’s Works, I: The Middle Ages, London: HMSO, 1963. 2 Gem, R., ‘A Recession in English Architecture during the Early Eleventh Century, and its Effect on the Development of the Romanesque Style’, Journal of the British Archaeological Association, 1975, 38: 28–49; Gem, R., ‘ABC: How Should We Periodize Anglo-Saxon Architecture’, in: L. A. S. Butler, R. K. Morris (eds), The Anglo-Saxon Church: Papers on History, Architecture and Archaeology in Honour of Dr H. M. Taylor, London: Council for British Archaeology, 1986, Fig. 96. 3 Turner, A. J., B. J. Muir, Eadmer of Canterbury: Lives and Miracles of Saints Oda, Dunstan and Oswald, Oxford: Clarendon Press, 2006, 122–25. 4 Domesday Book i, 1086, 205a. 5 Austin, D., Boldon Book, Chichester: Phillimore, 1982, 36–37. 6 Hale, W. H., Domesday of St Paul’s (Camden Soc. old series 69), London: Camden Society, 1858, 122–139; Horn, W., E. Born, (1979), ‘The Domesday of St. Paul’s Cathedral, London: A Portrait of Manorial Architecture in Twelfth Century England’, in: M. H. King, W. M. Stevens (eds), Saints, Scholars and Heroes: Studies in Medieval Culture in Honour of Charles W. Jones, Collegeville, MN: Hill Monastic Library, St. John’s Abbey and University, 343–417. 7 Hale, 1858, 129, 131, 134, 136. 8 The National Archives, SC11/74, printed in G. H. Fowler, ‘Extents of the Royal Manors of Aylesbury and Brill, circa 1155’, Records of Buckinghamshire, 1926, 11: 401–405 and Richmond, R., ‘Three Records of the Alien Priory of Grove and the Manor of Leighton Buzzard’, Bedfordshire Historical and Record Society, 1924, 8: 22–24. 9 Davis, R. H. C., Kalendar of Abbot Samson (Camden Soc. third series, 84), London: Camden Society, 1954, no. 77. Some eleventh-century prices are given by Harvey, S., ‘Domesday England’, in: H. E. Hallam (ed.), The Agrarian History of England and Wales II: 1042–1350, Cambridge: Cambridge University Press, 1988, 57. 10 Fowler, 1926, 403. 11 London Metropolitan Archives, CLC/313/L/H/002/MS25122/495. 12 Salter, H. E., Cartulary of Oseney Abbey, I (Oxford Historical Society 89), Oxford: Oxford Historical Society, 1929, no. 54. 13 Ibid., no. 112. 14 Kissan, B. W., ‘An Early List of London Properties’, Transactions of the London and Middlesex Archaeological Society, new series, 1940, 8, 58. 15 Barlow, F., M. Biddle, O. von Feilitzen, D. J. Keene, Winchester in the Early Middle Ages, Oxford: Clarendon Press, 1976, 37 (no. 19), 236.

RZ_wooden-architecture.indd 75

29.09.15 21:22

76

16 Riley, H. T., Munimenta Gildhallae Londoniensis, I (Rolls Series 12), London: HMSO, 1849, 328–329. 17 Zimmermann, W., ‘Ecclesia lignea und ligneis tabulis fabricata’, Bonner Jahrbücher, 1958, 158: 414–453; Ahrens, C., Die frühen Holzkirchen Europas, Stuttgart: Konrad Theiss Verlag, 2001, 163–191. 18 Martin, A. R., Franciscan Architecture in England, Manchester: Manchester University Press, 1937, 10, 13–14, 31. 19 Turner, Muir, 2006, 316–317; J. M. Todd, Lanercost Cartulary, Gateshead: Surtees Society, no. 36, 1997; Fowler, J. T., Rites of Durham, Durham: Surtees Society, 1903, 71; Arnold, T., Symeonis Monachi Opera Omnia 1 (Rolls Series), London: HMSO, 1882, 79. 20 Green, J. A., The Government of England under Henry I, Cambridge: Cambridge University Press, 1986, 29, citing Lambeth Palace Library, MS 51, f. 25. 21 For example, McNeil, R., ‘Two 12th-Century Wich Houses in Nantwich, Cheshire’, Medieval Archaeology, 1983, 27: 45–50. 22 Forrest, I., ‘The Survival of Medieval Visitation Records’, Archives (The Journal of the British Records Association), 2012, 125: 1–10; Forrest, I., ‘The Transformation of Visitation in Thirteenth-Century England’, Past and Present, 2013, 221: 3–38. 23 Zimmermann, 1958, 439. 24 Rich Jones, W. H. (ed.), Vetus registrum Sarisberiense, alias dictum Registrum S. Osmundi Episcopi (Rolls Series 78), HMSO: London, 1883, 275, 279, 282, 295, 297, 307, 309. 25 Bannister, A. T., ‘Returns of the Diocese of Hereford in 1397, I’, English Historical Review, 1929, 44: 283; Bannister, A. T., ‘Returns of the Diocese of Hereford in 1397, I’, English Historical Review, 1930, 45: 446; Boucher, A., R. Morriss, ‘The Bell Tower of St. Mary’s Church, Pembridge, Herefordshire’, Vernacular Architecture, 2011, 43: 22–35. 26 Simpson, W. Sparrow, Visitations of Churches Belonging to St. Paul’s Cathedral in 1297 and 1458, Camden Society, 1895, 45. 27 Woodruff, C. E., ‘Some Early Visitation Rolls Preserved at Canterbury’, Archaeologia Cantiana, 1917, 32: 162. 28 Wood-Legh, K. L., Kentish Visitations of Archbishop William Warham and his Deputies, 1511–12, Maidstone: Kent Archaeological Society, 1984, 159. 29 Mason, E., ‘The Role of the English Parishioner, 1100–1500’, Journal of Ecclesicastical History, 1976, 27: 24–27. 30 Drew, C., Early Parochial Organisation in England: The Origins of the Office of Churchwarden (St. Anthony’s Hall Publications, 7), London: St Anthony’s Press, 1954, 5–12; Burgess, C., ‘Pre-Reformation Churchwardens’ Accounts and Parish Government: Lessons from London and Bristol’, English Historical Review, 2002, 117: 307–309. For an example of such accounts, see Mellows, W. T., Peterborough Local Administration: Parochial Government Before the Reformation (Northamptonshire Record Society 9), Northampton: Northamptonshire Record 31 Owen, D. M., Church and Society in Medieval Lincolnshire, Lincoln: History of Lincolnshire 32 White, W., White’s History, Gazetteer and Directory of Lincolnshire, Sheffield: William White, 1872, 786. 33 Salzman, 1952, 192–193. All illustrations are by the author.

RZ_wooden-architecture.indd 76

29.09.15 21:22

RZ_wooden-architecture.indd 77

29.09.15 21:22

78

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 In Russia today, an increasing number of restoration initiatives demand constant reference to original archival documentation as the most trustworthy foundation for historical research. This situation has produced the need to establish a general typology of the written historical documents reflecting the history of Russian wooden architecture and to give a systematic overview of the Russian archives that contain various kinds of historical evidence. The substantial application of wide-ranging archival studies opens up new prospects for the theoretical interpretation and historical evaluation of Russian wooden architecture as a significant part of the nation’s art and culture, expanding and refining the previous methodology. The way in which archival documents are systematised has been determined by the administrative structure of Russia in various periods – Late Medieval (fifteenth to seventeenth centuries), Synodal (1700–1917) and Soviet (1917–91), within which it is possible to identify types of sources characteristic of that particular historical era.

Late Medieval sources At the time of the centralised Russian state, from the late fifteenth century to the early eighteenth, the basic territorial unit in the system of secular state administration was the uezd (a town or city and the lands associated with it), while the unit of ecclesiastical administration was the eparchy. The greater part of the territories now referred to as the Russian North belonged to the Novgorodian lands (present-day Novgorod, Pskov, Leningrad, Arkhangelsk and Murmansk regions, the Republics of Karelia and Komi) and belonged to the Eparchy (from 1589 Metropolitanate) of Novgorod, Over the course of the sixteenth and seventeenth centuries, the Vologda, Tot’ma, Ustiug, Dvina, Kargopol’, Vaga and other uezdy (in present-day Arkhangelsk and Vologda regions) that lay to the east of Novgorodian territory became part of the newly formed Vologda (from 1556), Velikii Ustiug (1682–1787) and Kholmogory (from 1682) eparchies. The centre of ecclesiastical administration in each eparchy was the Arkhiereiskii dom (literally Eparch’s House). The eparchy, with the bishop or archbishop (eparch) at its head, governed the local parishes (prikhody) – ecclesiastical districts of the ordinary populace with their own place of worship, monasteries and metochia (dependencies and representations of monasteries), controlling, among other things, the construction of churches. In the sixteenth and seventeenth centuries the state gradually restricted ecclesiastical privileges. The Church lost its autonomy as the prikaz system came into being. Prikazy were bodies of central state administration with responsibility for one branch of state affairs or a particular territorial area.

RZ_wooden-architecture.indd 78

29.09.15 21:22

79

One of the most important prikazy in the Muscovite state was the Pomestnyi prikaz. It had charge of all matters concerning land-holding, both pomest’ia (estates held in reward for service) and votchiny (those held by succession), and was responsible for the compilation of cadastres and census records (pistsovye knigi and perepisnye knigi). The records of the Pomestnyi prikaz have come down to us in an enormous quantity of rolls and books, only a small fraction of which has as yet been described and studied. The Pomestnyi prikaz fund is kept in the Russian State Archive of Ancient Documents in Moscow (RGADA). 1 In the early seventeenth century, for a centralised ecclesiastical administration of the clerical estate, several Patriarshie (Patriarch’s) prikazy were created, followed in 1649 by the Monastyrskii (Monastery) prikaz. The latter was created as the highest judicial organ for the clergy and the population of church-owned lands. Effectively it performed not only judicial, but also financial, administrative and police functions relating to ecclesiastical affairs. The RGADA and the Manuscript Department of the State Historical Museum in Moscow (OR GIM) possess an extensive complex of clerical documents from the Monastyrskii 2 and Patriarshie 3 prikazy: financial account books, legislative material and written records. In the places under the control of prikazy, between the late sixteenth and early eighteenth centuries there were officials called voevody in charge of the territories entrusted to them. All the matters connected with the governance of the urban centre and uezd were carried out in an office known as the prikaznaia izba or prikaznaia palata. The voevoda was responsible for administrative, police, judicial, financial and service-estate affairs. The records of local administration in the 1500s and 1600s – documents of the prikaznye izby and also of the eparchial Arkhiereiskie doma – can be found in the Russian section of the Historical Research Archive of the Russian Academy of Sciences St. Petersburg Institute of History (RS NIA SPbII R AN). The funds of all the institutions listed contain documents from which it is possible to glean a variety of information about the churches, pogosty and volosti within the structure of the eparchies. A general picture of the number and location of parishes and churches is provided by financial documentation of various levels. These mass sources are compiled in the form of ‘books’ (knigi): the account and duty books of the Patriarshii prikaz 4 and Arkhiereiskie doma, 5,6,7,8,9 boiarskie knigi (records of the highest class of feudal society, their estates, income and service history, payment ledgers and survey records of uezdy and more). Of particular importance are the cadastres, census records and dozornye knigi (updated registers) from the Pomestnyi prikaz. 10 The cadastres (pistsovye knigi) were compiled from the late fifteenth century, the time the centralised Muscovite state was being formed, to collect information on the socio-demographic and economic situation for purposes of taxation. These books contain among other things information about ecclesiastical buildings in Russian towns, pogosty (rural districts centred on a church), monasteries and convents. The scribes sent off to the Russian North systematically recorded one pogost after another, beginning

RZ_wooden-architecture.indd 79

29.09.15 21:22

80

as a rule with the administrative centre, where there was a church or a whole ensemble, since prosperity could be judged by the number of churches, their size, their architecture, church furnishings and icons with covers. The contents of sixteenth- and seventeenth-century cadastres are valuable indicators of the trends in and character of wooden church construction in the North, and in some cases of its typology as well. Clerical records from the prikazy represent an inexhaustible source of information for the history of individual churches and settlements. All the documents existing in this period can be grouped into the following main types: charters, formal replies, reports, instructions, judgements, memoranda and petitions. Particularly valuable sources are the eparchial documents giving approval for the construction of churches and appointing of priests in the RS NIA SPbII R AN and other collections, which they entered from the archives of the churches themselves. The precise dating characteristic of eparchiFig. 1: RGADA Fund 1201, schedule 1, file 79. 1677 ‘Census Record of Solovetskii Monastery Estates (volosti and saltworks)’, 203

al documents helps in tracing the constructional history of long-lost churches. Khramozdannye gramoty (church-building charters) regulated the construction and consecration of new churches and also the reconstruction of old ones. These documents contain information about the church architecture and plate, descriptions of the events that led to the destruction of previous churches, information about the names and status of settlements, churches and monasteries, the borders of eparchies and new parishes. The eparchs’ blagoslovennye gramoty (blessing charters) might prescribe the parameters of ecclesiastical buildings. A stavlenaia gramota was a document issued by a bishop to a priest to confirm the validity of his ordination. It indicated the church to which the priest was appointed and the year, month and day of its consecration. The group of late medieval historical sources recording timber buildings that actually existed in different regions of the Russian North is supplemented by another remarkable set of documents describing ‘virtual’ structures that were expected to be erected. The information about planned churches is contained in poriadnye zapisi – contracts between clients (peasant communities) and teams (artels) of builders that list in detail the features of the intended building – its spatial composition, height, decoration and interior structure. Not one of the churches built to a contract that has so far come to light has survived but a careful reading of the texts and conversion to our modern system of measurements makes it possible to reconstruct those buildings in accordance with the parameters agreed in the document. So far only a few dozen of these ‘verbal projects’ have been identified. These materials are scattered across various museum collections, archival funds and manuscript sections of libraries: the State

RZ_wooden-architecture.indd 80

29.09.15 21:22

81

Historical Museum (GIM), RGADA and RS NIA SPbII R AN, as well as the Russian State Historical Archive (RGIA) and the National Library of Russia in St. Petersburg (RNB), the Russian State Library in Moscow (RGB) and elsewhere. The clerical records have so far only been analysed and researched to a very small extent. This huge body of documents, if attentively read and studied, could enable us to reconstruct the course of church construction in the Russian North from the moment of its Christianisation through to the eighteenth century.

The Synodal period (1700–1917) Eighteenth to mid-nineteenth centuries The church reform initiated by Peter the Great and global changes in Russian history in the eighteenth century had a deep impact on the administrative and structural organisation of the Northern eparchies and their recordkeeping. In 1708 the country was divided up into eight gubernii (governorates). The northern territories became part of the Ingermanland (later St. Petersburg) and Arkhangelogorodskaia (Arkhangelsk) Gubernii. In 1727 the Novgorod Guberniia was formed from a considerable portion of the lands in St. Petersburg Guberniia. In the 1770s–80s, the Vologda and Olonets namestnichestva (from namestnik – ‘lieutenant’) were separated from the Arkhangelsk and Novgorod Gubernii and by the early 1800s they had also been granted the status of gubernii. Thus, at that time the boundaries of the gubernii were made to correspond with those of the church eparchies. Under Peter I, the Senate and kollegii (collegia, forerunners of ministries) became the highest institution of government and all executive power gradually passed to them from the prikazy. The kollegii were superior to the administrations of gubernii, provinces and uezdy. The Pomestnyi prikaz was replaced by the Votchinnaia Kollegiia 11 that was established in 1721. When the need arose to carry out the changes prescribed by the centre quickly and decisively at a local level, a further reform was implemented in the regions: the prikaznye izby were abolished and their functions shared between zemskie izby, the chancelleries of gubernii and provinces (voevody), municipal magistraty, courts and other newly created institutions. The records of those institutions are kept in the RGADA and RS NIA SPbII R AN. From the late eighteenth century, the administrative body for a guberniia was the gubernskoe pravlenie (a board made up of the governor and several other high officials). One of the bodies operating under the gubernskoe pravlenie was the department of construction (stroitel’noe otdelenie) responsible for the repair of old churches and the building of new ones. With the abolition of the patriarchate, administration of the Church at the highest level passed to the Most Holy Governing Synod. The records of the Synod are mainly to be found in the Russian State Historical Archive (RGIA), with parts in the RGADA and History Museum. Among the Synodal documents particularly noteworthy is the Economic Administration (Khoziaistvennoe upravlenie) fund that contains records of ecclesiastical and monastic property,

RZ_wooden-architecture.indd 81

29.09.15 21:22

82

the construction and repair of church and monastic buildings, lists of parishes and churches and insurance documents for the ecclesiastical property of eparchies. 12 The eparchy-level administrative organ was the consistory (dukhovnaia konsistoriia) (before 1744 such bodies went by various names – dukhovnyi prikaz, dikasteriia, kantselariia, dukhovnoe pravlenie), which was in the charge of the incumbent eparch and acted under his authority as a consultative and executive institution. One of the duties of the consistories was to monitor the construction and improvement of churches and the management of church property. An eparchy was subdivided into blagochiniia, groups of several neighbouring parishes overseen by a senior priest (blagochinnyi), who kept order in matters of faith in the parishes and regularly reported back to the consistory. The bulk of these documents are kept in the State Archive of Arkhangelsk Region (GA AO) in Arkhangelsk, the National Archive of the Republic of Karelia (NARK) in Petrozavodsk, the State Archive of Vologda Region (GAVO) in Vologda and the State Archive of Novgorod Region (GANO) in Velikii Novgorod, but a substantial number of sources in archives in St. Petersburg: the RGIA and the Research Archive of the Russian Academy of Sciences Institute of the History of Material Culture (NA IIMK R AN); and Moscow: the RGADA, also illustrate the process of interaction between church and secular authorities in the centre and periphery of the Russian Empire. Of the previous records the eparchial khramozdannye and stavlenye gramoty continued, while with the shift from taxation by peasant household to taxation of (male) individuals, cadastres and census books (pistovye and perepisnye knigi) were no longer required. They were replaced by state and ecclesiastical records of population accounting. The former took stock of all the taxable population, the latter only of the Orthodox Christian population and clergy. For us the Church documentation is especially important, particularly the klirovye vedomosti, records of the clergy that appeared in the last third of the eighteenth century. They included information about the building of the church in which a priest was serving and about church property. The klirovye vedomosti can be found in the funds of the Synod and Synodal institutions in the RGIA and also in the funds of the consistories, eparchial administrations, uezd religious administrations, monasteries and parish churches in the regional archives (GA AO, GAVO, GANO and NARK). A whole complex of sources has come to light in which matters concerning church buildings are examined. They include requests from the local clergy and parishioners for the construction and consecration of churches, records of the examination of buildings and various reports on the performance of particular assignments, inventories of church property and altar books, and so on. Furthermore, in the nineteenth century blagochinnye began to draw up special records about the state of the churches in their jurisdiction. All these documents are in the regional archives (the funds of the bodies of eparchy and guberniia administration, statistical committees, the monasteries, parish churches and other bodies).

RZ_wooden-architecture.indd 82

29.09.15 21:22

83

Beginning in the 1820s, the authorities’ determination to control church construction according to unified standards under the supervision of a professional architect produced extensive correspondence between all those involved in the process: parishioners themselves as initiators of the church building, parish priests, bishops’ aides, architects, church and civic commissions, all those responsible for approving the budgets, revising and refining the projects. This increasing bureaucracy has supplied modern-day researchers with a huge amount of documents. In all these sources it is possible from time to time to come across detailed descriptions of religious buildings and occasionally even depictions of them. Statistical information about the economic state of gubernii, lists of settlements and churches and a variety of cartographic and land use documentation were deposited in the funds of the guberniia statistical committees (as well as regional archives). The Fig. 2: GAAO, Fund 29, Schedule 4, Vol. 3, File 350, 44. From a file about the reconstruction of a dilapidated church in Purnema parish. 1858

materials of general land surveys from the second half of the eighteenth century and nineteenth century, technical drawings and inspection reports for a locality all help to establish the location of churches and their construction history.

1870s–1910s From the 1880s on, we can observe the first attempts by the imperial authorities not only to register the number of churches, clergy and parishioners, but also to process this information for the sake of preservation of the buildings. From

Fig. 3: GAAO, Fund 75, Schedule 1, File 670. File on the construction of a church in Nimen’ga province

the second half of the nineteenth century, a drive to preserve and restore the nation’s architectural legacy proceeded not only from the state, but also from archaeological societies, archival commissions, museums, congresses of archaeologists, artists and architects, societies of devotees of history, archaeology and ethnography. Particularly outstanding among them were the Academy of Arts, 13 Moscow Archaeological Society (M AO) 14 and Imperial Archaeological Commission (IAC), 15 the latter becoming the central official body responsible for the maintenance and restoration of architectural monuments in Russia. One of the most valuable categories of archival sources was generated by a professional and highly skilled approach to the mission of studying wooden edifices. The earliest initiatives in this field came in the 1870s from the Academy of Arts and its architects, who were involved in the first exploration of Northern architecture – Lev Dahl, Vladimir Suslov and Dmitry Mileev. The professional talents of these architects later ensured the successful activities of the Imperial Archaeological Commission, especially Mileev, who kept working for the commission until 1914. Of invaluable significance are the first professional plans and drawings as well as the travel notes, paperwork, reports and

RZ_wooden-architecture.indd 83

29.09.15 21:22

84

scholarly papers of the first researchers of the wooden architecture of the Russian North. The most important archival collections of Russian architects and restorers from the late nineteenth and early twentieth centuries are preserved in the RGIA, NA IIMK R AN and Research Archive of the Russian Academy of Arts (NA R AKh). A comprehensive study of these collections will give a panoramic overview of the proficient methods employed by Russian architects in the early twentieth century in preserving and restoring the masterpieces of Russian wooden architecture. Fig. 4: Manuscript section of the NA IIMK RAN, R-III, File 4223. Dossier on the Church of SS. Peter and Paul in Povenets, 1887, 16

The IAC, created in 1859, had as a goal the collection of data about ancient monuments in the Russian Empire, including the scholarly examination and evaluation of them. The commission also involved itself in the preservation and restoration of buildings. Henceforth no alterations to venerable churches were permitted without the consent of the IAC. Therefore a large number of files in regional archives and the NA IIMK consist of correspondence on questions of the reconstruction or demolition of old buildings. To gather information on existing churches the IAC devised a special questionnaire on churches, their history and furnishings. These were sent all over the Russian North in 1887 to be filled in by local priests in tabular form. The questionnaires were then sorted and partly published in the Issues of the Archaeological Commission, but the original documents are still preserved in its archive in the NA IIMK. 16 In a number of cases, for example, the dossier on the SS. Peter and Paul Cathedral in Povenets, this material makes it possible to reconstruct the appearance of a lost edifice. Of the commissions that appeared under the auspices of the Moscow Archaeological Society, the most noteworthy is the commission on the preservation of ancient monuments, especially churches, that was formed in 1876. The duties of its members included inspecting old monuments, making drawings of them and seeing to the restoration of those that had fallen into decline. We know now what hundreds of lost churches looked like thanks to the extensive use of photography to record architectural monuments from the 1880s onwards. The largest collection of photographs of wooden architecture is in the photograph section of the NA IIMK and the photograph library of the Shchusev State Museum of Architecture in Moscow (GM A, previously GNIM A). They contain negatives and prints of pictures taken by the majority of prerevolutionary researchers. Large collections of photographs are also to be found in the Russian Ethnographic Museum (REM, St. Petersburg) and the archive of the Russian Academy of Sciences Peter the Great Museum of Anthropology and Ethnography (Kunstkammer) (R AE R A N, St. Petersburg) and elsewhere. A photographic section exists in each regional archive, the archives of many local history museums, museums of wooden architecture and other places.

Soviet period (1917–91) The Bolshevik seizure of power in October 1917 brought radical changes to all spheres of the life of the state. In the first post-revolutionary years, the organs of central and local administration and the territorial structure of the country

RZ_wooden-architecture.indd 84

29.09.15 21:22

85

23,0

19,5

0

5

10 m

11,5 9,0

0,0

Fig. 5: Povenets, the SS. Peter and Paul Cathedral. Reconstruction by Andrei Bode

were being established and in a constant state of flux. All the old organisations were abolished, reformed and renamed. A host of new scholarly and scientific establishments and societies sprang up, many of which were short-lived. In the period of struggle against religion, churches themselves perished and, along with them, a huge stratum of ecclesiastical sources. While the bulk of sources from pre-revolutionary times are ecclesiastical clerical records, from the start of the Soviet era first place was taken by documents from scholarly establishments and materials from researchers looking into folk architecture. In the pre-war period the main direction of researchers’ activities was gathering information, investigating and recording extant architectural monuments. It is possible to identify a number of national-level institutions that worked in close connection with each other. In Petrograd-Leningrad there was the Department for the Conservation of Works of Art and Ancient Monuments under the People’s Commissariat for Education – the first state body for the conservation of monuments in the city and the forerunner of today’s Committee for the State Inspection, Use and Conservation of Historical and Cultural Monuments (KGIOP) – the Academy of the History of Material Culture (the present-day Russian Academy of Sciences Institute of the History of Material Culture) that traces its ancestry to the Imperial Archaeological Commission and the Ethnographic Section of the Russian Museum that in 1934 turned into the separate Ethnographic Museum; the peasant art section of the fine arts section of the State Institute of the History of the Arts (GIII) and the Hermitage. Members of one and the same group of specialists could work concurrently in all these organisations and others, moving from one to the other, and so the materials are kept in various places. In Moscow the conservation and study of folk art was the concern of the museum department of the scholarly section

RZ_wooden-architecture.indd 85

29.09.15 21:22

86

of the People’s Commissariat for Education, 17 the Moscow section of the State Academy for the History of Material Culture 18 and the State History Museum, which in 1925 began to organise expeditions to study history and daily life. The first set of archival documents dates from the first decade of the Soviet era (late 1910s–1920s) and comprises the materials from the expeditions to the North. This activity may seem unexpected, given the hardships of the first post-revolutionary years, but this research resulted in a large quantity of archival material documenting outstanding wooden buildings that would later disappear. A study of Igor’ Grabar’s and Petr Baranovskii’s expeditions (to the Northern Dvina and Pinega rivers in 1920–21, to Kizhi in 1926, to the White Sea, Northern Dvina and Nikolo-Korel’skii monastery in 1931) reveals precious evidence about authentic works of church architecture and peasant houses in the Russian North just before the dramatic events of collectivisation that took place in the late 1920s. This material is in the conservation section of the architectural archive of the Shchusev State Museum of Architecture (OKh A A GM A) 19 and the research archive of the Moscow State Integrated Art and Historical Architectural and Natural Landscape Museum-Reserve (NA MGOMZ). Multidisciplinary art-historical expeditions through the northern regions of Russia under the leadership of Konstantin Romanov gathered unique material on the ethnography and wooden architecture of Karelia, the Pinega and Mezen in 1926–28 that has not survived down to the present. This material is kept in the research archive of the Russian Museum of Ethnography, 20 the manuscript section of the research archive of the State Russian Museum (OR NA GR M), 21 ,22 and the NA IIMK R AN. 23,24,25 Konstantin Romanov’s contribution to the cause of preserving architectural monuments, both before the Revolution and in the 1920s–30s, was invaluable. A panoramic view of the peasant North can be seen in his photographs. 26 In the 1910s and 1920s a tremendous gathering of information about ancient works of architecture, taking stock of all the surviving cultural heritage, registering and recording (by photography and surveys) the objects, was carried out by the department of the People’s Commissariat for Education that was responsible for museum affairs and the conservation of artistic and historical monuments: Aleksandr Udalenkov, Konstantin Romanov, Fedor Kalikin and others. Registration material relating to the northern gubernii – questionnaires, reports, surveys, photographs, information about the stock-taking, preservation and repair of wooden churches and excerpts from the IAC church dossiers – are kept in the NA IIMK R AN. 27 The collection of negatives and prints of photographs taken by members of the department is in the photographic section of the NA IIMK R AN and the photographic library of the Russian Museum of Ethnography. The aftermath of the Great Patriotic War of 1941–45 saw a clear shift in Soviet state policy in favour of restoration, which had been almost totally neglected in the 1930s. The purposeful formation of institutions geared towards scholarly research, restoration and museum usage began.

RZ_wooden-architecture.indd 86

29.09.15 21:22

87

From 1944 the management and coordination of all work to conserve historical and cultural monuments was carried out through the Main Administration for the Conservation of Monuments (GUOP), which later repeatedly changed its subordination and name. Prominent specialists in wooden architecture were recruited to work in GUOP, including Petr Baranovskii, Igor’ Grabar’, Pëtr Maksimov, Sofia Zabello and Lev Lisenko. The material from that body – records of the inspection of buildings, surveys, sketches, drawings and photographs, dossiers and historical records – are kept in the GM A. 28 This is also the location of materials from the expeditions to study the monuments of the Russian North carried out by the Moscow Architectural Institute, the Academy of Arts, GM A and other organisations in the 1930s–60s. 29,

30

The personal funds of Bar-

anovskii, Maksimov, Dmitrii Sukhov and Iurii Ushakov contain preparatory matter for their published works, unique graphic art, photographs, descriptions of buildings, journals and other highly valuable materials. 31 , 32, 33, 34 A considerable photographic collection from northern expeditions throughout the whole Soviet era is kept in the GM A and IIMK, making it possible to determine the appearance of objects destroyed in recent times. Fig. 6: Manuscript section NA IIMK RAN. Fund 67, File 41, Bundle 3 (F. A. Kalikin, registration material for ancient monuments of Olonets guberniia, Petrozavodsk uezd (questionnaires, reports, surveys and photographs), 44

For several decades from the moment of their creation, the Moscow-based Central Scientific Restoration Design Workshops (TsNRPM) remained the foremost scientific restoration organisation in the country. The first work aimed at the conservation of examples of wooden architecture – research, restoration, planning of open-air museums – dates back to the 1950s–60s and was carried out under the guidance of the outstanding architect-restorers Alexander Opolovnikov and Boris Gnedovskii. The results of their work were presented in various books and articles in the 1970s–80s, but a significant portion still remains unpublished and is kept in the research archive of the TsNRPM, the conservation section of the architectural archive of the Shchusev State Museum of Architecture (OKh A A GM A), the stocks of the Kizhi State Open-Air Museum of History, Architecture and Ethnography, and elsewhere. As a result of the systematic multidisciplinary inspection of architectural monuments in the second half of the twentieth century, the creation of dossiers (‘passports’) on them and the restoration work undertaken, extensive stocks of registration and conservation documentation were formed in the archives of regional bodies for the conservation of monuments, scientific planning and restoration organisations. These are the St. Petersburg-based research and design institute NII Spetsprektrestavratsiia, the Karelian Republican Centre

RZ_wooden-architecture.indd 87

29.09.15 21:22

88

for the State Conservation of Objects of Cultural Heritage in Petrozavodsk, the regional Research and Production Centre for the Conservation of Historical and Cultural Monuments in Arkhangelsk, the autonomous regional cultural institution Vologdarestavratsiia in Vologda and the state budget-funded cultural institution Novgorod Scientific Restoration Administration in Novgorod. In the archives of these and other organisations, as well as firms carrying out restoration work, it is possible to glean the most detailed information about surviving churches and those lost in recent decades. They contain material from Vladimir Krokhin, Evgenii Vakhrameev, Boris Zaitsev, Leonid Krasnorech’ev and many other architect-restorers specialising in wooden architecture. The archives of the Kizhi, Malye Korely, Vitoslavlitsy and Kolomenskoe museums of wooden architecture contain extremely interesting collections of documents on those buildings that have been moved to them and more besides. Sources on timber architecture of the Russian North can be found in museum archives, from major local history museums (the National Museum of the Republic of Karelia and the Arkhangelsk Regional Local History Museum) and museum-reserves (the Novgorod State Integrated Museum-Reserve and the Vologda State Historical Architectural and Art Museum-Reserve) to small local history museums located in many towns and settlements in the northwest. Original historical documentation is of fundamental importance for practical research, especially in support of the restoration process. Investigation of primary sources will expand the range of objects to include those previously unexplored and broaden our vision of the church architecture of the Russian North. Archive documents will supply additional information on reconstruction and renovation of the surviving monuments and enable us to recreate the process of church building in the eighteenth and, especially, nineteenth centuries, when this sphere was managed and controlled by the Holy Synod. In contemporary Russia, where extensive restoration work is being undertaken, deep-reaching archival studies of historical sources and documents will help in understanding the broad experience of collaboration between different levels of church and civil authorities, the mistakes and losses caused by unprofessional and amateur actions and, of course, the inspiring achievements of cooperation between eminent architects and researchers, who have helped to preserve and maintain the unique legacy of Russian wooden churches.

RZ_wooden-architecture.indd 88

29.09.15 21:22

Russian State Archive of Ancient Documents

Moscow

State History Museum

Moscow

Historical Research Archive of the Russian Academy of Sciences St. Petersburg Institute of History

St. Petersburg

National Library of Russia

St. Petersburg

Russian State Library

Moscow

Russian State Historical Archive

St. Petersburg

Research Archive of the Russian Academy of Arts

St. Petersburg

Research Archive of the Russian Museum of Ethnography

St. Petersburg

Research Archive of the State Russian Museum

St. Petersburg

Research Archive of the Russian Academy of Sciences Institute of the History of Material Culture

St. Petersburg

Russian Academy of Sciences Peter the Great Museum of Anthropology and Ethnography

St. Petersburg

Research Archive of the Shchusev State Museum of Architecture

Moscow

Research Archive of Moscow State Integrated Art and Historical Architectural and Natural Landscape Museum-Reserve

Moscow

Research Archive of the Central Scientific Restoration Design Workshops

Moscow

PERIOD

Soviet (1917–91)

LOCATION

Late Medieval (C16–early C18)

INSTITUTION

Synodal (1700–1917)

89

Table 1: National Level Archives

RZ_wooden-architecture.indd 89

29.09.15 21:22

State Archive of Arkhangelsk Region

Arkhangelsk

Research Archive of the Arkhangelsk Local History Museum

Arkhangelsk

Research Archive of the Malye Korely State Museum

Arkhangelsk

Research and Production Centre for the Conservation of Historical and Cultural Monuments

Arkhangelsk

National Archive of the Republic of Karelia

Petrozavodsk

Research Archive of the National Museum of the Republic of Karelia

Petrozavodsk

Research Archive of the Kizhi State Museum-Reserve

Petrozavodsk

Republican Centre for the State Conservation of Objects of Cultural Heritage

Petrozavodsk

State Archive of Vologda Region

Vologda

Vologda State Historical, Architectural and Art Museum-Reserve

Vologda

Vologdarestavratsiia autonomous regional cultural institution

Vologda

NII Spetsprektrestavratsiia research and design institute

St. Petersburg

Research Archive of the Committee for the State Inspection, Use and Conservation of Historical and Cultural Monuments (KGIOP)

St. Petersburg

State Archive of Novgorod Region

Novgorod

Research Archive of the Novgorod State Integrated MuseumReserve

Novgorod

Novgorod Scientific Restoration Administration

Novgorod

PERIOD

Soviet (1917–91)

LOCATION

Late Late Medieval Medieval (C16–early (C16–early C18) C18)

INSTITUTION

Synodal (1700–1917)

90

Table 2: Regional Level Archives

RZ_wooden-architecture.indd 90

29.09.15 21:22

91

References 1 RGADA. Fund 1209. Pomestnyi prikaz, Votchinnaia kollegiia, Votchinnyi department. 2 RGADA. Fund 237. Monastyrskii prikaz (Kollegiia ekonomii). 3 OR GIM. Moscow Synod collection (Patriarch’s). 4 OR GIM. Moscow Synod collection (Patriarch’s). 5 RS NIA SPbII RAN. Fund 34. Vologda Arkhiereiskii dom. 6 RS NIA SPbII RAN. Collection 139. Collection of documents of the Ustiug eparchy. 7 RS NIA SPbII RAN. Collection 148. Collection of documents of the Kholmogory eparchy. 8 RS NIA SPbII RAN. Fund 171. Novgorod St. Sophia’s Arkhiereiskii dom. 9 RS NIA SPbII RAN. Fund 255. Kholmogory Arkhiereiskii dom. 10 RGADA. Fund 1209. Pomestnyi prikaz, Votchinnaia kollegiia, Votchinnyi department. 11 RGADA. Fund 1209. Pomestnyi prikaz, Votchinnaia kollegiia, Votchinnyi department. 12 RGIA. Fund 799. Service administration under the Synod. 13 RGIA. Fund 789. Academy of Arts of the Ministry of the Imperial Court. 14 OR NA IIMK RAN. Fund 4. Moscow Archaeological Society (1865–1920). 15 OR NA IIMK RAN. Fund 1. Imperial Archaeological Commission (1859–1919). 16 NA IIMK RAN. R-III. Dossiers on old churches. 17 OPI GIM. Fund of the museum section of Glavnauka of the RSFSR People’s Commissariat of Education (1918–44). 18 OPI GIM. Fund of the Moscow section of the State Institute for the History of Material Culture (GAIMK) (1919–37). 19 OKh AA GMA. Fund R-XIV. P. D. Baranovskii fund (1920–70). 20 NA REM. Fund 1. 21 OR NA GRM. Fund 98. Ye.E. Knatts fund. 22 OR NA GRM. Fund 187. Iu.N. Dmitriev fund. 23 OR NA IIMK RAN. Fund 1. Imperial Archaeological Commission (1859–1919). 24 OR NA IIMK RAN. Fund 2. State Academy of the History of Material Culture (1919–37). 25 OR NA IIMK RAN. Fund 29. K. K. Romanov (1882–1942), architect. 26 OR NA IIMK RAN. Fund 30. K. K. Romanov (1882–1942), architect, artist. 27 OR NA IIMK RAN. Fund 67. Department for museum affairs and the preservation of artistic and ancient monuments attached to the People’s Commissariat of Education. 28 OKh AA GMA. Fund 30. Main Administration for the Protection of Monuments (1930–60). 29 OKh AA GMA. Fund 31. Archival collection of materials on architectural monuments located in the former USSR (1930–60). 30 OKh AA GMA. Fund 32. Archival collection of materials from expeditions to investigate the architectural monuments of the Russian North and some regions of central Russia (1937–69). 31 OKh AA GMA. Fund R-XIV. P. D. Baranovskii fund (1920–70). 32 OKh AA GMA. Fund 35. Personal fund of the architect and artist P. N. Maksimov (1930s–60s). 33 OKh AA GMA. Fund 38. Personal fund of the architect and artist P. D. Sukhov. 34 OKh AA GMA. Fund 47. Personal fund of the architect and artist Iu. S. Ushakov.

RZ_wooden-architecture.indd 91

29.09.15 21:22

92

RZ_wooden-architecture.indd 92

29.09.15 21:22

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

RZ_wooden-architecture.indd 93

29.09.15 21:22

94

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 Stave churches, with their construction methods and ornamentation, are a distinctive contribution to Europe’s large and diverse cultural heritage from the Middle Ages. In Norway, there are twenty-eight surviving stave churches, in addition to the one that was moved to present-day Poland. In Sweden one survives, and the nave of a stave church-like wooden church is partially preserved in Greensted in Essex, England. Of the Middle Ages’ once rich diversity of wooden churches and stave constructions in Europe, it is therefore Norway where the vast majority are preserved. Urnes Stave Church in Luster, Sogn is the only one of these listed on UNESCO’s World Heritage List. It is one of the oldest surviving, dated to the 1130s, but with reused parts from an older church built not long after 1070 (Fig. 1). 1 In 2001, the Directorate for Cultural Heritage created a programme to improve the maintenance backlog of stave churches. The renovations are scheduled to be finished in 2015. During the course of the Stave Church Programme, the condition of all of the stave churches in Norway will be registered and the repair work executed. As part of the programme, a number of research-related studies were undertaken, among others in building archaeology, dendrochronology, material quality and tool markings. The purpose of these studies has been to obtain knowledge about topics that have previously been insufficiently emphasised in the research, including things like the stave church’s building process, craftsmanship and the quality of materials. Dating studies on the basis of growth rings of

Fig. 1: Urnes Stave Church from the north. Dendrochronologically dated to the 1130s, which fits in well with the stylistic dating. The northern portal and carved planks, along with many other materials, were reused from an older church at the site, built after 1070. The church is one of thirteen preserved stave churches with free-standing posts in the interior.

RZ_wooden-architecture.indd 94

29.09.15 21:22

95

pine have had as a goal both to date the timber and to provide knowledge about the wood’s provenance, age at time of felling, growth conditions and material quality. Some important building-archaeological issues have been investigated at individual churches to look at the durability of established perceptions about the development of stave church architecture. In the following pages, this article will discuss some preliminary results, look at them in the context of some established beliefs, and point out questions they raise concerning stave church research.

A short overview of the source material The written sources on the stave church’s older and oldest history are very few, as Jørgen H. Jensenius discusses in his article in this book. Therefore, it is the remains – archaeological traces, the churches themselves, preserved portals and materials from lost stave churches – that are the most important sources in stave church research. Most of the knowledge about stave churches builds on what we can deduce from them. Today’s 28 surviving stave churches in Norway represent less than three per cent of the minimum estimate of stave churches in 1350. 2 None survives in its original form. Around 20 churches are known from sketches and surveys. If one counts the stave churches that were more or less documented before demolition, the number approaches nearly double that, i.e. we have some certain knowledge of about five to six per cent. Around 130 portals and parts of portals are preserved, of which about 50 are from churches that have been demolished. If we add the portals of lost churches, the percentage rises, generously seen, to around ten per cent. Systematic surveying of preserved or reused stave church parts could probably expand the picture even more. The same could be said for a review of older church records and descriptions of lost churches. In addition to the Norwegian material, there are proven archaeological traces of wooden churches from the Middle Ages in the Nordic countries, on the Continent, and in the British Isles. 3 On the whole, it is nevertheless a fragmentary and – historically seen – randomly selected set of materials from what once existed. Our knowledge and expertise about these distinctive medieval churches will therefore be limited by the source material. General conclusions about stave churches’ carving and architecture will be, in the best case, very uncertain.

Building processes, tools and timber quality Timber quality, markings left behind by tools, and details of fabrication and assembly are important sources to help increase our understanding of the practical building process involved in constructing a stave church. A united discussion of the wood and material qualities of stave churches has not yet been published, but systematic observations have been made during the Stave Church Programme and will be compiled and made available as documentation. Carpenter Hans Marumsrud and forester and wood expert Kristen Aamodt have investigated the materials and craftsmanship in the two stave churches at Nore and Uvdal for the Stave Church Programme. 4 The churches are found

RZ_wooden-architecture.indd 95

29.09.15 21:22

96

at the top of Numedal, one of the main valleys in eastern Norway. Marumsrud and Aamodt have assessed the timber quality in different parts of the buildings, the trees’ age, twigs, the location of the heartwood, which part of the tree was used and, if the root ends face upwards or downwards (Fig. 2). The trees seem to have grown straight and steadily. The planks in the west wall in Nore are made from trees that had grown for 300 years. In the split material, the unstable spines of the heartwood were removed. The materials appear to have been used very consciously. In Nore, for example, the planks’ heartwood side faces inwards and all of the staves that were possible to observe had the root end facing down. In the ceiling construction, the root end was systematically used on the top of the rafters in the south, but at the very bottom in the north. Today we do not have an explanation why, but hopefully future research can give us some answers. In order to assess the markings left by tools and drying, Marumsrud and Aamodt performed experiments with older, traditional carpentry tools, making fresh timber into new wall planks. The new Fig. 2: Nore Stave Church seen from the southwest. The church, which has a central post, is dated to around 1170 and is one of the few known stave churches probably built with a transept. It has been changed and extended several times over the centuries but the core of the stave construction is preserved in situ.

materials were hewn from logs cut in half and were cut even and planed with older traditional tools on the heartwood side. After a short while, the planks dried radially and the flat surface got a slightly convex cross-section, similar to what one observes in the original wall planks. Marumsrud and Aamodt point out that the wood in Nore is not discoloured, suggesting it has not been exposed to direct sunlight and consequently had not been stored for a long time. The tool marks in the wood follow shifts of drying cracks, a clear sign that the wood was worked on before it had dried. 5 After experiments and visual analysis of the materials in the churches, the report concludes that both Nore and Uvdal were built with fresh wood. The study’s methods have not yet been applied in other stave churches and are still awaiting critical review and assessment. However, there is little doubt that the results give a new and promising direction for further studies and that they show potential for different academic approaches in stave church research. Growth ring samples from Nore have dated two staves to have been felled in the winter of 1166–67. On the basis of these test results, the report concludes that the stave church at Nore was built in the summer of 1167. 6 This conclusion presupposes that all the timber was felled during that same winter. It is tempting to make this conclusion, however, it is only based on two samples from the central mast and the nave’s south-western corner stave respectively (Fig. 3).

RZ_wooden-architecture.indd 96

29.09.15 21:22

97

Growth ring samples, material quality and reuse At Urnes, the situation is a little different. Nearly all accessible parts that could be growth-ring dated using contemporary methods have been dated. Considering the number of samples, this provides a solid basis for dating conclusions. Through these samples we have also gained a detailed picture of the materials’ quality and the age of the trees at the time they were felled. The investigations at Urnes have provided some interesting results that are significant for the assessment of the investigations in Nore and Uvdal (Fig. 4). Fig. 3: Ground plan of Nore shows where the dendrochronological samples are extracted and the dating results. (Drawing: Håkon Christie, Riksantikvaren)

Urnes Stave Church is dated to the 1130s7 and was partially built with materials from an earlier church. Two foundation pillars, a corner stave and the planks in the nave’s lower walls, richly carved planks and a portal in the north wall, and the gable planks in the east and west are reused from this earlier church, which was probably built shortly after 1070 and demolished around 1130. 8 A portion of the plain wall planks are sculpted with precisely crafted cross-sections forming a wave pattern in the wall’s surface. Inside, the staves have richly cut reliefs and capitals with some of Norway’s most advanced early Romanesque preserved architectural sculpture. Archaeological excavations have shown that the current church is the fourth on the site. No other stave church is discussed and described as much as Urnes, which is the subject, among other works, of two huge monographs, numerous articles and pictures used in an enormous range of contexts.9 The church is a key monument, both for research on stave churches and on pre-Romanesque and Romanesque architectural sculpture in Norway. The dating of materials felled for the current building and its predecessor, together with knowledge about the material quality and processing, are consequently of great import.

Fig. 4: Ground plan of Urnes shows where the dendrochronological samples were extracted from the lower walls and the dating results (Drawing: NIKU – Norwegian Institute for Cultural Heritage Research, antiquarian archive, Riksantikvaren)

RZ_wooden-architecture.indd 97

29.09.15 21:22

98

The reused materials from the earlier church are consistently processed such that the sapwood is removed, leaving only the heartwood. The trees were old when felled. This shows a conscious strategy to use only the firmest part of the tree and that there had been access to suitably stocky timber with a high proportion of heartwood. Because the sapwood is removed, it has not been possible to determine the felling years for the timber of these parts. Many wall planks have an outermost growth ring that does not exclude them from having been felled for one of the first two churches at this location. 10 The material quality and the craftsmanship on the current church differs markedly from the earlier retained parts. The timber was felled in a younger forest, the sapwood is retained, and adaptions between the church’s different parts are far rougher. 11 The dating of a sill, two staves and one of the foundation pillars was published in 1999. 12 These datings show the felling years to be the winter of 1129– 30, the winter of 1130–31, and the summer of 1131. Most of the planks can be dated to the same period. Because of the very stocky timber, some of the planks were not possible or very difficult to date. One plank received a surprising provisional dating, to around 1135 or later, but this has now been disproved after prolonged and time-consuming studies. The conclusion of Thomas Bartholin and Terje Thun is that this timber was felled after summer growth halted in 1131, i.e. sometime between the autumn of 1131 and the spring of 1132. 13 This means that the materials from the existing church, as far we can presently tell, were felled over a minimum period of two years and a maximum period of three and a half years. There is uncertainty here, in that some of the materials felled for the church have not been possible to date. So they could have been felled at an earlier or later period. This means the possibility that the church was built over a longer period than the present growth rings indicate cannot be excluded. We do not know the length of the building period or when the church was finished. Tool markings in the materials have not yet been systematically evaluated in the same manner as in Nore and Uvdal, but would probably produce interesting results. Compared with the conclusion on Nore’s building time, the growth-ring samples at Urnes show a two to three times longer time period. That does not mean building time but rather the time at which the materials were felled.

Dendrochronology – what is dated? The dating of Nore and the conclusions concerning the church’s building time are based as previously said on two growth-ring samples from two staves, i.e. from two absolutely essential parts of the church’s construction. Since the materials are not discoloured, the report concluded that the church’s other materials were felled simultaneously. This cannot be excluded, but based on the spread of growth-ring dates at Urnes, one should be careful not to draw hasty conclusions about Nore’s construction. Even if a span of two to four years is hardly a significant amount of time in a longer historical perspective, it is important when trying to find out about the stave church’s building process. For

RZ_wooden-architecture.indd 98

29.09.15 21:22

99

all we know, other stave churches may have materials that were felled over a longer period than they were at Urnes. It is quite possible that future investigations will show huge individual differences between the preserved churches. The growth-ring dating at Urnes gives substance to questions concerning the extent to which reused older materials were used in stave churches. Preserved fragments of carvings in the style found at Urnes from Torpo, Hopperstad and Feios (Rinde) among others, testify to the reuse of materials from earlier churches. 14 For materials without ornamentation, one has been referred in the research to conclusions based in part on markings left behind by re-axing and in part on the processing work done to cross-sections and joints. This can provide some clues to help estimate the relative age difference between various types of wall planks, but with the considerable uncertainty that will always reFig. 5: Hopperstad Stave Church dated to the 1130s, dendrochronological survey by measuring on the wall planks. The tapes carry markings for every tree ring on the actual plank.

sult, as different forms may occur at the same time in different places. Typology, as a dating method, is not very suitable with such numerically limited materials as found in stave churches. Growth-ring analyses with new testing methods have critically changed dating possibilities in Norway, for example of wall planks made of pine. Dendrochronological sampling methods have been further developed through the investigations at Urnes. Wall planks made of pine logs split in half provide the same growth-ring patterns on the heartwood side as with a drill sample in principle. Photographing the appropriate sample locations avoids destructive intervention. The sample location is marked with adhesive tape at the location of each individual growth ring and where the sample location is found in the building. Afterwards, the sample site with the markings is photographed. The photograph is then analysed under a microscope and compared with the appropriate reference curves. The adhesive tape with the markings shows the relationship between growth rings in a 1:1 ratio and is consulted when necessary. This method has proven to provide reliable results, which have made it possible to extensively analyse stave church planks, although round timber is still only possible to date with drill samples (Fig. 5). 15 The large number of measurements done with the photographic method have broadened the basis for local reference curves and also added a statistical basis for calculating the ratio between sapwood and heartwood in age-specific pine from local deposits. This could provide a basis for progress toward a likely growth-ring dating system where samples of the outermost growth rings are missing. 16 With the comprehensive measurements possible with the photographic method, one can also get a more detailed picture of the felling year and the progress of the construction process as at Urnes and not the least get a broader picture of the extent to which materials were reused in stave churches. The growth-ring dating at Urnes also points to a methodological challenge in using growth-ring dating in building research in general and stave church research in particular. The felling year and the building year are two different issues and must be separated from each other. One cannot conclude that the felling year for a tree is the same as the building year. There is also significant

RZ_wooden-architecture.indd 99

29.09.15 21:22

100

uncertainty concerning how much weight a few samples should be given for dating materials in an entire building and for determining the building year from that. Dendrochronology is important and central to building research, as the samples from Urnes show. Nevertheless, growth-ring dating does not replace building archaeology and the critical analysis of style. The results must be considered in light of other source material, archaeology and building archaeology and material analysis, along with relations connected to architecture and historical styles. It is precisely this interaction between disciplines that has proven so fruitful at Urnes.

Hopperstad and building typology In August 2008 and 2009, some investigations were carried out in the Hopperstad Stave Church as part of the Directorate of Cultural Heritage’s Stave Church Programme. 17 Hopperstad is given a key role in stave church literature and the presumed development of stave church architecture. 18 The purpose of the investigations was to look for a relationship that could cast light on the church’s building history and thus corroborate or disprove established views on the presumed development of the stave church’s construction. The preliminary results were published in the Society for the Preservation of Ancient Norwegian Monuments’ yearbook in 2010. 19 The main points will be briefly repeated here together with the results of a supplementary examination in March 2014 (Fig. 6). For more than 100 years, typology has provided the foundation of beliefs about the stave church’s architectural development, classifying stave churches by specific constructive characteristics. In a sense, this typology of stave churches was established as a kind of parallel to the history of style. The use of typology as a dating criterion has been debated, including whether it is methodologically justifiable to assert a type-based line of development founded on

Fig. 6: Hopperstad Stave Church viewed from the southwest.

RZ_wooden-architecture.indd 100

29.09.15 21:22

101

only 28 surviving stave churches and the handful of demolished stave churches known in measured sketches. 20 The discussion has rarely been based on demonstrable archaeological building conditions. How much weight do hypotheses have that are based on, conservatively speaking, five to six per cent of the material mainly derived from the inner fjords and valleys in southern Norway and from a limited period of the Middle Ages? This issue is related to stave churches where the nave has a number of free-standing internal staves or posts that support the upper roof (Fig. 7). The best-known examples are Borgund and Urnes. Some of these stave churches have all of the free-standing inner staves standing at full height from the nave’s ground beams as close as the middle stave in the church’s longitudinal axis that stands on a crossbeam supported by an arch. 21 For simplicity’s sake, let us call them group A. Urnes, which is a church in group A, has been dated as one of the oldest since Fig. 7: Exploded view of a stave church with interior posts (Borgund). The construction rests on a double frame with ground beams. The wall sills rest on the ends of the ground beams, while the free-standing interior staves are tapped into them. (Drawing: Håkon Christie, Riksantikvaren) Fig. 8: H. M. Schirmer’s drawing of the supposed evolution of the construction of stave churches with freestanding interior posts. From left: Urnes, Hopperstad and Borgund.

the beginning of stave church research, while those which we can call group B here have been seen as younger and more ‘developed’. Group B is characterised by the fact that the free-standing inner staves have one or two horizontal beams, string beams that lock the staves sideways. Between the string beams and the staves there are cross braces, the so-called St. Andrew's Crosses. The string beams are again supported by knee brackets made from tree roots in the angle between the horizontal beams and the vertical staves. It is the horizontal beams and the cross braces that are considered to be characteristic of this younger and more developed style (Fig. 8). Group B’s churches have, with three exceptions, two or more staves that are on the lowest horizontal beam, i.e. they have intervals between the number of staves that reach the full height from the floor. 22 But can stave church architecture and the individual building history of each church be put into such clear-cut categories? Three churches do not fit into such a clean dichotomy of types. One of these is Hopperstad, which is agreed to be one of the oldest preserved, based on historical style dating (Fig. 9). This dating is to a certain degree confirmed by the growth-ring dating of one of the staves in the nave, which was felled in the winter of 1130–31. 23 Other materials with a bark edge have not been growth-ring dated at Hopperstad, but the dating of original wall shingles without bark edges is consistent with the dating of the stave, such that it is preliminarily possible to justify dating the church to the 1130s. However, it seems that a number of planks in the chancel's north wall may

RZ_wooden-architecture.indd 101

29.09.15 21:22

102

have been reused from an older building when the church was built. These planks also have a different cross-section than the others. 24 With a dating roughly contemporary with Urnes, Hopperstad is a key monument in the discussion of stave church architecture’s presumed development. Are the string beams and the cross braces in the nave and the support beams and cross braces in the chancel later additions, or are they part of the original construction? The church’s exterior and interior are the result of extensive restoration and repair work in the 1880s, after the church went out of ordinary use. Fig. 9: Hopperstad southwest corner of the interior. The intermediate stave or post in front of the west doorway is resting on the horizontal string beam which connects all the free-standing staves around the nave. There is a similar solution at the eastern end of the nave, in front of the chancel entrance.

The construction with four portals, ground beams, wall sills, staves, wall plates and roof structures is, however, intact. The chancel screen is the only one preserved in any of the stave churches. The nave’s inner staves have cubical capitals with deep, blackened moulding profiles. Under the upper wall between the staves, nailed and profiled arc segments form a continuous arcade around the whole room. The chancel screen portal has a design that captures and varies the motifs on the arcading in the nave, here half-columns with non-profiled cubical capitals and a profiled archivolt. The staves are bound together horizontally by one string beam, which is supported by knee brackets. These also form a sort of lower simplified arcade around the whole room. Powerful cross braces span between the staves. These are plugged firmly against the capital’s shield surface and diagonally toward the corner between the string beam and the staves. The crossarms are uniformly wide, about 20 × 12 cm in cross-section. They are finely crafted on the surfaces visible from the middle of the nave and are lined with flat-bottomed, blackened profiles. Corresponding profiles are used on other structural parts of the church. At the point of intersection, each cross brace has a circular field marked with a Celtic cross surrounded by three circles. These Celtic crosses are very similar to the four consecration crosses that are preserved in the nave and chancel.

RZ_wooden-architecture.indd 102

29.09.15 21:22

103

Fig. 10: Hopperstad, upper part of the chancel on the east. The big cross braces are standing on beams that are tapered into the two free-standing posts in the chancel. The decoration of the braces is similar to the one of the braces in the nave. Note the double profile on the vertical brackets on the interior post, similar profiles are also used on the corner post in the nave.

The diagonal crosses in the chancel, two on each long wall and one on the east wall, are similar to those in the nave, but with longer crossarms (Fig. 10). Together with the underlying supporting beams, these crosses appear secondary. But that can hardly be the case. The cross arms’ upper attachment is meticulously and precisely tapped into the adjacent knee brackets and must have been installed before the lower side roof of the chancel was assembled. Inspection from internal scaffolding in spring 2014 showed that the rafter on the northern side roof is made from hand-planed, axed material. This is consistent with the workmanship of other original materials in the church. Therefore, the rafters cannot be from the restoration, as has been previously assumed. 25 The rafter is fastened on the top edge to one of the crossarms and not into the sill of the upper wall as at Urnes, i.e., the side roof’s support is partly anchored in the diagonal crosses in each of the chancel’s long walls. If the diagonal crosses were to be a secondary solution, it would also mean that the support of the side roof must also be secondary. That does not seem very reasonable and there are no visible traces of any such connection. A far simpler and more rational solution would have been to give the reinforcement a design and placement that did not come into conflict with an existing part of the construction. In the nave, the traces are more complex. The upper arms of the cross braces are nailed to the capitals’ shield surface. The crosses are clearly inserted from the rear. Here the surfaces of the capitals are visibly cut off, but not so deep that the blackened moulding of the surface is not to be found on several capitals. So much meticulous work on the profiles and surfaces could hardly have been

Fig. 11: The capitals on the interior posts in the nave are clearly cut off with an axe to facilitate the joints between posts and cross braces. Traces of the blackened profile were left after the axing was finished. They are still visible on several capitals.

done only to have it cut away if the diagonal crosses had been planned when the capitals were made (Fig. 11). From the traces alone it is not possible to determine whether this is a change made during construction or at a later date. At the same time, it is evident that the horizontal beam must be original. It bears the central staves in front of the western entrance and the chancel opening in the east. There is no evidence of any other support with strong beams and powerful knee brackets such as can be seen at, for example, Kaupanger and Urnes. What one can say with great certainty is that the cross braces in the chancel and their supports are original, together with the nave’s string beams. The design of the crosses in the nave and the choir are the same. Together, this indicates that the cross braces in the nave have come about as a result of changes during the construction stage, but it will probably not be possible to get unquestionable and definitive evidence. 26 The essential and decisive thing is that the string beams in the nave and the cross braces and supporting beams in the chancel must have been inserted in the construction during the building process at the time of erection. A number of stave churches belonging to group A also have diagonal crosses, St. Andrew’s Crosses, in their construction. This does not correspond with the typological evolution theory and consequently has been explained as being later changes and supports. 27

RZ_wooden-architecture.indd 103

29.09.15 21:22

104

Common to several of these are the cross braces, either in the chancel or in the nave, where the beams that carry the crosses are supported by knee brackets. The brackets are finished with a characteristic profile, i.e. a double torse of a similar shape to that found at Hopperstad. The diagonal crosses are also thought to be secondary in these other churches. This presupposes, however, that the double torse profiles that are frequently used in these churches must also have been carefully copied on new joints of later additions. Strangely enough, this copying of entirely subordinate details must have happened in four different churches. These are Kaupanger, Hopperstad, Årdal and Grinaker. The latter two were demolished in the nineteenth century and are only known from drawings. Three of these are on the western side of Langfjellene, dividing southern Norway, and one is in a wide rural area of eastern Norway, about 300 kilometres further east. Such convergence is unlikely. The facts, as far as they allow themselves to be tested with the surviving materials, point quite clearly to one conclusion in this author’s opinion: The cross braces and, to a certain extent, string beams, are original parts of the construction in these churches. These are structural joints that, in other words, can be followed back to the 1130s in preserved stave churches. The idea of a gradual development of stave churches with free-standing staves lacks foundation in facts. Using typology as a dating criterion lacks concrete grounding in clear building-archaeological conditions and largely uses secondary sources like measuring sketches as supporting evidence. Such source material involves significant critical challenges that have not been discussed: How is it possible to draw Fig. 12: G. A. Bull: ink drawing of Feios Stave Church (Rinde) from a sketch, 1854

far-reaching conclusions about the details of a church’s building history on the basis of a principal sketch done in one day, especially where the object disappeared 150 years ago? This also raises some theoretical and methodological issues in stave church research. What is it that one chooses to compare? How does one choose what to compare and to what end? What does the similarity between sculpture, construction and detail consist of?

Similarity in sculpture The characteristic cubical capitals in Hopperstad once existed in a number of stave churches, as mentioned above. Such similarities in sculpture are normally seen in art history as a contemporary phenomenon, as is the case when it comes to these portals and their churches that are dated stylistically to the second quarter of the 1100s. 28 Feios stave church with cubical capitals was demolished in 1866, but it is known from measurement sketches by G. A. Bull from 1854 (Figs. 12, 13). Some simple investigations of reused stave church materials from this

RZ_wooden-architecture.indd 104

29.09.15 21:22

105

church were done in the Stave Church Programme. In the present church from 1866 there are five wall planks and some fragments of planks besides some reused materials from the previous stave church. The materials were examined to see if it was possible to date them dendrochronologically. So far, this has only been successful with one of the planks. This one has a bark edge and its outermost ring has been dated to 1173. 29 In comparison, there are growth-ring samples with bark edges dated to 1131 for Hopperstad (one sample of the outermost growth ring) and to 1137– 38 for Kaupanger (three staves). 30 The plank from Feios is dated to respectively 42 and 35 years later. If the dating were representative for the portal and the church in its entirety, this would throw new light on the current style-historical dating. Hopperstad and Kaupanger’s growth-ring dating fits like a glove with the stylistic dating, but not Feios (Rinde). How can this be explained? Are the seemingly uniform cubical capitals that existed in inner Sogn partly a result of a stylistic fixation or petrification in architectural sculpture? Fig. 13: Hopperstad Stave Church, portal in chancel screen. This is today the only preserved portal of this kind in any stave church.

Considering the dating from Feios, one can argue that it only consists of one plank, and that from a church which, aside from scattered re-axed materials, is only known from a rapidly executed measured sketch. In principle, the plank could be from a rebuilding, an extension or repair. We also do not know where it was placed in the church. The sketch provides no information about the details and quality of materials or workmanship. Even the portal that Bull drew could have been reused. At Feios, there is also a preserved plank with carving in Urnes’ style, probably from the late 1000s or around 1100. 31 It may seem farfetched that the church would have reused materials from two different periods. On the other hand, it is proven in Urnes that four churches were built within 100 years and in Kaupanger three. 32 The source material is, so far, not sufficient to draw conclusions on Feios and the situation calls for the further investigation of preserved materials from Feios, as well as Hopperstad and Kaupanger. The growth-ring dating from Feios is a warning against arriving at general conclusions too hastily on the basis of a) one piece of material without a certain structural context, b) apparent superficial similarities, c) secondhand evidence, as is the case with capitals that are only known from drawings. Nevertheless, it is natural to sketch a possible connection between at least Kaupanger and Hopperstad, a possibility that growth-ring dating does not preclude. But what does this connection imply?

RZ_wooden-architecture.indd 105

29.09.15 21:22

106

Similarities and differences in construction The constructions in Hopperstad and Kaupanger have some striking similarities when it comes to the support of the chancel’s upper walls. Both churches have horizontal binding beams between the chancel’s free-standing staves and between these staves and the nave (Fig. 14). Hopperstad has, as discussed, two cross braces on each long wall, while Kaupanger has one vertical plank. This represents two solutions to the same problem: how to support the upper wall’s sill. The chancel in Hopperstad has a frame of ground beams on its north, east and south sides while Kaupanger only has a single ground beam on its east side. Kaupanger has had two portals with cubical capitals as in Feios, Hopperstad, Grinaker and Fortun, and probably also a so-called bow portal similar to the chancel’s north portal in Hopperstad. 33 The knee brackets in the corners of the upper walls and the choir’s binding Fig. 14: Fortun (Fantoft) Stave Church portal on the west side. The church was set alight and burned down in 1992.

beams have, as already stated, a similar style in Hopperstad and Kaupanger: In both churches these knees are finished with a characteristic double torse profile. The design is similar, but there are some differences. The dimensions in Hopperstad are more powerful and probably more significant, but the execution is not quite the same. The finishing of the brackets in Hopperstad continue in the bracket’s line over the torse profile, while in Kaupanger they come down at a different angle. In Hopperstad there are clear scratch marks on the staves. These marks correspond both with the knee brackets’ torse profile and with the brackets’ lower closure. It would be reasonable to interpret these as marks made by the carpenter or carpenters during the construction. These scratches are found on all free-standing staves with profiled brackets in Hopperstad, but have not yet been found in Kaupanger. Similarly, the eastern cross brace’s mounting between the chancel’s staves is different in Hopperstad and Kaupanger. The cross-section of the binding beam between the nave and the choir is also different in these two churches. In Hopperstad, the cross-section is rectangular, while in Kaupanger it is shaped like a ‘T’.

Some final methodological considerations This leads to questions about what can be deduced from similarities, or rather degrees of similarity. What similarities and differences are significant for reaching conclusions about craftsmen, workshops, craftsmanship and workshop traditions? Are markings and lack of markings relevant, as in Hopperstad? What about workmanship, joints and various assembly sequences? Craftsmen have different ways of doing the same work, depending on their individual experience and whom they have learned from. 34 This will necessarily leave its marks. How can one distinguish individual solutions from shifting workshop traditions and master builders? Is the variation in solutions for the same building the result of different building historical phases or different craftsmen working simultaneously? The Stave Church Programme has systematically investigated tool markings and the treatment of materials in an effort to acquire more knowledge about building processes. This is an area that requires different expertise than the traditional disciplines of art and architectural history. New methods extend

RZ_wooden-architecture.indd 106

29.09.15 21:22

107

the use of buildings as sources. Knowledge about a building’s craftsmanship and the tools historically used is necessary to get a closer understanding of the stave churches as historical relics and to delve deeper into them. The results of these studies in the Stave Church Programme are still being compiled, but when they are, they could provide important new knowledge that has not been previously available in stave church research. There is a challenge when marrying handicraft and academia. Observations and assessments of tool use, tool markings and material processing must be made available for verification and critical source assessment. They must be documented in a form that allows them to be used in a scientific context. In art and architectural history research, the analysis Fig. 15: Sketch showing the principal solutions of the chancel construction in Hopperstad (top) and Kaupanger (bottom). (Drawing after H.-E. Lidén, 1978)

of architectural sculpture, the processing of the surfaces and materials, and spatial relationships are key methods of understanding architecture, workshops and craftsmanship traditions, style influences, dating questions and so on. In the analysis of stave church sculptures, the likeness or similarity of seemingly unimportant details may be the key to ascertaining the relationship between portals and craftsmen. 35 The visual analysis of materials and tool markings is not essentially different from an analysis of style, but it requires different knowledge, the craftman’s knowledge. Is it possible to establish fruitful criteria for coming closer to understanding craftsmanship traditions and various workshops in stave church research? I believe it is, to a certain degree. Experience gained from the Stave Church Programme is promising. The task for future stave church research requires the participation of all the different disciplines that are involved.

References 1 For discussion and historical overview on Urnes see Hohler, E. B., Norwegian Stave Church Sculpture, Vol. I, Oslo: Scandinavian University Press, 1999, 234–241. For the church’s dendrochronological dating see Storsletten, O., ‘Nær en datering av de gjenanvendte deler i Urnes stavkirke!’, Fortidsminneforeningens årbok 2002, Oslo: Fortidsminneforeningen, 2002, 91–99. For a general presentation and the latest discussion of stave churches see Anker, L., The Norwegian Stave Churches, Oslo: Arfo, 2005. 2 Christie, H., ‘Stavkirkene: Arkitektur’, in: K. Berg (ed.), Norges kunsthistorie 1, Oslo: Gyldendal, 1981, 185. 3 For the latest overview see Ahrens C., Die frühen Holzkirchen Europas I–II, Stuttgart: Konrad Theiss Verlag, 2001. 4 Marumsrud, H., K. Aamodt, Nore og Uvdal stavkirker. Tømmerkvaliteter og materialfremstilling, unpublished report, 2002, antiquarian archive, Norwegian Directorate for Cultural Heritage. An overall discussion of the use of materials in stave churches is not available.

RZ_wooden-architecture.indd 107

29.09.15 21:22

108

Craftsmanship historian Jon Bojer Godal has published a number of works on traditional craftsmanship and use of materials. See for example Godal, J. B., ‘Timber Used in the Church from Haltdalen’, in: E. Seip (ed.), A Stave Church for Iceland, Oslo: NIKU, 2000, 91–97 and Godal, J. B., ‘Om materialkvalitet i ein del mellomalderhus’, in: Berg, A. (ed.), Norske tømmerhus frå mellomalderen 6. Hus for hus. Tillegg og tidfesting, Oslo: Landbruksforlaget, 1998, 216–252 (with English and German summaries). Hohler 1999 discusses portal production and woodcarving, especially in Vol. I, Ch. 5 and Vol. II, Ch. 4. 5 Marumsrud and Aamodt, 2002. 6 Ibid. 7 Storsletten, 2002. 8 Krogh, K. J., Urnesstilens kirke, Oslo: Pax Forlag, 2011, 169, 214. 9 The literature on Urnes is comprehensive. Hohler 1999 II discusses and gives an overview of the research up to 1999. The current church is extensively described in Christie, H., Urnes stavkirke, Oslo: Pax Forlag, 2009. The predecessor is described and discussed in Krogh, 2011. The monographs about Urnes are independent of the Stave Church Programme. 10 Krogh, 2011, 120. 11 Ibid., 201, 223. 12 Christie, H., O. Storsletten, T. Thun, ‘Dendrokronologisk datering av gamle norske hus’, in: Fortidsminneforeningens årbok 1999, Oslo: Fortidsminneforeningen, 1999, 141–152 (with English summary). These studies are continued in the Stave Church Programme, see below. 13 Supplerende analyser på fotografier utført i 2013, unpublished report, 11.1.2015, NTNU – Museum of Natural History and Archaeology, laboratory for dating. A-284 Urnes stavkirke – stavkirkeprogrammet, no. 09/00215, antiquarian archive, Norwegian Directorate for Cultural Heritage. 14 See Hohler, 1999, I; and Krogh, 2011, for a closer description and discussion. 15 Bartholin, T. (2012), Fotometoden. Splintstatistik og fældningsår. Dendrokronologisk datering uden indgreb og uden barkkant. Erfaringer fra Urnes, Kaupanger og Hopperstad stavkirker og Tømmerhuse, unpublished ms 5.3.2012, antiquarian archive, Norwegian Directorate for Cultural Heritage. See also Storsletten, 2002, 92 ff. 16 Bartholin, 2012; Krogh, 2011, 214. 17 Befaringsrapport Hopperstad stavkirke 21.8.08, Notat befaring 19.08.09, A-291 Hopperstad stavkirke, antiquarian archive, Norwegian Directorate for Cultural Heritage. 18 Dietrichson, L., De norske stavkirker, Christiania: Alb. Cammermeyers Forlag, 1892, 239; Bjerknes, K., ‘Romansk og gotisk i våre stavkirker. En studie omkring Hopperstadkirken’, in: Fortidsminneforeningens årbok 1944, Oslo: Fortidsminneforeningen, 1947, 7–42; Bugge, A., Norske stavkirker, Oslo: Dreyer, 1953, 25, 27; Hauglid, R., Norske Stavkirker. Dekor og utstyr, Oslo: Dreyer, 1973, 164; Lidén, H.-E., ‘Den gamle kyrkja i Årdal’, in: D. Krossen (ed.), Bygdebok for Årdal II., Årdal, 1978, 760 ff; Nordhagen, P. J., ‘Stavkyrkjene’, in: N. G. Brekke, P. J. Nordhagen, S. S. Lexau, Norsk arkitekturhistorie. Frå steinalder til det 21. århundre, Oslo: Samlaget, 2003, 101. Other views from Hohler, 1999, II, 72; Anker, P., Stavkirkene, deres egenart og historie, Oslo: Cappelen, 1997, 271 f; Anker, L., The Norwegian Stave Churches, Oslo: Arfo, 2005, 33 ff. See also Hohler 1999 II, especially Ch. 2, Ch. 9, and 1999 I Catalogue for the field’s historical overview and discussion. 19 Anker, L., ‘Typologiens tvangstrøye’, Fortidsminneforeningens årbok 2010, Oslo: Fortidsminneforeningen, 2010, 129–144.

RZ_wooden-architecture.indd 108

29.09.15 21:22

109

20 Anker, P., 1997, Hohler, 1999, II, Anker, L., 2005, see also footnote 18. 21 Group A consists here of Urnes, Kaupanger, Årdal (demolished), Hafslo (demolished), Fortun ( burnt down at Fantoft 1993). The last three are known from surveys and partially from photographs, see the Norwegian Directorate for Cultural Heritage for the individual churches. Fortun is also described in the literature, see Bjerknes, K., ‘Fantoftkirken’, in: Fortidsminneforeningens årbok 1940, Oslo: Fortidsminneforeningen, 1942, 49–84; Lidén, H.-E., ‘Fortun – Fantoft. Forholdet mellom original og rekonstruksjon’, in: Berg, A., E. Elster, E. B. Hohler, H.-E. Lidén, O. Storsletten, S. Tschudi-Madsen (eds), Kirkearkeologi og kirkekunst, Øvre Ervik: Alvheim & Eide, 1993, 55–67. 22 Group B consists of Borgund, Gol, Hegge, Høre, Lomen, Heddal, Ringebu. Flesberg has been changed such that it cannot be determined what group it fits in. The three wooden churches that do not fit in are Hopperstad, Lom and Torpo. Hopperstad is the centre of the debate in this context and is therefore kept out. Lom is often counted as secondarily changed from A to B, but the building’s history here has not been clarified and it is therefore kept out of the discussion. The research agrees that Torpo has its original diagonal cross and horizontal beam. At the same time, all the free-standing staves are held in the foundation pillars and therefore the church does not fit into the typology’s pure categories. 23 A-291 Vik i Sogn og Fjordane dendrokronologisk analyse av Hopperstad stavkirke, unpublished report 18.12.2012 from NTNU – University Museum, section for archaeometry. A-291 Hopperstad, antiquarian archive, Norwegian Directorate for Cultural Heritage. 24 Ibid. 25 Anker, L., 2010, 134. 26 Ibid., 136. 27 Ibid., 139 ff, see also note 18. 28 Hohler, 1999, II, 70 ff. 29 Feios (Rinde) Stavkirke, Stavkirkeprogrammet, Dendrokronologiske analyser 2013, unpublished report 25.9.2013, Thomas Bartholin, Scandinavian Dendro Dating, A-289 Feios, antiquarian archive, Norwegian Directorate for Cultural Heritage. 30 Ibid. 31 Kaupanger stavkirke. Dendrokronologiske analyser 2014, unpublished report, 25.7.2014, Thomas Bartholin, Scandinavian Dendro Dating, A-287 Kaupanger, antiquarian archive, Norwegian Directorate for Cultural Heritage. 32 Krogh, 2011, 204. 33 Ibid., 194 ff.; Lidén, H.-E., ‘The Predecessors of the Stave Churches of Kaupanger’, in: Bjerknes, K., Lidén, H.-E., The Stave Churches of Kaupanger, Oslo: Fabritius & Co., 1975. 34 Hohler, 1999, I, 184. 35 Ibid., 28; Hohler, 1999, II, 46. All illustrations are by the author except where noted.

RZ_wooden-architecture.indd 109

29.09.15 21:22

110

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

Andrei Bode Wood was the dominant building material used in Russia for many hundreds of years. The abundance of timber and the ease with which it could be worked made wooden structures readily available and inexpensive. Furthermore, in comparison with masonry edifices, wooden houses were warmer and drier, and therefore more comfortable for habitation. Expertise in building with wood was accumulated and perfected over the centuries, being handed down from generation to generation. This made it possible to develop the most rational construction techniques and striking architectural forms. Besides the study of archaeological data and written documents, particular attention needs to be devoted to the building material and the methods of preparing and working it, the tools employed and methods of ensuring the functioning of various elements in the structure of bearing and supported volumes. All these aspects are most informative sources and evidence of the history of Russian wooden architecture.

Materials for construction The best and most widely used material for construction was pine. We know that in olden times buildings that were particularly important or needed to be especially durable were made of oak – for example, the first wooden St. Sophia Cathedral in Novgorod, put up in the late tenth century, or the wooden walls and towers of the Moscow Kremlin. The longest-lasting timber is larch, but due to its less extensive distribution in the European part of Russia its use was limited, chiefly to northwestern regions. Larch was actively employed for construction in Siberia. Among coniferous species, the least valuable for building purposes is reckoned to be spruce. It was used mainly for subsidiary service structures. The trees considered most suitable for building were those that had grown slowly on dry, elevated sites. They yield dense, fine-grained timber that is less susceptible to decay. When selecting pine logs preference was given to trunks with a large core. For the lower parts of the building in contact with the damp ground, twisted logs were selected as they possessed the greatest density and resin content. Timber for construction was felled in winter when the sap leaves the wood and it becomes drier and more resinous. If the felled logs are stripped of bark in spring and immediately worked, as the wood dries a natural layer of resin forms on its surface which goes on to protect the material from moisture penetration. For building in Russia, timber was mainly used in the form of logs. The earliest surviving churches are made of logs 20–24  cm in diameter. Preference was given to relatively thin logs, possibly because they are less susceptible to cracking. Usually, though, logs 26–30 cm thick were used in construction. Some

RZ_wooden-architecture.indd 110

29.09.15 21:22

111

Fig. 1: Types and shapes of roof constructions and corner joints in Russian traditional wooden architecture3: 562

seventeenth-century structures were assembled from logs up to 50, even 80 cm across. 1: 13 Buildings made of thick logs, including dwellings, are found in Siberia. Straight-grained logs were split in two or into several parts with the aid of wedges to produce either half-round timber or planks. This process left the surface of the planks uneven and they had to be finished with an axe. Such planks were wide and thick. In contrast to sawn timber, their natural layers remained whole and so they were more resistant to atmospheric influences and served much longer. To produce beams, logs were squared off with an axe. 2: 12–14 The axe was the main carpentry tool. Saws came into widespread use only at a later time. Moreover, the surface of wood worked with a saw becomes more open. An axe, by contrast, cuts away layers of the timber, leaving a smooth surface that absorbs considerably less moisture. To make grooves a different tool – the adze – was used. Openings were cut using a gouge. The earliest axes were

RZ_wooden-architecture.indd 111

29.09.15 21:22

112

short and thick, leaving the wood with a distinctive wavy surface. For finishing work, craftsmen employed a curved scraper or drawknife with two handles. In the eighteenth century axes with broad, thin blades came into use. A surface worked with one of these is smoother. In the nineteenth century the carpenter’s tool kit expanded significantly.

Srub and its roofing The basis of practically any traditional wooden building is the srub, a cell of logs laid horizontally and fitted closely together. There are a fair number of different ways of joining the logs at the corners. The most common in all types of building is known as ‘v chashu’ or ‘v oblo’ s ostatkom, a halved joint with the ends of the logs protruding. 4: 24 This method produces corners that are warm and strong. In the similar joint ‘v lapu’ the ends of the logs are cut off square with the wall, forming a neat, cleanly worked corner. This was usually used when constructing the upper tiers of church buildings and sanctuaries, and also on corners that pointed into the interior of a church. When connecting logs, extensive use was made of hidden lugs fitting into sockets within a joint to prevent movement. Within each wall the narrower ends of the logs were placed alternately in each direction to cancel out any slope. If a log was curved, it would be laid so that the outside of the curve was upwards. To make the logs in adjoining layers fit more closely together, a lengthwise groove would be cut in one of them. This operation was generally performed on the upper log, but it has been observed that in buildings from the sixteenth and seventeenth centuries the opposite was quite often the case. There was also a method known as ‘v rezh’, when the logs were set only a little way into one another, leaving slits. This was used in the construction of civil engineering structures – bridges and embankments, and also for the structural elements of the upper parts of churches that would then be covered by roofs. The dimensions of buildings or the rooms within them were usually determined by the average length of a log, somewhere between 6 and 8 metres. It is believed that in earliest times a building would be increased in size simply by

Fig. 2: The house of the Petunov family in the village of Velikaia Guba, Karelia. The building combines living and service quarters under one roof. Early 20th-century photograph 6: 31

RZ_wooden-architecture.indd 112

29.09.15 21:22

113

placing one srub next to another. In extant dwelling houses we see that when it became necessary to extend logs lengthwise, that was done inconspicuously at the junction with a transverse wall. Often in peasant houses an extension was made by building three walls onto the main srub, with the unconnected ends being reinforced with upright pillars or else set into a groove cut in the logs of the main body of the building. The combination and addition of log cells is often found in various types of peasant dwelling complexes that include both living quarters and premises for keeping animals and farm use. 5: 6–10 A more reliable connection of the various parts of a building was provided by joining the walls of additional sruby into those of the main log frame. To link a wall of one srub to the other, the above-mentioned ‘v chashu’ or ‘v lapu’ joints were used. Placing additional log frames on different sides of the main srub would give a tall building greater stability and create a pleasing architectural composition. Fig. 3: The Church of St. Nicholas (1600) in the village of Panilovo. This church was entirely log-built, including the tent roof and cupola. 1920 photograph 9: 88

Extending logs lengthwise was necessary for the construction of fortress walls or large public or monastic service buildings. The joint was made using a hook carved on the end of a log or a vertical mortise and tenon. These two methods would be alternated in a wall, preventing the lengthwise and crosswise movement of the logs. Additionally, for greater stability long walls were connected with rhythmically spaced cross walls. Sruby usually stood directly on the ground, but boulders were placed under the corners and the middle of the walls. Sometimes short lengths of thick logs were used as foundations, being dug into the soil vertically. 7: 14 The strip foundations of stone or brick that can be seen beneath many old churches were inserted during repairs in the second half of the nineteenth century or the beginning of the twentieth. The simplest way of roofing a srub is a flat ‘deck’ of logs fitted tightly together. Such roofs, covered with birch-bark and turf, have survived down to the present in Yakutia. We occasionally find single-slope roofs on small barns, bathhouses and even some dwelling houses in remote regions, but the most widespread type was the two-sloped pitched roof. Traditional pitched roofs had a substructure of logs. They differ between themselves in height and the number of log purlins supporting the slopes of the roof. The most archaic method was to set the logs into each row of the gable ends. Widely spaced log purlins and framed roofs are typical of nineteenth- and early twentieth-century buildings. Church architecture presents a very wide variety of different roof shapes. Besides simple pitched roofs, the ridged (i.e. structurally linear) variants include tall, wedge-shaped ones with more gently sloping skirts (politsy) at the

RZ_wooden-architecture.indd 113

29.09.15 21:22

114

Fig. 4: The churches of the Luroma Pogost (17th–18th centuries). Early 20th-century photograph 12: 127

bottom, cascade roofs and bochki – literally ‘barrels’ – with an ogival section. The pitched roofs also developed into the more complex crossed (kreshchatye) roofs, which could be plain with (eight) flat slopes or elaborated in the form of two intersecting bochki. 8: 13f The most striking method of topping wooden churches was the shatër or tent roof, examples of which reached a height of 18 metres. Shatry could be either log-built (with gaps between the logs) or have a framed structure. The log type is older. 10: 28, 106 Sometimes the two kinds of construction were combined; the lower part of a tent roof might, for example, be made of logs, while the upper part had a frame. 11; 20–21 Framed shatry invariably had a central pillar, while log ones could be built without. Some forms of church superstructure had more elaborate curved lines. One example is the kub, resembling a ‘four-sided onion dome’. In the seventeenth and eighteenth centuries, roofs of this type were used extensively in the basin of the River Onega. They were made from logs and usually topped with five domes. In the basins of the Vaga and Northern Dvina, many churches between the 1700s and early 1900s had an eight-sided superstructure in the form of a large ‘onion dome’ with a single cupola on top. Simpler roof shapes became the basis for the development of more elaborate compositions. For example, in the basins of the Pinega and Mezen there were five-domed superstructures that combined a tent roof with crossed bochki. There is an example of the amalgamation of a kub with crossed bochki and nine domes. The tops of the many-domed churches of the Lake Onego area take the form of tiered structures incorporating a complex combination of bochki and domes.

RZ_wooden-architecture.indd 114

29.09.15 21:22

115

Fig. 5: St. George’s Church (1493) in the village of Luksovichi stands out for its striking poval – the flare in the upper part of the walls. Photo, 2013

A distinctive feature of old churches and chapels was the poval – an outward flaring of the upper part of the srub. On the earliest buildings it is possible to find steep povaly five to seven logs high. On churches with pitched roofs, apart from rare exceptions, flares were made only on the longitudinal walls. On churches with a centrally planned superstructure there were flares on each side at the top of the srub. With tiered superstructures povaly would be made on each stage. Highly unusual are the seventeenth- and eighteenth-century tent-roofed churches of Karelia that have two or three povaly on their sruby and broaden out towards the top. Stability was achieved in these buildings by fitting the logs tightly together and adding internal horizontal ties. On nineteenth-century churches, the flares were small if they existed at all. Povaly were only made on the main bodies of churches, their sanctuaries, and chapels. Their origin and significance remains unexplained to this day. The roofs of traditional wooden buildings were made of planks. Something considered a sort of classic feature of wooden architecture is the nail-less roof that was used very extensively on service buildings and dwellings as well as churches. 13: 562 It has a two-sloped substructure of logs. Cut into the logs in the lower part of the slopes were trunks of young spruce trees harvested together with a root and cut in such a way as to form a hook. These pieces, known as kuritsy (‘chickens’) on account of their shape, supported wooden gutters in which the lower ends of the roof planks rested. On the ridge of the roof the craftsmen placed a specially trimmed log with a V-shaped groove below – the okhlupen’ that pressed down the upper ends of the planks and covered the seams. The tips of the log purlins protruding from beneath the roof were covered with a bargeboard called a prichelina. The juncture of the two pricheliny was covered

RZ_wooden-architecture.indd 115

29.09.15 21:22

116

Fig. 6: The Dormition Church (1774) in the town of Kondopoga – one of the tallest Russian wooden churches. Photo, 2014

by a vertical board, the polotentse (‘towel’). There are several different variants in the construction of nail-less roofs, as well as local peculiarities. The gutters could, for example, be supported not by kuritsy but by the extended ends of logs in the gable walls. The way the kuritsy and okhlupen were mounted might also vary. Roofs with straight slopes were usually covered with planks. On old buildings the outer surface of each plank was made slightly concave like a trough to make water flow off better. In the 1800s and 1900s, roofing planks were made with two grooves cut along the edges of the surface. Tall, pitched roofs and shatry

RZ_wooden-architecture.indd 116

29.09.15 21:22

117

Fig. 7: Interior of the refectory of the Church of Saints Peter and Paul (1698) in the village of Puchuga. Early 20th-century photograph 15: 101

were covered with several rows of overlapping boards. Shortened boards were known collectively as gont. Domes and bochki were covered with even shorter little boards or shingles, lemekh, laid on like tiles. The lower edges of gont and lemekh were usually embellished with stepped carving. On many churches in the Pomor’e area (the White Sea region) we find shingles with triangular and rounded ends. 14: 13 Covering with gont or lemekh is generally characteristic for curved shapes, although there were churches in the basin of the Northern Dvina and in Vologda region that had tent roofs covered with small shingles.

Elements, details and decorations Floors and ceilings in wooden buildings were fairly simply made. Floors normally consisted of half-logs laid on beams. In small structures (chapels, barns) the floors could be without beams, made of half-logs set directly into the walls during the process of assembling the srub. Very expressive log-ceilings can still be found in some old peasant dwellings and bathhouses. In nineteenth-century buildings they usually made the ceilings from broad thick planks laid overlapping on timber joists. Trapezoidal ceilings supported by two beams are considered archaic. They are observed in the oldest dwellings that were heated by stoves without chimneys. It is supposed that in olden times churches did not have ceilings. There are a few rare examples of two-sloped ceilings that have survived in fifteenth- or sixteenth-century churches. Churches in the 1600s and 1700s were quite often given flat herringbone-pattern ceilings by placing half-logs at an angle between the joists. In the eighteenth century, framed ceilings of a type called nebo (‘heaven’) were widely used in churches. The design of the frame resembles a sun with rays running out from the sides. Nebo

RZ_wooden-architecture.indd 117

29.09.15 21:22

118

ceilings were covered with decorative painting. In church refectories (large rooms added to a church that substantially increased the capacity of the building), which could be up to 10–12 metres in width, the crossbeams supporting the flat ceilings were supported by carved pillars. The treatment of window and door openings is extremely varied. These elements were invested with symbolic significance as links between the outside world and the internal domestic space. The simplest kind of window was a small one cut in two adjoining logs. It was closed on the inside with a sliding shutter. Such windows are recorded in the oldest churches and chapels. Sometimes in primitive hunting lodges and small service buildings, the windows and doors were simply cut through the walls and the edges were not strengthened at all. It was not possible to make large openings in that way, however, as with time the wall could become deformed. To avoid that, the ends of the cut logs were reinforced with posts of squared timber. Door Fig. 8: Porch of a house in the village of Brusenets. Early 20th-century photograph 17: 16

frames made of two jambs, as the simplest way of strength-ening the opening, are found in barns and other ancillary buildings. In the earliest churches, the frames of window and door openings consisted of four pieces of wood. A very interesting and archaic feature was the rounding of the opening by trimming the inner side of the timbers. Buildings of a later period, the 1700s for example, typically have openings framed by three pieces of wood, without the bottom lintel. In the nineteenth century, the framing of openings followed a course of rationalisation and ever greater lightening of structural members. By the twentieth century, they had turned into a casing of thin wooden bars or planks. Window and door openings were not always rectangular. They could have more elaborate shapes. Some seventeenth- and eighteenth-century churches have presented five-sided windows, where instead of a single upper lintel the builders installed two at an angle. Doorways were quite often made with a top made of several pieces of wood imitating an arch. Elaborately shaped door frames in churches were frequently covered with decorative painting or carved in an interpretation of the recessed portals of masonry churches. In the very early days window openings were filled with wooden frames holding sheets of mica. In later buildings of the eighteenth and nineteenth centuries, glazed windows came into use. The first doors were assembled from thick planks fastened together with dowels. Sometimes they were covered on the outside with sheets and bands of metal and painted. Doors like that were heavy and strong. Later, in the 1800s and 1900s, doors were made lighter with panelling.

RZ_wooden-architecture.indd 118

29.09.15 21:22

119

Wooden buildings, while basically made of logs, also contained no small amount of framed elements. The framework of such structures consisted of a bottom rail and frequent uprights linked by beams that support the rafters of the roof. The space between the uprights was filled with boards set into vertical slots. 16: 12 In the early days the lightweight upper storeys of wealthy dwellings had a frame structure. This is clearly seen in an old depiction of the wooden royal palace in Kolomenskoe from the second half of the seventeenth century. An example of a framed superstructure is provided by the second storey of the western annexe (porch) of the seventeenth-century Church of the Resurrection in the village of Kushereka, now in the Malye Korely Museum. Admittedly, that part of the building is not original: It was recreated on the basis of traces in the course of restoration. A framed porch or narthex on a tall log undercroft (lowest above-ground storey) was a feature of the eighteenth-century Church of St. George in the village of Pocha that has now been rebuilt in the Vologda Region Museum of Wooden Architecture. An external gallery or porch running around the main sruby on two or three sides was very common in church architecture. Most often the galleries were supported on cantilever beams – logs projecting from the wall. Examples are also known that have their own separate log base a few tiers high. The closed areas of walls on the galleries were filled in with boards set between the pillars at a slight slant so as to form a herringbone pattern. The windows on the galleries did not have frames containing glass or mica and were usually closed with hinged or sliding shutters. Covered walkways between buildings – a church and a bell tower or two adjoining churches, for example, were constructed in a similar manner to the galleries. Porches also had a framed structure. Churches or more expensive dwellings were usually constructed on top of a high ground floor (podklet’, undercroft) with an ancillary service function. For that reason external steps turned into elaborate structures with several flights and intermediate landings roofed with bochki or shatry. The landings of the porches could be supported by cantilevers or posts; sometimes by a single massive carved pillar. The landings of high porches were quite often placed on separate sruby. The decoration of old timber buildings consisted of the carving of details. In traditional wooden architecture, however, the majority of elements given decorative treatment are part of the structure or have a definite function. The richest embellishment was to be found on the upper part of the building, such as the details of the roof. The kuritsy were carved in the form of birds or animals. The ends of the gutters were given intricate shapes, while the pricheliny (bargeboards) were covered by complex rhythmic carving. The end of the okhlupen’ on the ridge was shaped into a horse’s head (but not on churches), while beneath it hung the carved polotentse. A decorative crest of carved upright boards quite often rose from the okhlupen’, giving the building a more striking silhouette. Carving adorned the brackets that held up the overhangs of the roof. The variety of their decorative shapes was also very great: spindles, scallops, hooks, volutes and much more. The brackets supporting galleries and porches

RZ_wooden-architecture.indd 119

29.09.15 21:22

120

were not usually carved. The porches themselves, though, were very ornate. The supporting pillars were carved and carved boards (podzory) were suspended beneath the overhangs of the roof. The parapets around open galleries and porches on the oldest buildings were made of boards placed tightly together either horizontally or slightly slanted. On later buildings the parapets consisted of vertical boards cut in imitation of balusters or thin pieces of timber. In traditional wooden architecture, all the connections between structural and decorative elements were made using insets, tongues and grooves with a minimal use of metal fastenings. The preparation and working of details took a lot of time, of course, but on the other hand that made it possible to assemble the framed parts of the building quickly. It should be noted that all the techniques and shapes employed in traditional wooden construction are amazingly simple and rational. They are the product of the logic of shape-creation, the characteristics of wood as a building material and centuries of accumulated experience. The attractiveness of wooden buildings, especially churches, was also achieved by carefully judged proportions. That, as a rule, determined the architectural and artistic look of the structure. Some churches are massive and resemble fortress towers; others are slender and graceful. Traditional wooden churches were usually modelled on others constructed earlier. For that reason a composition of shapes might be repeated many times in different buildings. Nevertheless, one cannot find two identical churches – the proportions and details are different in each case. Dwelling houses became particularly richly decorated in the second half of the nineteenth and early twentieth centuries. The gables of the main facades of houses acquired balconies decorated with carving. Elaborate carving full of symbolism appeared on window surrounds. Many decorative elements at that time were no longer parts of the structure. They were made from boards and nailed on afterwards. Their designs combined the traditions of folk architecture with elements of the classical orders and various stylisations. Decorative elements in different regions display their own local traditions, influenced to some degree by the ethnic cultures of the native peoples of Russia.

References 1 Zabello, S., V. Ivanov, P. Maksimov, Russkoe dereviannoe zodchestvo (Russian Wooden Architecture), Moscow: Izd-vo Akademii arkhitektury SSSR, 1942. 2 Krasovskii, M. V., Kurs istorii arkhitektury. Vypusk 1. Dereviannoe zodchestvo (A Course on the History of Architecture. Issue 1. Wooden Architecture), Petrograd: T-vo R. Golike i A. Vilborg, 1916. 3 Brunov, N. I., A. I. Vlasiuk, A. I. Kaplun, A. A. Kiparisova, P. N. Maksimov, A. G. Chiniakov, Istoriia russkoi arkhitektury [The History of Russian Architecture], Moscow: Akademia Arkhitektury SSSR, 1956. 4 Krasovskii, M. V., 1916. 5 Opolovnikov, A. V., Russkoe dereviannoe zodchestvo. Grazhdanskoe zodchestvo

RZ_wooden-architecture.indd 120

29.09.15 21:22

121

(Russian Wooden Architecture. Civil Architecture), Moscow: Iskusstvo, 1983. 6 Zabello, S., et al., 1942. 7 Ibid. 8 Ushakov, Iu.S, ‘Narodnoe dereviannoe zodchestvo’ (Wooden Folk Architecture), in: V. I. Piliavskii, A. A. Tits, Iu. S Ushakov: Istoriia russkoi arkhitektury [The History of Russian Architecture], Leningrad: Stroiizdat, 1984, 7–79. 9 Zabello, S. et al., 1942. 10 Opolovnikov, A. V., Russkoe dereviannoe zodchestvo. Pamiatniki shatrovogo tipa. Pamiatniki kletskogo tipa i malye arkhitekturnye formy. Pamiatniki iarusnogo, kubovatogo i mnogoglavogo tipa (Russian Wooden Architecture. Buildings of the Shatër Type. Buildings of the Klet’ Type and Small Architectural Forms. Buildings of the Tiered, Kub and ManyDomed Type), Moscow: Iskusstvo, 1986. 11 Ushakov, Iu.S, 1984. 12 Zabello, S. et al., 1942. 13 Brunov, N. I. et al., 1956. 14 Krasovskii, M. V., 1916. 15 Zabello, S. et al., 1942. 16 Ushakov Iu.S, 1984. 17 Zabello, S. et al., 1942. All illustrations are by the author except where noted.

RZ_wooden-architecture.indd 121

29.09.15 21:22

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

RZ_wooden-architecture.indd 122

29.09.15 21:22

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-

RZ_wooden-architecture.indd 123

29.09.15 21:22

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

RZ_wooden-architecture.indd 124

29.09.15 21:22

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

RZ_wooden-architecture.indd 125

29.09.15 21:22

126

The traces of a ciborium have been found in situ

The tie of the main building

only in St.  Sophia Cathedral (Fig. 7)   – four round holes in the floor in which wooden pillars originally stood. 13: 88–89 A slab with three holes in it was found in the floor of the sanctuary of St. Nicholas Cathedral. Two of the holes could have been sockets for the ciborium pillars, while one could have been for the altar cross. 14: 271 Probably this slab was used for a second time in the late twelfth century. It is obvious that in the majority of churches the wooden pillars of the ciborium stood on the altar itself or on its base, The tie of the northern gallery Fig. 4: St. Sophia Cathedral in Novgorod. 1045– 50. The ties between the main building and the galleries. Oblique tabled scarf joint. Drawing by Grigorii Shtender

which is why we cannot find traces of them. 15: 80

Lintels of openings, doors and windows It has been possible to find traces of the lintels of door openings in some buildings. The number of beams laid above the openings ranged from one to five. As has already been mentioned, sometimes components of the tie bands were also used for these purposes. One of the five beams in the lintel above the doors in the south facade of St. Nicholas’s Cathedral is part of a wooden tie. The restorers have often found the imprints of door frames, but the method of their installation is not exactly clear. It is possible that the frames were fixed by wooden wedges, and that the lower beam of the frame was placed between limestone slabs. 16: 104 At the same time, in the Transfiguration Cathedral of the Mirozhskii Monastery in Pskov, the upper and lower beams of the frame of western portal were set into the masonry for 1.2 metres, the side parts of the frame were narrower. 17 The same method was used by the builders of St. Paraskeva’s Church in Novgorod. On the threshold of the portal of the south porch, the mark of the beam into which the side parts of the door frame were inserted during the building of the church has been found. A similar beam was obviously placed in the upper part of the opening. 18: 50–51

Fig. 5: St. Nicholas Cathedral in Novgorod. 1113. Two bands of ties in the apses. Drawing by Liudmila Druzhinina

RZ_wooden-architecture.indd 126

29.09.15 21:22

127

Fig. 6: The Transfiguration Church on Nereditsa Hill in Novgorod. 1198. Altar screen. Reconstruction by Vladimir Sarab’ianov

RZ_wooden-architecture.indd 127

29.09.15 21:22

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

RZ_wooden-architecture.indd 128

29.09.15 21:22

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

RZ_wooden-architecture.indd 129

29.09.15 21:22

130

NAME

DATE

St. Sophia Cathedral in Novgorod

1045–1050

St. John the Baptist side chapel

Mid-11th century

Nativity of the Virgin side chapel

12th century

St. Nicholas Cathedral in Novgorod

1113

Church of the Nativity of the Virgin in the St. Anthony Monastery in Novgorod

1117–19, 1122?

St. George’s Cathedral of the Iur’ev Monastery in Novgorod

1119

St. John’s Cathedral in Pskov

1143–44?

Transfiguration Cathedral of the Mirozhskii Monastery in Pskov

1130s – early 1150s

Dormition Church in Staraia Ladoga

1150s–60s?

St. George’s Church in Staraia Ladoga

1150s–60s?

St. Clement’s Church in Staraia Ladoga

1153

Annunciation Church on Miachino Lake (Novgorod)

1179

SS. Peter and Paul Church on Sil’nische (Novgorod)

1185–92

Transfiguration Church on Nereditsa Hill (Novgorod)

1198

St. Paraskeva’s Church in Novgorod

1207

St. Pantaleon’s Church in Novgorod

1207

Church of the Nativity of the Virgin at Peryn (Novgorod)

1230s?

Table 1: Works of Novgorodian architecture of the 11th to mid-13th centuries whose wooden elements have been studied

The best place to understand the idea of the construction of the wooden platform of the upper galleries is the church on Nereditsa Hill (Fig. 9). Here the wooden beams in the side chambers of the catechumena were set in the masonry from west to east, while the beams of the central part were placed north to south. They rested on the beams of the side chambers, with the craftsmen using halved joints to tie them in. 29: 204 We cannot see any traces of a railing in the majority of buildings, making it possible to assume that this was not an obligatory feature for the balconies. Auxiliary wooden structures were also used during the process of building stone churches. Restorers have found traces of scaffolding, the imprints of centerings and supports for them. It is difficult to study the wooden elements in monuments of medieval stone architecture because of the poor preservation of wood and the loss of many elements in fires. Nevertheless, it is obvious that wooden elements were essential features of all early Russian buildings. The churches could not have stood without timber parts. Techniques typical for traditional wooden architecture were also used in masonry architecture. Carpenters took part in the work of ancient Russian building teams as well as masons.

RZ_wooden-architecture.indd 130

29.09.15 21:22

131

Acknowledgements

I am very grateful to Julia Morozova (Novgorod, Russia), curator of the archive of the Novgorod Scientific Restoration Office (NNRU), for her kind help during my study of the archive’s collections, to Dr. Denis Jolshin (St. Petersburg, Russia), who helped to write the part of the text about window frames and to Sergei Lalazarov (St. Petersburg, Russia), who gave me access to his research material. This study has been completed thanks to funding from the Russian Foundation for Humanities, project No. 15-04-00251 (Architecture of Novgorod the Great (11th–15th centuries): building materials and constructive elements).

References 1 Rappoport, P. A., Building the Churches of Kievan Russia, Aldershot: Variorum, 1995. 2 Voinova, I. L., S. V. Lalazarov, Staraia Ladoga, krepost’, tserkov’ Georgiia XII v. Kompleksnye nauchnye issledovanniia [Staraia Ladoga, the Fortress, 12th-century St. George’s Church. Multidisciplinary Researches], Vol. 3, Book 12, Report on the study of the facades (after the removal of the plaster), Archive of the St. Petersburg ‘Spetsproektrestavratsiia’ research institute No. 5320, St. Petersburg, 1993. 3 Bol’shakov, L. N., P. A. Rappoport, ‘Raskopki tserkvi Klimenta v Staroi Ladoge’ [Excavations of St. Clement’s Church in Staraia Ladoga], in: A. N. Kirpičnikov, V. M. Masson, E. N. Nosov (eds), Novoe v arkheologii Severo-zapada SSSR [New Things in the Archaeology of the North-Western USSR], Leningrad: Nauka, 1985, 111–116. 4 Shtender, G. M., ‘Vosstanovlenie Nereditsy’ [The Restoration of the Nereditsa], Novgorodskiii istoricheskii sbornik [Novgorod Historical Anthology], Issue 10, Novgorod: Novgorodskaia Pravda, 1962, 169–205. 5 Voinova, I. L., S. V. Lalazarov, 1993. 6 Shtender, G. M., ‘K voprosu ob arkhitekture malykh form Sofii Novgorodskoi’ [On the Question of the Small-Scale Architectural Elements of St. Sophia in Novgorod], in: V. N. Lazarev, O. I. Podobedova, V. V. Kostochkin (eds), Drevnerusskoe iskusstvo. Khudozhestvennaia kul’tura Novgoroda [Early Russian Art. The Artistic Culture of Novgorod], Moscow: Nauka, 1968, 83–107. 7 Kovaleva, V. M., ‘Altarnye pregrady v trekh novgorodskikh khramakh XII v’, [Altar Screens in Three 12th-Century Novgorodian Churches], in: Drevnerusskoe iskusstvo. Problemy i atributsii [Early Russian Art. Questions and Attributions], Moscow: Iskusstvo, 1977, 55–64. 8 Sarab’ianov, V. D., ‘Novgorodskaia altarnaia pregrada domongol'skogo perioda’ [The Novgorodian Altar Screen of the Pre-Mongol Period], in: A. M. Lidov (ed), Ikonostas. Proiskhozhdenie – Razvitie – Simvolika [Iconostasis: Origins, Evolution, Symbolism], Moscow: Progress-Traditsiia, 2000, 312–359. 9 Chukova, T. A., Altar’ drevnerusskogo khrama kontsa X – pervoi treti XIII v. Osnovnye arkhitekturnye elementy po arkheologicheskim dannym [The Sanctuary of the Early Russian Church of the Late 10th Century – First Third of the 13th Century. The Main Architectural Elements on the Basis of Archaeological Data], St. Petersburg: Peterburgskoe Vostokovedenie, 2004. 10 Sarab’ianov, V. D., 2000. 11 Shtender, G. M., ‘K voprosu o galereiakh Sofii Novgorodskoi (po materialam arkheologicheskogo issledovaniia severo-zapadnoi chasti zdaniia)’ [On the Question of the Galleries of

RZ_wooden-architecture.indd 131

29.09.15 21:22

132

St. Sophia in Novgorod (On the Basis of Archaeological Studies of the Northwestern Part of the Building)], Restavratsiia i issledovaniia pamiatnikov kul'tury [The Restoration and Study of Cultural Monuments], Moscow: Stroiizdat, 1982, 2: 6–27. 12 Sarab’ianov, V. D., 2000. 13 Shtender, G. M., 1968. 14 Bulkin, V. A., ‘Novgorodskoe zodchestvo nachala XII veka po novym arkheologicheskim materialam’ [Novgorodian Construction of the Early 13th Century on the Basis of New Archaeological Materials], in: Drevnerusskoe iskusstvo. Rus' i strany vizantiiskogo mira: XII vek [Early Russian Art. Rus’ and the Countries of the Byzantine World], Moscow: Nauka, 2002, 270–288. 15 Chukova, T. A., 2004. 16 Lalazarov, S. V., ‘Arkhitektura tserkvi sv. Georgiia’ [The Architecture of St. George’s Church], in: V. D. Sarab’ianov (ed): Tserkov’ sv. Georgiia v Staroi Ladoge. Istoriia, arkhitektura, freski … [St. George’s Church in Staraia Ladoga. History, Architecture, Frescoes…], Moscow: Progress-Traditsiia, 2002, 69–124. 17 Nikitin, V., I. Golubeva, ‘Edinstvennyi na Rusi. Istoriia sozdaniia, zhizni i restavratsii SpasoPreobrazhenskogo sobora Mirozhskogo monastyria. Chast’ pervaia. XII vek’ [Unique in Rus’. The History of the Creation, Life and Restoration of the Transfiguration Cathedral of the Mirozhskii Monastery. Part 1: The 12th Century], Pskovskaia guberniia, No. 33 (454), 2–8 September 2009. 18 Shtender, G. M., ‘Nauchnyi otchet ob issledovanii i restavratsionno-konservatsionnykh rabotakh po pamiatniku arkhitektury ts. Paraskevy Piatnitsy v Novgorode 1954–1964’ [A Scholarly Report on the Study and Restoration and Conservation Work on the Architectural Monument the Church of St. Paraskeva Piatnitsa in Novgorod 1954–64], archive of the Novgorod Scientific Restoration Office R-537, Novgorod, 1965. 19 Gladenko, T. V., L. E. Krasnorech’ev, G. M. Shtender, L. M. Shuliak, ‘Arkhitektura Novgoroda v svete poslednikh issledovanii’ [The Architecture of Novgorod in Light of the Latest Research], in: Novgorod. K 1100-letiiu goroda. Sbornik statei [Novgorod. To Mark the City’s 1100th Anniversary. An Anthology], Moscow: Nauka, 1964, 183–263. 20 Nikitin, V., I. Golubeva, 2009. 21 Rappoport, P. A., 1995. 22 Mikhailov, S. P., ‘Issledovanie sobora Ioanna Predtechi v Pskove’ [A Study of the Cathedral of St. John the Baptist in Pskov], Kratkie soobshcheniia Instituta arkheologii AN SSSR [Bulletins of the USSR Academy of Sciences Institute of Archaeology], Moscow: Nauka, 1982, 172: 74–79. 23 Gladenko, T. V. et al., 1964. 24 Rappoport, P. A., 1995. 25 Lalazarov, S. V., ‘Chetvertaia restavratsiia tserkvi sviatogo Georgiia v Staroi Ladoge’ [The Fourth Restoration of St. George’s Church in Staraia Ladoga], in: Khram i kul'tura. Chteniia pamiati Nikolaia Efimovicha Brandenburga (1839–1903). Tezisy dokladov konferentsii [Church (Building) and Culture, Symposium in Memory of Nikolai Efimovich Brandenburg (1839–1903). Synopsis of the Lectures at the Conference], St. Petersburg: St. Petersburg Culture Foundation, 1995, 80–93. 26 Krasnorech’ev, L. E., Sobor Rozhdestva Bogoroditsy Antonieva monastyria 1117–1119 gg. Dnevnik issledovanii 1982–1985 gg. [The Church of the Nativity of the Virgin in the

RZ_wooden-architecture.indd 132

29.09.15 21:22

133

St Anthony Monastery 1117–19. A Journal of Research 1982–85], archive of the Novgorod Scientific Restoration Office R-1601, Novgorod, 1983. 27 Mikhailov, S. P., ‘Pervonachal’noe ubranstvo inter'era sobora Ioannovskogo monastyria vo Pskove’, [The Original Decoration of the Interior of the Cathedral of St. John’s Monastery in Pskov], in: Drevnerusskoe iskusstvo. Khudozhestvennaia kul'tura X – pervoi poloviny XIII v. [Early Russian Art. The Artistic Culture of the 10th – First Half of the 13th Century], Moscow: Nauka, 1988, 95–100. 28 Lalazarov, S. V., 2002. 29 Shtender, G. M., 1962. All illustrations are by the author except where noted.

RZ_wooden-architecture.indd 133

29.09.15 21:22

134

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

Karel Kuča Free-standing wooden bell towers belonging mainly to village churches (not necessarily wooden themselves) are typical for many countries, especially Poland, Germany, the Czech Republic, Slovakia, Hungary, Romania, Ukraine, Belarus, Lithuania, Finland, Sweden, Denmark and the Netherlands, while a few examples have also survived in Norway, France, England and Russia. They are scattered – in varying densities – across the whole northern half of Europe. 1 Within that wide area at least several thousand wooden bell towers exist, representing only a fraction of the bell towers that were built there. It is very interesting to compare similarities and differences between them. This article is focused on fascinating similarities between bell towers in Bohemia (the historical land in the west of the Czech Republic) and Sweden, which are important at a European scale.

Types of wooden bell towers Wooden bell towers usually seem to be quite simple buildings, similar to stone or brick ones: four walls, a base and an upper stage, and a roof. Their shape is often prismatic or polygonal, their stages can be stepped. When we go inside, however, we see a more complicated structure of pillars, beams and braces. There are many types of bell-tower structure, but in general there are only two basic structural approachFig. 1: Rtyně v Podkrkonoší, Úpice district, Bohemia. Columnar type of bell tower; built in 1591–92 (dendro-dating). Plan and cross-section (1983). Photo, 2008. A similar bell tower at Kočí (Chrudim) that is connected to the church front was built in 1499 (d-d) and rebuilt in 1666 (d-d).

es – plus a combination of the two. The first basic type is a post-structure, which consists of four vertical pillars at the corners of a rectangle, sometimes with more pillars between them, and is therefore made up of rectangular frames. In some countries (including Bohemia), diagonal buttresses or braces are usually added to such a core, so these bell towers have a stepped silhouette. The second basic type consists of triangular frames. In this case there is only one line of pillars supported by diagonal braces. This type of bell tower is shown in Fig. 2. The bell tower in Fig. 3 has exactly the same structure but its lower part is not clad. 2 A bell tower is not a building put up for its own sake or merely as part of an ornamental composition with a church. Its function is obvious – it was built mainly as a technical device to allow for the ringing of bells. Ringing is carried out by swinging the bells (an exception to this principle is found in some Orthodox countries, mainly Russia, where the bells are fixed and they are sounded by setting the

RZ_wooden-architecture.indd 134

29.09.15 21:22

135

Fig. 2: Osek, Sobotka district, Bohemia. Plan and cross-section (1983). Photo, 2004

clapper in motion.) So the bell tower must be able not only to bear the static weight of the bells, but also to sustain the dynamic forces connected with ringing them. The simplest and at the same time most efficient solution is represented by the second of the basic types of bell tower described above. It also appears that the first structural type, based on rectangular frames, would have considerably more difficulty in absorbing the dynamic forces. This prompts the question of why the first, rectangular, type still exists. We can probably find many answers to this: Rectangular bell towers are primarily towers. Along with the function of containing a belfry, they were also used as watchtowers for defensive purposes. Additionally they have a symbolic quality. We could discuss this at length, but the most interesting and important question arises when we look at the map. In fact there are only two regions in all of Europe where the second triangular-frame type of large wooden bell tower appears: Bohemia 3, 4 and Sweden. 5, 6 Fig. 3: Lavad, Småland region, Sweden. Photo, 2007

In Bohemia, there are about 100 surviving wooden bell towers in total. In midwestern and northeastern Bohemia only a few examples of triangular-frame bell towers still exist. We know of other bell towers of this type that are now lost. Besides that, there are about 40 bell towers of the same type that are more or less ‘hidden inside’ their outer shell, making the structure hard to recognise from the outside. Finally, there are about 20 bell towers with a mixed structure that combines the post-structure and triangular-frame principles. In Sweden, where about 450 bell towers have survived, there are dozens of bell towers of the triangular-frame structural type that are enclosed. There are also many bell towers of this type with board cladding or the like. This points to a very long tradition of a triangle-frame structure in both Bohemia and Sweden. Beside those areas, there is also a long strip along the northern coast of mainland Europe where bell towers can be found with a triangular-frame structure of similar form, but very simple and reduced in size. They are very common

RZ_wooden-architecture.indd 135

02.10.15 16:54

136

in the northern part of Poland (regions which belonged to Germany before 1945), in the northern states of Germany (Mecklenburg-Western Pomerania and Schleswig-Holstein), in Denmark and the Friesland region of the Netherlands. There are hundreds of such modest belfries that cannot really be called towers. Some are indeed old, but most are very recent, and they have been built constantly, even nowadays. Their function is strictly utilitarian. A larger open triangular bell-tower structure is an exception in these regions. Examples can be found at Hohenaspe, Krummendiek, Seester and Heiligenstedten in the German state of SchleswigFig. 4: Lancken, Ludwigslust-Parchim district, Mecklenburg-Western Pomerania, Germany. Photo, 2007

Holstein, at Wijnjewoude in the Friesland region of the Netherlands, and elsewhere. 7 Importantly, on the northern coast of Europe there are also large wooden bell towers, either free-standing or connected to the front of a church. Their shape is almost prismatic and their structure based on posts. According to research and dendrochronological dating carried out in recent years by Stefan Amt of the University of Hanover, they were first built this way in the fourteenth century and the same or very similar design was used until the eighteenth century, thus making them contemporary with the triangular-frame type of smaller bell tower.

The Protestant aspect of Bohemian and Swedish bell towers A couple of questions arise. First, why can large triangular-framed structures only be found in Sweden and Bohemia (and in smaller form on the northern coast lying between these two countries)? Second, why do bell-carrying buildings with the same structural principles appear in such distant territories as Sweden and Bohemia? There is one aspect to be considered: Bohemia (at the peak era of wooden bell-tower development) and Sweden and the Baltic coast were Protestant lands. Bohemia of course remained mostly Protestant only until the Thirty Years’ War, that is to say, until the middle of the seventeenth century. But that cannot be the answer, or at least not the whole answer. Which group of triangular bell towers is older – the Bohemian or the Swedish? Can we find any mutual influences between them? There is no doubt about historical connections between the northern European coast and Sweden. On the other hand, Bohemia historically had no direct ties with Sweden through constructional traditions, dynasties of architects or the like, as was the case, for example, with Renaissance and Baroque architecture between Bohemia and Italy. There is, however, one important exception: During the Thirty Years’ War (1618–48) Bohemia was occupied by Swedish forces for a long time and the Swedes took a lot of Bohemian art treasures back to Stockholm and other places. This period was indeed the only really significant point of intersection between Bohemian and Swedish history.

RZ_wooden-architecture.indd 136

29.09.15 21:22

137

Returning to Bohemia in detail, we know that a bell tower of the triangular type existed in 1602 in the town in Rakovník – and it has survived down to the present day. It was probably built in the late 1580s. A similar bell tower at Kvílice (Slaný district) was also built in the sixteenth century; others at Osek in 1642/1653 (dendro-dating) and Samšina (Sobotka) around 1650. The Swedes might indeed have encountered this type of bell tower in Bohemia. Besides these four old surviving bell towers, there are only two more – a little one in Neprobylice (Slaný), which was built in 1896 as a copy of an Fig. 5: Rakovník, cemetery church, Rakovník district, Bohemia. Jan Willenberg 1602. Photo, 2008

older structure, and another from 1906–07 in the town of Me ˇlník, designed by the architect Antonín Wiehl, which was inspired by this old architectural tradition. A detailed look at Sweden shows that a triangularframe structure without cladding is typical for the southeastern regions of Småland, Östergötland, Södermanland and Närke. It is generally the only type of bell tower in those parts. There are also several examples in Västergötland and other regions. In the Småland region there are notable bell towers in Alvesta, Angerdshestra, Arby, Askeryd, Bottnaryd (1686), Brahekyrkan (on Visingsö Island in Lake Vättern), Djursdala, Dörby, Drev, Frödinge (1694), Granhult, Hagby, Halltorp, Hemmesjö (old church), Hossmo, Jät (old church), Karlslunda, Karlstorp, Kläckeberga, Kråksmåla, Kvanneberga (chapel), Lannaskede (old church), Lekaryd, Mellby, Moheda, Mortorp, Myresjö (old church), Nävelsjö, Norra Sandsjö, Öjaby, Oskar, Pelarne (1675–1699), Sjösås, Torsås, Tveta (1670s), Vissefjärda, Vallsjö (old church) and Voxtorp. In the Södermanland region examples can be found in Bälinge, Bärbo, Barva, Bergshammar, Bettna, Björnlunda, Blacksta, Floda, Forssa, Halla, Helgesta, HusbyOppunda, Kila, Kungsör (King Charles Church), Lästringe, Lid, Lista, Mörkö, Muskö, Näshulta, Nykyrka, Råby Rekarne, Råby Rönö, Runtuna, Sättersta, Sköldinge, Torö, Vadsbro and Vrena. In the Östergötland region there are bell towers in Åtvidaberg, Gusum, Jonsberg, Kullerstad, Nykyrka, Rönö, Skällvik, Tidersrum, Torpa, Ulrika and Veta; and in the Närke region at Almby, Glanshammar, Kvistbro (1662), Skagershult and Tångeråsa. Several examples can be also found in Västergötland: Brandstorp (1708), Daretorp (1699–1702), Habo (1760), Kungslena (1679) and Velinga (1686). In this region, mainly to the south of Lake Vänern, there are more bell towers of the triangular type but with smaller dimensions: Gillstad (1768), Gökhem, Göteve (seventeenth century), Kälvene (1780–89), Längjum (1804), Lavad (1780), Skörstorp (1700), Slädene (eighteenth century), Södra Lundby (1740), Utvängstorp (1735), Väla (1769) and Vättak (first half of the eighteenth century). All of these have their structure exposed (without cladding). Some have a small boarded bell chamber on the top.

RZ_wooden-architecture.indd 137

29.09.15 21:22

138

We can see that Swedish bell towers of the triangular-frame type date mostly to the seventeenth and eighteenth centuries. Some could be much older. It is hoped that dendrochronology will reveal the true age of many. Even now, though, the bell tower at Söderköping in Östergötland might provide an answer to our questions. It was built by Anders Staplamakare in 1582 – almost exactly the same time as the one in Bohemian Rakovník and many years before the Thirty Years’ War. There is also a small bell tower in Borås in Västergötland that was built in 1617. So, from these examples, we can conclude that the development of bell towers in Sweden and Bohemia was totally independent. In both then-Protestant territories, the same optimal structural type for bell towers was developed during the sixteenth century.

Technical aspects

This does not mean, however, that there could not have been influences later. Up until now we have only been considering ‘clearly visible’ triangular-frame structures. Bell towers based on the triangular structural principle with exter-

Fig. 6: Church of St. Lawrence, Söderköping, Östergötland region, Sweden. Photo, 2014

nal cladding are typical for Bohemia. From the structural point of view they are of the same type as those without cladding. About twenty bell towers of this design have survived in the Czech Republic, including those at Vysoc ˇany (Nový Bydžov, probably from the second half of the seventeenth century), Libušín/Hradište ˇ (Slaný), Želenice (Slaný, sixteenth century; with an eighteenth-century mansard roof), Nabdín (Velvary, with more horizontally elongated proportions, Lhotice (Humpolec, possibly sixteenth century), Rakovník (St. Bartholomew’s Church, 1630), Hor ˇešovice (Slaný, possibly sixteenth century), Skr ˇivan ˇ (Kr ˇivoklát, 1660), Be ˇlec ˇ (Kr ˇivoklát, before 1717), Senomaty (Rakovník), Malín (Kutná Hora), Sků ry (Slaný, dated to 1728 but probably older), Vepr ˇek (Velvary), Obr ˇíství (Me ˇlník), ˇisuty (Slaný, before 1665). Tur ˇany (Slaný, 1666) and R The bell tower at Sudome ˇˇ r, built in 1616, is an important example: It proves that this type of building was being constructed at the same time as unclad ones. The main pillars and braces as well as the little pillars of the enclosed belfry stage are of the same age. It might even be a remnant of an older tradition of column-type structures. It is necessary to note that we can identify the structural type from outside only if the bell tower has the typical six-sided base. If there is a four-sided (rectangular) base (as in Sudome ˇˇ r ), we have to step inside to recognise this type. Many Swedish bell towers of triangular structure and silhouette have their structure concealed by cladding too. So we can continue to find similarities

RZ_wooden-architecture.indd 138

29.09.15 21:22

139

Fig. 7: Sudoměř, Bělá pod Bezdězem district, Bohemia. Plan and cross-section. Photo, 2010

or differences between this subtype of Swedish and Bohemian bell towers as well. It is particularly typical for the southwestern and central parts of Sweden (the Uppland, Dalsland, Västmanland and Västergötland regions). There are some two hundred wooden bell towers of this sort. It is the most common Swedish type and it is not possible to describe all the variants and to name all villages where they stand. The basic structure of these bell towers is usually identifiable from outside, but sometimes not very clearly. About twenty Bohemian bell towers – as already mentioned above – can be considered ‘mixed cases’, i.e. a triangular structure incorporated into a rectangular, columnar one. They represent the triangular and columnar systems comˇáslav), Vršce (Libán bined. Such bell towers have survived in Žleby (C ˇ), Kozojedy (Nový Bydžov, 1689), Kopec Svatého Jana (Hradec Králové, perhaps around 1583), ˇeský Dub, 1671 or earlier), Týn nad Rovenskem (Turnov, 1630), Vlastibor ˇice (C Jabkenice (Mladá Boleslav), Jezbor ˇice (Pardubice, probably from the early seventeenth century, repaired in 1753), Bojanov (Nasavrky, 1730), Krásná Hora (Havlícˇků v Brod), Me ˇbor ˇ, perhaps ˇník (Nový Bydžov, 1689), Horní Studenec (Chote ˇ the second half of the eighteenth century), Ježov (Cechtice, dendro-dated to 1554/55, repaired 1822/23), Všebor ˇice (Ledec ˇ nad Sázavou), Psinice (Libán ˇ, probably seventeenth century), Slatina nad Úpou (c ˇeská Skalice), Veliny (Holice, 1753) and Mlade ˇjov v ˇ c echách (Sobotka, dendro-dated to 1605/06, repaired 1793). Sometimes the triangular bell-carrying part was added later (at Jabkenice), but generally the whole structure was built at once (at Veliny). In Sweden, there are also several bell towers employing a combination of the columnar and triangular-frame structural principles, for example in the Uppland region, at Giresta, Hökhuvud, Roslags-Bro, Österlövsta and Tierp.

RZ_wooden-architecture.indd 139

29.09.15 21:22

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

RZ_wooden-architecture.indd 140

29.09.15 21:22

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

RZ_wooden-architecture.indd 141

29.09.15 21:22

142

Fig. 11: Lučice, Chlumec nad Cidlinou district, Bohemia. Plan and cross-section (1983). Photo, 2008

The principle of a triangular frame can be used also as an ‘additional feature’: Two big diagonal cross braces put into a columnar type of bell tower. The feature is characteristic of several bell towers in northeastern Bohemia: at Luc ˇice (Chlumec nad Cidlinou, from 1718 or earlier), Semín (Pr ˇelouc ˇ, second half of the eighteenth century), Chote ˇborky (Jarome ˇˇ r, probably from the sixteenth century), Zebín (Jic ˇín), Tr ˇebosice (Pardubice, probably after 1754, but part of its structure was dendro-dated to 1539/40; with double diagonal cross braces), ˇ edice (Holice, before 1754, with Michalovice (Ledec ˇ nad Sázavou) and Horní R triple diagonal cross braces). This feature makes the non-ideal structure of this type of bell tower stronger. We do not know if these bell towers were built in this way or if the diagonal cross braces were added later. This type of structure has not been found in Sweden yet either, but we can point to a similar feature in bell towers in, for instance, Naverstad (Bohuslän region) and Roslags-Bro (Uppland region). And thus, there are still many reasons to study relations between Bohemian and Swedish bell towers and probably between the building traditions for this specific type of structure in other European countries as well.

Acknowledgements

All information about bell-tower structures is based on the author’s own field studies of Swedish bell towers in 2007, 2011 and 2014 and an ongoing study of Bohemian ones since 1983.

RZ_wooden-architecture.indd 142

29.09.15 21:22

143

References 1 Kuča, K., J. Langer, Dřevěné kostely a zvonice v Evropě. Svazek 2. Zvonice, Prague: Paseka, 2009. 2 Ibid. 3 Kuča, K., ‘Dereviannaia kul’tovaia arkhitektura na territorii Cheshskoi respubliki XV–XX vekov’ [Wooden Religious Architecture on the Territory of the Czech Republic 15th–20th Centuries], in: Dereviannoe zodchestvo [Wooden Architecture], Issue III, Moscow–St. Petersburg: Kolo, 2013, 29–62. 4 Kuča, K., Dřevěné a polodřevěné kostely, kaple a zvonice České republiky, Prague: Academia, 2015. 5 Kuča, K., J. Langer, 2009. 6 Stockhaus, B., Klockstaplar, Stockholm: Forvännen, 1940, 337–363. 7 Kuča, K., J. Langer, 2009. All illustrations are by the author.

RZ_wooden-architecture.indd 143

29.09.15 21:22

144

RZ_wooden-architecture.indd 144

29.09.15 21:22

145

4. Empirical Study for Preservation and Restoration: Experience and Perspectives 4 . 1 T H E R E S TO R AT I O N O F WO O D E N A RC H I T E CT U R A L M O N U M E N TS I N R U S S I A . C O N T E M P O R A RY M E T H O D S A N D A P P ROAC H E S / Andrei Bode 4 . 2 T H E C H U RC H O F T H E T R A N S F I G U R AT I O N AT K I Z H I P O G OS T – S O M E R E F L E CT I O N S O N T H E B U I L D I N G A N D I TS R E S TO R AT I O N /

Arnt M. Haugen

RZ_wooden-architecture.indd 145

29.09.15 21:22

146

T H E R E S TO R AT I O N O F WO O D E N A RC H I T E CT U R A L M O N U M E N TS I N R U S S I A . C O N T E M P O R A RY M E T H O D S A N D A P P ROAC H E S

Andrei Bode Wooden log-built structures are an original and distinctive part of Russia’s cultural heritage. Mass-scale timber construction predominated over many centuries. That permitted the accumulation of colossal constructional experience and the development of striking architectural forms. Early towns and cities built entirely of timber have long since perished. Due to the short life of the material, a relatively small number of historical wooden buildings remain today. Despite the tremendous losses, however, Russia still has around 130 wooden churches built before the end of the eighteenth century. There are additionally around 100 edifices from the 1800s and early 1900s, bearing the features of classicism, ‘Empire’ and other architectural styles. Besides these, in the regions of the Russian North there are around 300 surviving chapels dating from the eighteenth to early twentieth centuries and countless dwelling houses and service buildings from the late 1800s and early 1900s. The restoration of wooden architectural monuments in Russia presents a colossal field of operation. At the same time a difficult situation has arisen in this sphere. A huge number of buildings are in an untended and abandoned state. There are problems regarding legislation, qualified manpower, technology and other aspects. The quantity of well-restored objects remains small. Here I will focus on questions of the methodology employed in restoring wooden buildings, examine various approaches and analyse planned restoration solutions and their implementation. The evolution of the restoration of wooden architecture in Russia has on the whole been in line with general European tendencies. The first attempts to carry out such work date from the early twentieth century. That was a time that saw the mass-scale reconstruction of ancient wooden churches with a substantial alteration of their appearance. Researchers and architects took a negative view of such refurbishments and saw the value of the buildings as lying in their original appearance. At the same time, early twentieth-century restoration was marked by a fairly relaxed attitude to authenticity. Light is shed on pre-revolutionary restorations in Russia by a special publication by a group of authors led by Alexei Shchenkov, who has a PhD in architecture, entitled Architectural Monuments in Pre-Revolutionary Russia. 1 The theoretical issues attendant on restoration work in Soviet times are revealed in Evgenii Mikhailovskii’s monograph on The Methods of Restoration of Architectural Monuments. 2 The restoration of wooden buildings in that period is presented by Igor’ Shurgin in sections of a summarising work that was also written under the leadership of Alexei Shchenkov in Architectural Monuments in the Soviet Union. 3

RZ_wooden-architecture.indd 146

29.09.15 21:22

147

Restoration projects in the Soviet Union after the Great Patriotic War After the Russian Revolution hardly any restoration of timber buildings was carried out. The large-scale restoration of wooden architectural monuments unfolded only after the Second World War. At that time the predominant aim in such work on both wooden and masonry buildings was to remove later additions, to recreate the original appearance and achieve artistic integrity. The most significant restoration projects of the post-war years were carried out on the Dormition Church in Kondopoga (1774), the Dormition Cathedral in Kem’ (1711–17), the Kizhi Pogost complex (eighteenth to nineteenth centuries), the Church of SS. Peter and Paul on Lychnyi Island (1620; eighteenth century), the Church of the Virgin in the village of Kholm (sixteenth to eighteenth centuries), the Church of the Deposition of the Robe in the village of Borodava (1485), and the Transfiguration Church in the village of Spas-Vëzhi (1713). In the late nineteenth and early twentieth centuries, the majority of those buildings were given wooden cladding, metal roofs and new extensions and porches in place of old ones. The restorers of the 1950s and ’60s inherited the attitudes of the early twentieth-century architects. To return buildings to their original appearance, they resolutely removed later additions and cladding and carried out reconstruction of the wooden roofs. In doing so, they performed a fair amount of non-authentic reconstruction of porches, galleries and decorative details. As a result of restorations of this sort, the buildings acquired an exceptionally expressive and stylistically whole outward appearance. The architect-restorer Aleksandr Opolovnikov established a methodology for the restoration of wooden edifices in accordance with the tendencies prevailing at that time. 4 Leonid Krasnorech’ev, who worked on buildings in the Novgorod region, was a committed follower of Opolovnikov. 5

Fig. 1: The complex of churches in the Kizhi Pogost from the 18th and 19th centuries, restored and returned to its original appearance in the 1950s. The Transfiguration Church is undergoing repeat restoration. Photo, 2014

RZ_wooden-architecture.indd 147

29.09.15 21:22

148

Fig. 2: The 1669 Church of the Ascension from the village of Kushereka, was moved to the ‘Malye Korely’ open-air museum in Arkhangelsk and restored to its original appearance. Photo, 2009 Fig. 3: The Church of St. John the Divine in the village of Bogoslov, restored on a partby-part basis. Photo, 2006

Particularly strong alterations were made during the restoration of architectural monuments that were being moved to open-air museums. The dismantling of a structure opened up extensive opportunities to replace and reconstruct lost parts of a building. The re-sited churches of the Nativity of the Virgin from the village of Perëdki (early sixteenth century), of St. Nicholas from the village of Vysokii Ostrov (eighteenth century), of the Ascension from the village of Kushereka (formerly Kusheretskoe) (1669) and of St. George from the village of Vershina (1672), have a large number of dubious reconstructions made on the basis of marks on the walls or by analogy. On the other hand, the removal of the buildings to museums preserved them from destruction. The reconstructions carried out on the edifices at that time are historically plausible and display excellent artistic taste. It appears that this is really the way things were. The removal of later additions made it possible to uncover and demonstrate the beauty of traditional Russian wooden buildings and make them part of the world’s cultural heritage. One example of a problematic restoration is provided by the Church of St. Barbara in the village of Iandomozero (1650). The building has a complex constructional history. It was extended both upwards and outwards. During the last phase of construction irreversible changes took place: When the walls were clad with timber, the old window openings were enlarged. Opolovikov analysed the constructional history of the building and came to the conclusion that the optimal restoration solution would be a return to the pre-cladding state with all the eighteenth-century extensions. After the cladding was removed, the question of recreating the original windows arose. In lieu of dismantling and

RZ_wooden-architecture.indd 148

29.09.15 21:22

149

reassembly, this was done by means of crude insets. A portion of the windows were left in the later state, but their construction was not in accordance with the unclad logs. Traces of belts of decoration were discovered on the log facades, but those were not restored. 6: 39–55 This example shows the impossibility of totally restoring the appearance of an edifice at any particular stage in its life. Dwellings and service buildings are easier to restore since they are relatively late works without a complex constructional history. Yet they too have undergone numerous changes and losses over their lifetime. The majority of restored objects are to be found in open-air museums. All the dwelling houses in museums have been restored to their original appearance. The restoration of their details and parts is made easier by an abundance of analogies. However, not all buildings in the post-war period were restored to their original appearance. In some edifices that were left in situ, later additions were preserved during restoration. The Church of St. John the Divine in the village of Bogoslov on the River Ishnia (1687) came down to us with a refashioned superstructure and cladding on the walls, a lost gallery and a bell tower that was added at a later date. In Opolovnikov’s project drawings we find various solutions: removal of the bell tower with the reconstruction of the original porch; removal of the cladding without the reconstruction of the gallery but with preservation of the later bell tower. 7: 247–253 The way this object was treated shows us an attitude that contemplated alternatives, a deviation from strict insistence on restoring the original appearance and a search for a more flexible part-by-part approach. The upper part of the Church of St. Nicholas in the village of Nizhnii Pochinok (1717) had been refashioned. Study of the building provided no evidence to picture how its top may originally have looked. During restoration the later shapes of the superstructure and metal roofs were retained on top of the main log srub (frame).

Restoration projects in Russia (1970s–2010s) Restoration projects in recent decades have displayed a greater variety of approaches. On a par with all-of-a-piece restorations, the practice of keeping all later additions has emerged. A number of new projects have been appearing in which opening up of the original structure is combined with the preservation of later parts. In all-of-a-piece restorations, major removals remain quite frequent. For example, in the cases of the Church of John the Baptist in the village of Shirkovo (1694) and the Church of the Prophet Elijah in the village of Samino (1692), to achieve a harmonious appearance the bell towers built on in the nineteenth century were removed. The restoration of the Church of the Nativity of the Virgin and its bell tower in the village of Gimreka (1695) entailed the removal of cladding. Many details and decorative elements were restored. The ensemble undoubtedly lost in authenticity, but it acquired an expressive and integral appearance. This project precisely continued the course laid down by Opolovnikov’s post-war restorations.

RZ_wooden-architecture.indd 149

29.09.15 21:22

150

Fig. 4: The 1665 Church of St. John Chrysostom in the village of Saunino (Kargopol'), restored back to its original appearance. Photo, 2010

A fairly authentic result was produced by the restoration of the Church of St. Demetrius of Salonica in the village of Verkhniaia Uftiuga (1784). This building was not changed much over its history. It escaped cladding and retained many authentic details. The restoration was made with a high degree of confidence. On this project, for the first time in restoration practice, experience was obtained in the revival of historical carpentry techniques. 8 The Church of the Deposition of the Robe from the village of Borodava (1485) was moved to the compound of the Kirillo-Belozerskii Monastery. It is one of the oldest wooden edifices in Russia. It was refashioned many times and has been restored more than once. During the first restoration, the cladding and metal roofs were removed. At the time of the 2009–10 restoration further research was carried out, as a result of which new information about the original appearance came to light. On the basis of traces, various types of nail-less roofs were recreated. It is presumed that the church looked like this in the fifteenth century, however, there are no analogies for such nail-less roofs and details in any known Russian church. It seems to me that reconstructions like this which raise questions would be better left on paper and not implemented on such a unique edifice. A restoration of the original appearance was carried out on St. George’s Church from the village of Semënovskoe (1685) that has been moved to the Kolomenskoe Museum-Preserve in Moscow. During restoration all later additions were removed and the original forms were reinstated. There were sufficient grounds for the restoration of the details of the main srub, but the porch was made on the basis of analogies. All objects moved to museums consistently continue to be restored to their original appearance.

RZ_wooden-architecture.indd 150

29.09.15 21:22

151

Now I shall examine a number of buildings restored with the preservation of a look that took shape at a later stage of construction. The Church of St. George in the village of Iuksovichi (1495) was refashioned in the seventeenth and nineteenth centuries. The restoration envisaged the retention of all the changes and additions, including the cladding. Inside, however, authentic antiquity is preserved: log walls and carved pillars. The only change made during the restoration was the removal of the nineteenth-century porch and the hypothetical reconstruction of the ancient porch. Fig. 5: The 1755 Church of the Prophet Elijah in the village of Tsypino, restored with the retention of all later alterations. Photo, 2012

The complex of the Il’insko-Vodlozersky Pogost comprises a church, a bell tower and an enclosure with a gate. The Church of the Prophet Elijah (1798) has been through at least four stages of construction. During restoration all the later additions were preserved, including the cladding of the walls and the metal covering of the domes. A large portion of the cladding had been lost and was made anew. The large volume of the building covered in cladding looks rigid and doesn’t go with the old log enclosure. A recent example of the historical approach is the restoration of the Church of the Prophet Elijah in the village of Tsypino (1755). The building was heavily reworked in the nineteenth century: the entrance, the top, the cladding of the walls and the metal roofs. In the 1980s a plan of restoration was drawn up to restore its original appearance. In the early 2000s an amendment was made to the plan calling for the combination of the removal of the cladding with reconstruction of the later domed top with metal roofs. At the time of the restoration, however, a new plan was adopted – to precisely preserve the external appearance as it was at start of the twentieth century. While the later appearance of the exterior was preserved, on the inside the log walls and traces of the refashioning of window and door openings were almost always uncovered and the original details of the interior were reinstated. Summing up, as we have seen, a full-blown authentic reconstruction of the appearance is not achieved either through all-of-a-piece restorations or with the preservation of later additions. Finally, I shall examine a special small group of cases in which an uncovering of the exterior is combined with the preservation of a considerable portion of later changes. Such restorations began to appear in the 1980s. The Church of St. Nicholas in the village of Liavlia dates from 1584 and is the oldest surviving tent-roofed church in Russia. The original tall, octagonal frame of logs was surrounded by a gallery. Over three centuries the main srub was reduced in height by 2.5 metres. In the late nineteenth century the gallery was removed, the walls were covered with cladding, the roof was refurbished and a narthex was added. In the course of restoration the ancient srub was freed of cladding, but it was not restored to its original height. The old windows

RZ_wooden-architecture.indd 151

29.09.15 21:22

152

were opened up and traces of the lost gallery were revealed. The original wooden shingle covering of the tent roof, the cupola and the ogival-sectioned bochki roofs of the priruby (side parts) was rebuilt, while the later narthex was retained.9: 24–29 The Church of St. Nicholas in its restored form has practically no questionable reconstruction. This instance demonstrates the uncovering of the ancient parts of a building with a minimum of reconstruction. The Church of St. Nicholas in the village of Soginitsy (1696) was also reworked over the course of its history. The church and bell tower were built at different times. During restoration, the main srub of the church was completely returned to its original appearance, but the later cladding was retained on the bell tower along with the later treatment of the belfry. Different restoration approaches were applied to different parts of the ensemble. The resultant contrast is softened by the fact that the church is separated from the bell tower by an open porch. Another Church of St. Nicholas, in the village of Nënoksa (1762), has also been Fig. 6: The 1655 Church of the Presentation in the Temple and St Michael in the village of Krasnaia Liaga, preserved with late 19th-century reconstructions. Drawing, 2010

through a series of reconstructions. In the late 1990s a project was undertaken containing five alternative approaches to its restoration. Methodologically this was very useful for the adoption of the optimum solution. Each alternative was assessed in terms of the preservation of authenticity and degree of documentary evidence. An option was chosen that included only trustworthy reconstruction with minimal resort to analogies. It called for the uncovering of the srub of the church and the refectory in conjunction with retention of the later narthex. On the main srub, the kokoshniki were recreated on the basis of traces, as were the shingles of the roof. The later narthex was restored with cladding and painting of the walls. Standing next to St. Nicholas Church is the 1834 bell tower restored in the 1990s. In terms of architecture and paint colour, it resembles the narthex of the church. The Trinity Church in Nënoksa (1727) is being restored with preservation of the later additions. The Nënoksa religious complex demonstrates a variety of approaches to restoration. The Church of the Presentation in the Temple in the village of Zaostrov’e (1688) has also come down to our times with significant changes and losses. This is a unique building, imitating the forms of masonry churches. It had an exceptional roof design modelled on the elaborate pozakomarnoe pokrytie that followed the shape of a church’s vaults, but in the nineteenth century this was replaced by a simple four-slope roof. The bochki over the priruby were also replaced by pitched roofs. There is insufficient information for a full reconstruction of the original appearance of the edifice. The walls bear indeci-

RZ_wooden-architecture.indd 152

29.09.15 21:22

153

pherable traces of a gallery and a walkway to another church that has long since been lost. A further consideration is that the later four-sloped roof is practical for the reliable preservation of the building. The restoration plan proposed a restoration by parts. The late metal roof is combined with partially revealed zakomary (rounded gables usually associated with vaults). Organised guttering is being planned. The roofs of the sanctuary and narthex are being reconstructed on the basis of traces. On the walls, the original windows can be found alongside later ones. A modern, simply shaped porch is proposed. The restoration plan for the Church of the Presentation represents a development of the part-bypart approach. The present-day state of expertise Fig. 7: The 1584 Church of St. Nicholas in the village of Liavlia, restored part-by-part. Photo, 2010

accumulated in the twentieth century is summed up in my own recent article. 10 The main idea on which attention is focused is the impossibility of recreating the original appearance completely and authentically and the advisability of the method of restoration by parts. This approach is very rarely taken with wooden architectural monuments, while in other forms of restoration and in international practice it has long been firmly established. Precisely this approach makes it possible to avoid unjustified removals and dubious reconstructions, while at the same time retaining most fully the elements of the building in their authentic state and revealing its historical and artistic biography. And thus, a variety of methods are employed in Russia for the restoration of wooden architectural monuments. The dominant trend is to completely remove cladding and restore the original appearance. As we have seen, however, the results are varied. A building’s authenticity is diminished by too many removals and rebuilt parts. The shortcoming of restorations that preserve all the later elements is a failure to reveal the building’s history and the early stages of construction, which are the most interesting from an architectural and artistic point of view. Timber buildings are easier to reconstruct than masonry ones. The wooden architectural monuments that have survived are made up of elements from different times and have many additions of varying quality, as well as losses, including those that can no longer be made good with reliable authenticity. The artistic integrity and stylistic unity of a building undergoing restoration is usually achieved at the expense of its authenticity. On the other hand, results that are interesting and judicious with regard to the preservation of a building are provided by an analytical or differentiated approach, when the appearance of different periods is preserved in various parts of the building, as we see, for example, in the churches in Soginitsy, Nënoksa and Zaostrov’e. A fairly large amount of uncovering on the main, as a rule oldest, volume makes it possible to present the artistic integrity of traditional

RZ_wooden-architecture.indd 153

29.09.15 21:22

154

Fig. 8: The 1688 Church of the Presentation in the Temple in the village of Zaostrov’e. The restoration plan called for the uncovering of the oldest parts of the church while preserving the later alterations. See also: Fig. 11 on p. 16. Drawings, 2013

wooden architecture. The retention of later roofs, extensions and porches only underlines the antiquity of the main body and does not produce the impression of dubious reconstructions that is present in many all-of-a-piece restorations. The best way to preserve the authenticity of an object undergoing restoration is to minimise reconstruction and removals. In the majority of works of wooden architecture, the optimum balance between reconstruction and removal can only be achieved with an analytical approach, differentiated for each differently dated part of the building. This makes it possible to reveal the history of the building to the greatest extent, while at the same time preserving its authentic later parts and details. However, despite the promising and attractive qualities of the differentiated method, it cannot be considered exclusively preferable. Each building has its own unique history and, furthermore, the situation regarding preservation is different in each case. This determines an individual approach for each building and the particular method of restoration that is optimal for its preservation.

RZ_wooden-architecture.indd 154

29.09.15 21:22

155

References: 1 Shchenkov, A. S. (gen. ed.), Pamiatniki arkhitektury v dorevoliutsionnoi Rossii. Ocherki istorii arkhitekturnoi restavratsii [Architectural Monuments in Pre-Revolutionary Russia. Outlines of the History of Architectural Restoration], Moscow: TERRA-Knizhnyi klub, 2002. 2 Mikhailovskii, E. V., Restavratsiia pamiatnikov arkhitektury (razvitie tvorcheskikh kontseptsii) [The Methods of Restoration of Architectural Monuments (The Development of Creative Conceptions)], Moscow: Izd-vo literatury po stroitelstuvu, 1971. 3 Shurgin, I. N., ‘Restavratsiia pamiatnikov dereviannoi arkhitektury v Rossii’ [The Restoration of Wooden Architectural Monuments in Russia], in: A. S. Shchenkov (gen. ed.), Pamiatniki arkhitektury v Sovetskom Soiuze. Ocherki istorii arkhitekturnoi restavratsii [Architectural Monuments in the Soviet Union. Outlines of the History of Architectural Restoration], Moscow: Pamiatniki istoricheskoi mysli, 2004, 440–452; 542–549. 4 Opolovnikov, A. V., Restavratsiia pamiatnikov narodnogo zodchestva [The Restoration of Works of Folk Architecture] Moscow: Stroyizdat, 1974. 5 Krasnorech’ev, L. E., Restavratsiia i issledovanie pamiatnikov dereviannogo zodchestva. Po opytu rabot v Novgorodskoi oblasti [The Restoration and Study of Wooden Architectural Monuments. On the Experience of Projects in the Novgorod Region], St. Petersburg: Dmitrii Bulanin, 1999. 6 Opolovnikov, A. V., Russkoe dereviannoe zodchestvo. Pamiatniki shatrovogo tipa. Pamiatniki kletskogo tipa i malye arkhitekturnye formy. Pamiatniki iarusnogo, kubovatogo i mnogoglavogo tipa [Russian Wooden Architecture. Buildings of the Tent-Roof Type. Buildings of the Klet’ Type and Small Architectural Forms. Buildings of the Tiered, Kub and Multi-Domed Type], Moscow: Iskusstvo, 1986. 7 Ibid. 8 Popov, A. V., Shurgin, I. N., O vossozdanii russkoi plotnichnoi tekhnologii XVII–XVIII vekov [On the Reconstruction of Russian Carpentry Techniques of the 17th–18th Centuries], Moscow: 1993. 9 Opolovnikov, A. V., 1986. 10 Bode, A. B., ‘K voprosu o metodike restavratsii pamiatnikov dereviannogo zodchestva’ [On the Question of the Methodology of the Restoration of Wooden Architectural Monuments], Restavratsiia i issledovanie pamiatnikov kul'tury [The Restoration and Study of Cultural Monuments], Moscow-St. Petersburg, 2013, 6, 309–326. All illustrations are by the author.

RZ_wooden-architecture.indd 155

29.09.15 21:22

156

T H E C H U RC H O F T H E T R A N S F I G U R AT I O N AT K I Z H I P O G OS T – S O M E R E F L E CT I O N S O N T H E B U I L D I N G A N D I TS R E S TO R AT I O N

Arnt M. Haugen The summer church at Kizhi Island The Kizhi Pogost in Karelia is one of few remaining northern Russian wooden church ensembles, consisting of three buildings: a summer church, a winter church and a bell tower. To my mind there is no building comparable to the summer church at Kizhi – the Church of the Transfiguration of Our Saviour. According to Alexander and Yelena Opolovnikov, this church is the ‘sole survivor of all the multi-domed cathedrals built during the reign of Peter the Great’ (Fig. 1). 1 The building is a construction of pine logs topped with twenty-two onionshaped domes covered with aspen shingles. The top of the upper dome reaches 37 metres above the ground and the cross on top adds a further 5 metres. It has been suggested that the design could have been drawn up by Peter the Great himself and, according to the legend, the master builder, Nestor, flung his axe into the lake when the work was finished, exclaiming that ‘The world has never before seen, and never again will see, anything like it!’2 It is not the size itself Fig. 1: The Church of the Transfiguration, Kizhi, February 2011. The bottom tier has been removed for restoration, and the building rests on a lifting system based on simple jacks.

that makes this edifice so special; several other log churches in northern Russia are taller, and there are certainly bigger log buildings too. 3 Still, this is without doubt among the most spectacular and complicated log buildings ever built. The construction of the church is well thought out and executed; although it must be right on the very limit of what is possible to make with a simple log construction technique. The building is taller than it is wide, and its huge weight, estimated to be more than 600 tons, 4 could be enough to crush the bottom logs if the distribution of the pressure were uneven. When the wind sweeps in from Lake Onego or the flat terrain around, there is no shelter for the tall church and the wind forces must exert significant pressure on the building, moving load forces around the huge timber construction. The twenty-two domes covered with aspen shingles make the church look spectacular, but they also represent a danger to the durability of the building as they collect moisture from rainwater and dew, leading to vulnerable spots (Fig. 2). Furthermore, the spectacular shape reduces the effect of sun and wind so that the walls will not dry as quickly as on a plain log building. Altogether this church has a lot of challenges that in theory should make it vulnerable to weather and time, and if it were built like this today we would not expect it to last very long. Still the Church of

RZ_wooden-architecture.indd 156

29.09.15 21:22

157

the Transfiguration has been standing on its site for more than 300 years now. Even though it has had a supportive structure for the past few decades, it stayed up unassisted for more than a quarter of a millennium. This paper reflects on some selected properties of this particular building and its construction, but it also has the objective of discussing issues relating to log building construction and restoration in general, which may also be relevant to other buildings and restoration projects. It represents a view from an outside perspective and is based on general knowledge and experience of log construction, not of Russian wooden architecture in particular.

The building history The Church of the Transfiguration was built in 1714 to replace a predecessor on the same site that had burnt down. The new church was built on a simple foundation consisting of loose rocks and lime mortar and it measures approximately 20 × 29 metres at ground level. 5 The main volume of the church Fig. 2: The 22 domes are covered with aspen shingles. Fig. 3: Sketch of the church plan, based on old drawings and photos. The thick lines illustrate log walls; the thin lines imply interconnecting logs below floor level. The sketch may contain inaccuracies.

has an octagonal shape, but four of the walls have smaller annexes attached to them, giving the plan of the church the shape of a cross (Fig. 3). Three of the annexes are square, the fourth, containing the sanctuary, 6 has five exterior walls – the three shorter ones at the east end forming an apse. In addition to this main plan, the church has a quite large refectory annexe on the west side, enclosing the lower part of the building. 7 About halfway to the top of the church, the walls of the main octagonal volume end. A smaller octagonal volume is set on top of this, on a bridge of logs laid crosswise. Yet another even smaller octagonal volume is set on this again, and above that comes the cylindrical ‘neck’ of the large onion-shaped dome that forms the pinnacle of the great church. The other twenty-one domes are placed on barrel-shaped roofs, seventeen of them above the annexes, lining up with the four cardinal directions, thus following the main pattern of the church’s plan. The final four domes are inserted between the others, each on top of one of the four remaining walls. All the roofs are covered with birch bark and then shingled with aspen. 8 In the 1950s, before the church was restored by the architect Alexander Opolovnikov, it had sheet metal roofs and boarding on the walls. (Fig. 4) Those elements were removed by Opolovnikov to recreate what was perceived to be the original look

RZ_wooden-architecture.indd 157

29.09.15 21:22

158

Fig. 4: The Church of the Transfiguration (1912), before Opolovnikov's restoration. (photo: Kizhi State Open Air Museum of History, Architecture and Ethnography)

of the monument. This was still quite early in the history of log-building restoration and good methods of changing the timber in this kind of building had not yet been developed. Attempts to change timber were made, but since there was no means of replacing complete lengths of logs, or even parts of logs in their full diameter, most of the repairs only fulfilled aesthetic purposes and had no structural significance. Because of this, the construction suffered increasing deterioration over the following years. In 1982 all the interior was removed to storage and a heavy metal frame was installed inside to support the building and secure it from falling over. Since its inclusion on the UNESCO list of World Heritage Sites in 1990, the approach to the repair of the summer church has been thoroughly discussed. In 2012 the first tier of logs in the church walls was completely restored and put back on the foundation.9

RZ_wooden-architecture.indd 158

29.09.15 21:22

159

Fig. 5: The round notch. Fig. 6: Example of a lap joint with a dovetail.

The construction As was already stated, the Church of the Transfiguration seems to be a daring piece of work, even for a master builder. Still, after all these years we must recognise the fact that this building is very well designed, and that the planner and the master builder must have been extremely skilful. One of the factors crucial for the success of this project was the layout of the construction. The ground plan of the summer church is quite complicated for a log building, but it is also in a way simple and logical. In the lowest tier, below floor level, some logs also run across the octagonal base of the structure connecting the walls, and holding the bottom logs in position. Above the ceiling a similar arrangement of intersecting logs is used. The design leaves relatively short lengths between joints; most logs only run freely for between 3 and 6 metres before they are fixed by a joint. This is good practice since a log will become increasingly susceptible to bending forces when it exceeds a length of 5–6 metres. It also provides very practical lengths for the carpenters to deal with as long logs are heavy to move and short logs do not lie still while being worked with an axe. The layout of the logs also results in some triangular structures in the lower part of the construction that play a role in the stability of the whole bottom tier. Another important fact is that the weight is distributed across many joints, because even though the walls of a log building do to some extent act as a solid slab, the logs are not completely rigid. The joints are a stabilising factor and the huge forces from the weight are taken down to the ground by the vertical ‘stacks’ of joints, in principle performing a function similar to columns. Another significant factor is the shape of the joints. There are two main types of notches in this building. One is the so-called round notch (Norwegian: vagenov, Fig. 5), where half a cylinder is cut away across one log, so that the transverse log can rest in that socket. 10 The other type is a kind of lap joint with a half dovetail in different variations (Fig. 6). The round notch has protruding ends, the lap joint does not, and this is likely to be the factor that determined where the different types of joint were used. Both serve to lock the construction, and they are quite solid and stable. What is more important in this case, though, is that these kinds of joints distribute the weight in a convenient way,

RZ_wooden-architecture.indd 159

29.09.15 21:22

160

so that the pressure from above is evenly spread without damaging the logs. (Notches with tapered sides, for example, would be less suitable to carry heavy weight and can actually make the logs crack.) The shape of the layout and the notches were very important choices in making this building constructionally sound. The round notch is very rare in Norway and, with very few exceptions, cannot be seen in extant buildings there. It is almost exclusively found during archaeological excavations, often in foundations, and most of the finds date back to 900–1200. One theory is that the round notch soon evolved into a more complicated joint with a trapezoidal section. The reason for this is believed to be that the round notch collects water due to its shape, which is also supposedly why there are no surviving buildings containing it. 11 The lack of buildings with this kind of joint in Norway provides no empirical proof for this theory. If it really is true it should at least be Fig. 7: The notches are very tight and carefully fitted.

possible to see the effect on the Church of the Transfiguration. This building has round notches in very weather-exposed positions, and it has been standing on a narrow part of Kizhi Island, close to the lake, for more than three centuries. It has undoubtedly been regularly exposed to high humidity and rainy weather throughout those 300 years. It should thus be possible to establish whether damage caused to a very high degree by rot and other moisture-loving organisms is related to the areas around the round notches. But this does not seem to be the case. The parts of the church that have so far been repaired give the impression that the damage is mainly located where the roofs meet a wall and other similar situations. These are well-known weak spots in any wooden building, and it is very interesting that the round notch does not seem to make a difference. In other words, damage is mainly situated where it is common in any log house, regardless of the type of joints. In the Church of the Transfiguration the notches are very carefully fitted, which might also be a factor to take into consideration. (Fig. 7) Anyway, this is an important observation that perhaps should make us look differently at how log-building techniques evolved in different parts of Europe. Existing theories are mainly based on speculation and are not supported by unquestionable, empirical facts. A new study suggests that the round notch disappeared in Norway due to practical reasons: The round notch requires timber with very little taper, which is rarer in Norway than further east. 12

RZ_wooden-architecture.indd 160

29.09.15 21:22

161

The restoration During a restoration project there is always a variety of choices and decisions to be made. The questions to be considered can be divided into two main areas. First, the philosophy of the restoration – drawing a line between what to conserve and what to restore. Second, one has to choose the practical methods of restoration and conservation. The philosophy must be decided ahead of the restoration, while the methods will often be adjusted during the working process due to new insights into the issues and the object. Though these are two separate questions, they are also closely connected and may be regarded simply as a theoretical and a practical approach to the same issue. First and foremost, the appropriate level of restoration must be established. Should everything be repaired as is, with as little change of materials and construction as possible, aiming to conserve the present situation? Should the repairs be made with strict reference to one point in time, or should they also preserve the traces of the building’s history? To determine this we need to decide what we want to achieve, like which ‘truth’ we want to preserve, and what story we want the object to tell. Traditionally, when building conservation was in its infancy, it was an important goal to get back to a presumed original state and even to wipe out ‘incorrect’ modifications made over the course of time. 13 In recent decades this has gradually changed and now it is often seen as equally important to preserve all aspects of an object’s biography. Traces of change are appreciated as testimony to a building's history and it might be those very details that make it important for us. If the original fabric itself is considered to be the most valuable, the repairs will be done to retain as much of it as possible. The value of original elements is determined by assessments of age, rarity, craftsmanship and other factors. In other cases the structure of the building is considered more important; this might be motivated by a wish to preserve technical aspects, but the goal might also be to keep the appearance of the building mostly unchanged. The practical approach to the restoration of log buildings can be divided into two main strategies: 1. Complete dismantling of the building, with repair work being done in a suitable place before reassembly. 2. Restoration piece by piece without complete dismantling, while the building remains on its site. Complete dismantling has some clear advantages, such as the opportunity to do all the work in a controlled environment. It also gives the restorers the necessary space in which to work and the practical process will be almost the same as when the building was put up in the first place. The challenge with this approach is that it demands very thorough preparation and, even with the most accurate documentation and measurement, it is extremely hard to get every log back precisely where it was. Every bit of the building must be thoroughly measured, documented and marked to make it possible to carry out an accurate repair and an exact reassembly of a completed restoration. However, when the

RZ_wooden-architecture.indd 161

29.09.15 21:22

162

building is totally dismantled there is no physical reference left. The only reference is the documentation, which is in fact secondary information and can never be guaranteed to be entirely accurate, as all documentation will be an interpretation of sorts. Even if it were precise, it would never be possible to prove it. This uncertainty might lead to errors which will accumulate and the challenge will increase with the size of the building. Finally, complete dismantling actually removes the monument from its site and even when every precaution is taken there will always be a risk that it will never be re-erected, due to such things as human error or natural disasters. Total removal also has a symbolic aspect when it comes to a monument of some importance, even if temporary. Restoring a log building without complete dismantling makes it easier to avoid problems resulting from a lack of reference. Since the building is still there, it will always be possible to measure reference points and to see how in fact different parts of the building are made, if in doubt. The challenge of this approach is the opposite one: How working space should be created. It is in theory possible to split a whole, damaged log and remove it from the wall without causing any major problems. It is not, however, possible to put a log of similar size back in without a minimum of extra space between the adjacent logs. This means in practice that to make larger repairs without dismantling the building, it has to be either lifted above the area being repaired or lowered below that point. In either case the upper part of the walls has to rest on some kind of construction. For the restoration of the Church of the Transfiguration on Kizhi Island, a specially designed lifting system has been created, consisting of a metal construction and jacks, making it possible to split the building horizontally at any level needed. The building is divided into seven tiers consisting of several layers of logs. The tiers are dismantled, one at a time, then taken to a workshop where a full restoration is performed on that part of the building. Even though this restoration is carried out while the building is still on the site, it is really a combination of the two approaches discussed above and the method has advantages inherent in both of them. It permits dismantling to an extent that provides a comfortable workspace, and the dismantled parts can be brought to a workshop and repaired under controlled conditions. Still, only a limited fraction of the building is dismantled at any one time. The repairs can be done in the same way as if the object were fully dismantled, yet the building as a whole never leaves its original site. Supplementary documentation of the standing building can be gathered at any time during the process and the symbolic importance of the monument is maintained during the whole restoration. Since this is also a World Heritage Site, complete removal of the monument would undoubtedly lead to serious issues over the UNESCO status. It would be very difficult to find a good argument for keeping a removed monument on the list. An important issue is the method of log repair. Among other factors this depends on the requirement to keep original material, and the extent to which the log still performs its function as a structural component. In a log building all the logs work together to a large extent. This means that every single log is do-

RZ_wooden-architecture.indd 162

29.09.15 21:22

163

ing part of the job and so each one should be treated as a load-bearing part and repaired accordingly. If one log is not performing its function it might affect the balance of the construction, overloading some parts, and that could be the start of a chain of deformations, followed by damage. In the Kizhi project, log repair methods have been thoroughly discussed and the methods employed are based, among other things, on international acknowledged principles, as well as guidelines given directly by ICOMOS. 14 The most important statement of these principles was that any repair that involves the load-bearing part of a log must be carried out in such a way that the log can still take the same load as it originally did. This might mean that some logs need to be completely exchanged for new ones instead of being patched. It also means that damage to the surface of a log need not necessarily be repaired at all, as long as it is outside the load-bearing part of the timber. In unfortunate cases, cosmetic patching can act like a lid, keeping the underlying wood from drying and thus leading to more rot later. The biggest challenge in this restoration project is the huge size and the weight of the building. This is also to some extent decisive for the restoration methods. In this project it is crucial that the properties of the structure are kept and restored, probably more crucial than for any other log building restoration. A failure in this building might have major consequences and there is no room for mistakes.

Settling, shrinkage and movement of the building There are two main reasons for shrinkage in log buildings. First, timber will always shrink to some extent because of drying; second, a log building will always develop small deformations between the logs. This is actually an important mechanism that makes a log wall tight, distributes pressure and creates friction between the layers of logs, which in turn makes the building more stable. In a restoration project like the one on Kizhi Island, settling and shrinkage is always a big challenge because it is hard to predict accurately. Also, the need to exchange timber is usually unevenly distributed around the building. To avoid adverse settling of the Transfiguration Church, several precautions are being taken. The fundamental factor is the quality of the craftsmen. The Kizhi project has highly skilled craftsmen making precise adaptations of new logs, inlays and other kinds of repair. This is crucial to minimise the level of shrinkage because a precisely matched log will have a larger contact area for distribution of the pressure and thus less deformation. As mentioned above, the shape of the notches also distributes pressure well, reducing deformation. In addition, the quality of timber plays an important role regarding shrinkage. The heartwood of pine timber has more resin and tar compounds in it and is less affected by shrinkage than the sapwood. Logs with a large content of heartwood will therefore be less susceptible to shrinkage. For the Church of the Transfiguration a great effort has been made to pick trees with the best possible qualities. 15 The project has also included creating a system to simulate the load of the building on the restored tiers in the restoration hall. This is done by

RZ_wooden-architecture.indd 163

29.09.15 21:22

164

using jacks in combination with specially adapted devices spanning vertically over the walls. Exposing the wall to such loads will also initiate deformation of the new replacement logs and adapt them to the right positions. This will accelerate the shrinkage and deformation process and will also make it easier to perform adjustments during re-erection if needed.

The question of structural reinforcement Another major challenge in the project is the question of reinforcement. Several places in the building have obvious weaknesses, yet it has survived for 300 years. The question is when to trust the old construction and when to improve it. Sometimes what is seen as a technical improvement from a modern point of view might have a negative effect on the old structure. Any rigid element mounted on a log wall might cause trouble, since one of the main advantages of a log construction is its flexibility. Today we have advanced methods for calculating the forces and capacities of wooden construction, i.e. tables and mathematical formulas based on years of experience. Still, it is not possible to make an entirely accurate calculation, and even the best of engineers will have to leave a safety margin on top of a calculation to be sure. Since wood is not homogeneous, it is not even possible to determine the characteristics of one single log precisely. Hence, the capacity of old buildings will almost always be underestimated. Furthermore, in the past, house timber was often chosen by completely different criteria than are used today. Because of this it is not possible to make an accurate calculation for a log building and any calculation will never be more than an educated estimate. Old buildings are often better than we believe because the quality of the timber is far better than we expect. 16 As a result of this, empirical facts must always be given considerable importance when the capacity of an old timber structure is assessed. The only things that can be proven to some extent are whether a structure has survived, and what degree of damage it has sustained. As of July 2014 approximately thirty layers of logs from the summer church had been restored and reassembled at the building site. The walls had all been repaired and test assembled as close to the original positions from 1714 as possible, and had been exposed to loads to make them settle like this. Despite all the precautions taken, controlling the movement of the building proved a challenge. In the bottom tiers, pegs were inserted between the layers of logs to keep the walls in position. Movement caused by settling made it necessary to adjust the walls after reassembly and the pegs made this difficult. Because of this, the use of pegs was abandoned. The tendency is for the building to move back towards the deformations accumulated through the centuries, in the shape it was before the restoration started, and making the new logs adapt to this. The challenge is how to get the church to stay in its original position from 1714, which is the goal of the project. To stop this movement it is crucial that the building is held in the desired position for as long as it takes to stabilise the building. Held in the right positions, the deformed logs will acquire new deformations, the same way as hap-

RZ_wooden-architecture.indd 164

29.09.15 21:22

165

Fig. 8: Binding posts.

pens in a new building. The biggest challenge is how to achieve this. The logical way to keep all the horizontal logs in position is to support them with vertical beams. This could be done with a free-standing construction on the outside of the church that is not attached to the building, but supporting every log from the outside at two points. A construction like that would have to be temporary and the method requires that the building be able to stand by itself when the construction is removed. There are grounds to believe, though, that when the main octagonal volume is re-assembled and the tie-beams in the upper part are put back, the building should be able to stay up by itself. After all, sitting on a poor foundation with severe rot damage it has survived 300 years. It would be very strange if it could not stay up in a repaired state. Another option is to use ‘binding posts’ (Fig. 8). This is a common method of permanent reinforcement on log walls of a certain size in Norway and it is frequently used on churches. Many Russian timber churches also have such binding posts; the question is whether their primary purpose is reinforcement, or are they there to carry the cladding? Another question is whether the binding posts on the Church of the Transfiguration should be considered an added element or part of the construction.

Conclusions

The huge weight and complicated construction of the building make this restoration project very challenging and it puts a great onus on the solutions and choices made. There is no room for mistakes. For this reason, it is also not possible to choose to patch up or keep rot-damaged material. The physical requirements of the building determine the level of repair to a large extent. Yet there is still a possibility of taking different approaches in some respects.

RZ_wooden-architecture.indd 165

29.09.15 21:23

166

What makes the Church of the Transfiguration particularly interesting as a case study in restoration is that practically all existing challenges regarding the restoration of log buildings can be found in this very project. While there is no other project comparable to the restoration of the summer church, the project itself will be relevant to almost any other log building restoration. Thus, the experience acquired here is extremely useful and it will be very valuable to utilise it with other projects and to develop the techniques and the knowledge of restoration.

Acknowledgements

I am very grateful to the Kizhi State Open Air Museum of History, Architecture and Ethnography, and to the people working with the restoration project for their great hospitality and for giving me the opportunity to gain information about this spectacular project. I am particularly thankful to Alexander Ljubimtsev and Olga Bukchina from the museum, and Vitaliy Skopin from the contracting company Zaonezhie for providing me with detailed information for this article. I also wish to acknowledge the help provided by my dear colleagues Leif Anker and Dorina Sylvia Dobnig, in giving me useful input, and Anders Amlo for preparing the photos and illustrations. Special thanks to Andrew Powter, who has been my mentor during my visits to Kizhi.

References 1 Opolovnikov, A. V., Opolovnikova, Y. V., The Wooden Architecture of Russia. Houses, Fortifications, Churches, London: Thames and Hudson, 1989, 168. 2 Ibid. 3 For example, the Dormition Church in Kondopoga, on the western shore of Lake Onego, is 42 metres high; 4 Bukchina, O., Ljubimtsev, A., Skopin, V., 2015: oral information. 5 The Church of the Transfiguration of Our Saviour (8 April 2015), retrieved from: http://kizhi.karelia.ru/architecture/en/russkie-zaonezhya/tserkov-preobrazheniya-gospodnya 6 In Russia the word ‘altar’ is used for this part of the church. Buchina, O., 2015: oral information; the word ‘sanctuary’ is used in this paper as is common in English. 7 The refectory was, and still is, used as a meeting place for parishioners. Before the Church reorganisation in Russia in 1650–60, the refectories were used as dining rooms, where the Holiday after the Church service could be celebrated. This tradition was prohibited in churches by the Patriarch Nikon, but in all the monasteries it is still used for the old purpose. The name ‘refectory’ is still used for this part of the church. Buchina, O., 2015: oral information; Nikon (1605–1681) was Patriarch of Moscow and all the Rus’ from 1652 to 1658. Retrieved from: http://global.britannica.com/biography/Nikon (22 June 2015). 8 The Church of the Transfiguration of Our Saviour (8 April 2015). Retrieved from: http://kizhi.karelia.ru/architecture/en/russkie-zaonezhya/tserkov-preobrazheniya-gospodnya 9 Popov, N. L., Ljubimtsev, A. Yu., Kuusela, A. S., The Detailed Report on Preservation of Kizhi Pogost Monuments (Kizhi Pogost, C544) in 2012, Petrozavodsk: Kizhi Museum, 2012, 14. 10 Mackie, B. A., Notches of all Kinds. A Book of Timber Joinery (3rd ed.), Pender Island, BC: Log House Publishing Company, 1990, 19; Berg, A., Norske tømmerhus frå mellomalderen.

RZ_wooden-architecture.indd 166

29.09.15 21:23

167

Band 1, Alment oversyn, Oslo: Landbruksforlaget, 1989, 19. 11 Berg, A., 1989, 19–29. 12 From an upcoming book about different types of notches in Norwegian log buildings; Godal, J. B., 2015: oral information. 13 Viollet-Le-Duc, E., The Foundations of Architecture – Selections from the Dictionnaire Raisonné, New York: George Braziller, 1990, 195; the works of Viollet-le-Duc are classical examples of early restoration theory. 14 Popov, N. L., Glushko, I. M., Kuusela, A. S., Lyubimtsev, A. J., Nabokova, O. A., The Detailed Report on Preservation of Kizhi Pogost Monuments (Kizhi Pogost, C 544) in 2010, Petrozavodsk: Kizhi Museum, 17. 15 Skopin, V., 2014: oral information. 16 A sample from a rafter made of pine in Værnes church, Norway (dated 1140) proved to have extreme qualities, among other factors a density which is normal for harder wood species; Godal, J. B., Renmælmo, R., Berntsen, P., Kvalitet på material i Værnes kyrkje – eit forsøk på ein karakteristikk, unpublished, 2010. All illustrations are by the author except where noted.

RZ_wooden-architecture.indd 167

29.09.15 21:23

168

RZ_wooden-architecture.indd 168

29.09.15 21:23

169

Conclusion 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 : D I F F E R E N T A P P ROAC H E S TO N E W K N OW L E D G E / Siri S. Lexau

RZ_wooden-architecture.indd 169

02.10.15 16:29

170

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 : D I F F E R E N T A P P ROAC H E S TO N E W K N OW L E D G E

Siri S. Lexau Archaeology and geometry as possible indicators In this anthology, we have concentrated on some limited areas of research into wooden architecture in Russia and parts of northern Europe. The texts represent recent research in this field within architectural history, which through different approaches investigates technological and aesthetic aspects of the use of wood as a main structural material. The investigations reveal diverse and inventive methods of timber construction, both through physical, extant evidence and by interpretation of former buildings or stages in construction on the basis of archaeological records and written sources. The wooden architecture of the Middle Ages represented a huge number of buildings, an architectural heritage that for the most part has now been lost. So what can archaeological excavations tell us about superstructures that are no longer visible or adequately documented? The first chapter deals both with the occurrence and number of remains and traces stemming from wooden buildings and with possible ratios readable from foundation traces and archaeological discoveries. Jørgen Jensenius discusses how different solutions for bringing the loads of a structure down to the footings could influence the choice of either earth-fast or ground-set footings, and that these methods may have existed as parallel traditions depending on the design and load of the superstructure. Ratios and rules of proportions, such as the rafter span, may indicate possible connections between footings and construction system, between the choice of footings and the scale of the building. This question has, according to the author, barely been discussed seriously. All the same, he very strongly advises us as researchers against making deductions on poorly based assumptions. It is not possible to deduce the original superstructure of buildings from possible ratios and proportions, but the indications may give rise to reflections on possible geometric coherence. Mark Gardiner also points to some thinly based assumptions that have shaped research on medieval England’s timber churches. His text challenges the assumptions that 1)  many (even most) masonry churches were preceded by a timber structure, and 2)  that masonry is technologically superior to timber construction. He argues that timber in certain cases was chosen for other reasons and evidence for this may be found in later excavations. Some timber buildings were also major edifices, such as the abbey church of Bury St. Edmund’s, replaced by a masonry building in the eleventh century. This is in keeping with a major shift towards stone churches at the time. Excavations have also shown that the construction of churches was different from that of secular buildings, something that may result in distinctions that have previously been overlooked. Further investigation of the wooden churches of medieval England is much needed and

RZ_wooden-architecture.indd 170

29.09.15 21:23

171

Gardiner has pointed to fields that need closer examination. At the same time he also warns us against failing to distinguish between scientific evidence and creative imagination, while pointing out that imagination may be an important prerequisite for continued question posing and experiments – which is a very important tool for all research practice. The religious architecture of Novgorod presents a similar pattern of timber edifices being replaced by stone structures. The authors Marina Rodionova, Victor Popov and Nadezhda Tochilova state that fire was the main reason for replacing wooden buildings with masonry ones. However, in addition, such factors as economic preconditions and aesthetic and material preferences may have been in operation there as elsewhere. It is also anticipated that the foundations of wooden churches will to a large extent have been lost due to the deeper foundations needed for later stone churches. Practical studies of the medieval city of Novgorod’s wooden church architecture are all but non-existent to this day, but archaeological finds and the stylistic features of surviving wooden parts, notably the ‘columns of the oaken St. Sophia, provide a rich basis for a discussion of possible influences and dating. Stylistic influences may be derived from a broad abundance of possible sources, such as Byzantine, Scandinavian or Slavic traditions, or it may be the other way around, depending on definitive dating based on stratification and dendrochronological evidence. The shape of the wooden churches also remains a topic for further investigation. Later edifices from the seventeenth century indicate that ‘round’ ground plans may have been used for wooden churches, as polygonal faceted structures would make it possible to scale up buildings otherwise limited by the maximum available length of logs. Polygonal and especially eight-sided buildings may have been inspired as well by medieval number symbolism, Byzantine traditions or the ideal rotunda of the Holy Sepulchre in Jerusalem, as Richard Krautheimer indicates in his book Early Christian and Byzantine Architecture (1965/1986).

Written sources shed light on the past Should we wish to obtain accurate accounts of medieval wooden architecture through written texts, the sources can unfortunately give us few answers. However, as Jørgen Jensenius states in his article, a combination of sources may together give indications of the planning, design and assembling of wooden churches. St. John, with his allegorical description of the New Jerusalem (Revelation 21:10), inspired medieval church builders to treat the church interior as an earthly version reminding the congregation of this beautiful cityscape. Parallel to such ideals, the monastic orders preaching humility and the frugal use of economic resources erected simple churches. The significance of relics as essential artefacts belonging to the churches also contributed to more or less standardised elements, but not standardised design and workmanship, in a church interior and its crypt. The system seems to be more important than exact measures, specific instruction of aesthetic elements or types of construction. That makes the specific and accurate recreation of demolished churches a

RZ_wooden-architecture.indd 171

29.09.15 21:23

172

forlorn hope lost in the mists of the centuries, as it seems that the interpretations of the decisive rules could be diverse. Jensenius also confirms that, ‘Any demands for ecclesiastical conformity may have been met with arguments for local diversity’ (page 59). We may, however, find information on certain qualities belonging to impressive churches in such texts as the Laxdæla Saga quoted by Jensenius. Rune carvings may also inform us of the names of builders and craftsmen, while several texts relating to church dedication practice may reveal aspects of rituals and traditions that point to certain fixed elements. In England, medieval historical records are relatively rich compared with other parts of Europe. Those sources have, however, been exploited only to a limited extent by architectural historians, and Mark Gardiner is doing pioneering work in casting new light on this material. Some researchers such as Richard Gem, mostly unpublished, have already investigated sources such as saints’ lives, charters and chronicles. The archives of important churches, such as St. Paul’s in London, have also been studied. The dimensions of buildings, such as barns, on the estates belonging to that church were recorded. In conjunction with extant buildings, these sources give fairly good indications of shape and scale. However, wooden buildings, traditionally symbols of humility and modesty, were in general probably deemed of less interest and value and therefore not recorded. In the sphere of religious architecture, they were often minor buildings such as chapels or parish churches, and are therefore reflected less in the sources than stone buildings. Visitations are normally recorded, and these give information about necessary repair work to buildings belonging to the church. Gardiner is, however, somewhat pessimistic concerning the possible results of studying these resources, as he doubts that they will give substantial information about the wooden churches as buildings. In Russia, archival documents produced by different administrative structures in various historical periods have been systematised in the Russian State Archive of Ancient Documents (RGADA) and a wide range of other institutions, such as museum collections, archival funds and manuscript sections of libraries. One of the richest sources for medieval buildings is the records collected under the authority of the Pomestnyi Prikaz, a body of the central state administration. Evgeny Khodakovsky and Arina Noskova provide us with information on how this abundant material is organised and labelled, and it seems that the resources available are much more extensive than in other areas of Europe. Some records are even detailed enough for churches to be reconstructed from the descriptions, but of course not thereby corresponding with possible existing churches and their actual structures. From the imperial period of the eighteenth and nineteenth centuries, particularly after the introduction of photography, abundant material documenting wooden architecture is available to researchers. Unfortunately, a lot of physical churches and archives disappeared during the Soviet period, but, according to the authors, in the ‘1910s and 1920s a tremendous gathering of information about ancient works of architecture, taking stock of all the surviving cultural heritage, registering and recording (by photography and surveys) the objects was carried out by the department of

RZ_wooden-architecture.indd 172

29.09.15 21:23

173

the People’s Commissariat for Education responsible for museum affairs and the conservation of artistic and historical monuments’ (page 84). Only a small fraction of this enormous quantity of archival sources has been investigated, so there is reason to believe that many pages in the history of Russia’s wooden architectural heritage still remain to be written.

Wooden constructions – primary sources? When we move on to physical buildings, it is possible to acquire more exact knowledge of their material history. This chapter deals with visible, reconstructed and/or documented buildings. Although a substantial part of the medieval wooden buildings in the areas we are dealing with here has disappeared, having some physical objects or parts of objects available can provide extensive information. The works that have come down to us from this very rich heritage represent different ‘lines of descent’ of constructional techniques, development stages and local variants. So how can we move closer to possible conclusions about their creation processes? Norway has in its 28 stave churches a unique heritage from the Middle Ages that is practically unparalleled. Two exceptions are the Hedared Stave Church in Sweden and Greensted Church in England. Nonetheless, the 28 churches represent all that is left from approximately 1,000 churches that existed in Norway around the thirteenth century. The surviving churches have very different shapes, structural systems and decorative ornamentation. Through the efforts of the Society for the Preservation of Norwegian Ancient Monuments founded by J. C. Dahl in 1844 and the Norwegian Directorate for Cultural Heritage, many of the churches were measured and documented through drawings and photographs. Dahl was the first to understand the value of this heritage and as early as 1836–37 he was able to publish a collection of plates in Denkmale einer sehr ausgebildeten Holzbaukunst aus den frühesten Jahrhunderten in den innern Landschaften Norwegens. The earliest collected monograph De norske stavkirker (The Norwegian Stave Churches) was written by Lorentz Dietrichson, the first professor of art history in Norway, and published in 1892. The following year, he published Die Holzbaukunst Norwegens in Vergangenheit und Gegenwart together with the architect Holm H. Munthe. One may imagine that the scientific basis for research on stave churches was deficient, as many of the measured drawings of such complicated structures were executed in a hurry. A large number of them were about to be demolished as they were too small, old, not heated or in a very bad condition for religious use, and dedicated enthusiasts tried to document them and to attract attention before they were destroyed. Later, the remaining churches came to be regarded as an important cultural legacy, but only one of them, Urnes, is listed as a World Heritage Site. And funding for maintenance has, to be honest, not been abundant. In 2001, the Directorate for Cultural Heritage initiated the Stave Church Programme. Under this project, the condition of all the Norwegian stave churches was registered, and the need for repair, restoration and preservation was documented. Leif Anker, in his article here, discusses both the results and possible

RZ_wooden-architecture.indd 173

29.09.15 21:23

174

consequences of the investigations carried out during the programme. Dendrochronology, studies of wooden structures and joints as well as decorative programmes, united with photographic methods and sampling, have called into question earlier results based on typology and stylistic criteria. Workmanship, tool marks and the age of the timber seen in relation to the construction period and similarity with other churches may give new conclusions to earlier research based on inadequately grounded assumptions. One distinctive result of the Stave Church Programme is the very close cross-disciplinary collaboration established between archaeologists, art historians, building archaeologists and artisans regarding further research on stave churches. All professionals now regard such close partnership between disciplines as an essential part of more profound investigation into the history of this wooden heritage. In Russia, most of the remaining wooden churches date from the seventeenth and eighteenth centuries, and Andrei Bode presents us with an elaborate survey of features belonging to the Russian tradition of building in wood. Studies of the relationship between tools and material properties, different ways of using the logs and other structural parts, and the detailing of building elements give us an impression of a cultural heritage full of diversity. As Bode states, no two buildings are identical. In this wooden architecture, proportions and shapes of buildings are mainly determined by the log-based technique. This is a prerequisite for the maximum dimensions of wall parts and the way additional building units are linked together. One might say that the limits set by maximum log length were pushing builders to invent ever new building shapes and amazing features, constructed for functional and aesthetic reasons. Structural parts are also often carved and additionally used as decorative elements of a building, like the ‘St. Andrew’s crosses’ in Norwegian stave churches. Also, secondary building elements without structural importance, such as exterior stairs with intermediate landings and roofs, are used for both their functional and aesthetic values. Their designs often united traditions of folk architecture with elements of classical orders and various local styles. Ilya Antipov studies the wooden elements of masonry buildings, as all stone constructions were dependent on wooden parts – for structural reasons as well. In addition, he lists a wide range of details, interior features and foundation parts that were made of timber: ‘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’ (page 118). A substantial amount of these elements have been lost, but drawings of structures and joints made by researchers provide us with documentation of existing (or now lost) visible elements or traces of wooden parts.

RZ_wooden-architecture.indd 174

29.09.15 21:23

175

Karel Kuc ˇa has studied a minor, but nonetheless very important adjunct to the architecture of churches: free-standing bell towers. He has found that of the several thousand bell towers that existed in the past in Scandinavia and eastern Europe, there is a special type to be found in Sweden and Czech Bohemia: bell towers with triangular frames, free-standing or integrated into other structures. This is the most stable construction for absorbing the dynamic forces of heavy bells in motion, but why can such large triangular-framed structures only be found in these two regions, and why in such distant territories? Kuc ˇa proposes one possible reason: the Thirty Years’ War of 1618–48, when Bohemia was occupied by Sweden. A 1602 drawing by Jan Willenberg shows a bell tower of the triangular type in a town in Rakovník district, Bohemia, and several others built in the sixteenth and seventeenth century also existed. Accordingly, there is a possibility that the Swedish occupying forces brought the tradition directly back to their home country. Dendrochronology may reveal the age of the Swedish towers and thus reveal whether this is a tradition that appeared in Sweden only after the occupation of Bohemia.

Past restorations and contemporary responsibility The last chapter of this anthology deals with the documentation and preservation of medieval wooden architecture, a responsibility that all territories possessing this vulnerable heritage have to take seriously. Restorations of wooden churches have in many cases been guided by (often nationally coloured) romantic perceptions of the architectural heritage of a distant past. Around 1900, many of the Norwegian stave churches were clad in weathered and simple materials that did not at all accord with people’s imaginations of a rich decorative carving culture of the Middle Ages. Huge Viking ships had been excavated in the late 1800s and archaeologists had uncovered a rich and advanced ornamental culture very suitable for representing the proud past of a newly liberated nation, which for centuries had been tied to and subjugated by other powers. Therefore in certain cases the cladding was removed from stave churches, some of them were also moved to protect them from demolition, and they were given an elaborate exterior similar to the church regarded as an aesthetic ideal: Borgund. Thanks to records made by persons connected to cultural heritage associations in Norway, we know their approximate general dimensions and shape thanks to drawings and photographs, but a lot of original parts added over the centuries have been lost. Andrei Bode discusses restoration practice in Russia, where the majority of medieval or later buildings were given wooden cladding, metal roofs, and new, imaginative extensions and porches in place of old ones. Additionally, Russia’s wooden architecture was heavily restored in the late nineteenth and early twentieth centuries. Restoration architects of the mid-twentieth century inherited the attitudes of the early twentieth-century architects. To return buildings to their original appearance, they resolutely removed later additions and cladding and carried out reconstructions of the wooden roofs.

RZ_wooden-architecture.indd 175

29.09.15 21:23

176

Today our objective is to convey a historical record of a building based strictly on confirmed evidence. We seek to preserve as much as possible of a building’s material authenticity down through the ages, and not to take it back to one particular stage of history according to changing taste and ideals, without historical or material evidence. To achieve this, dealing with the fragile, existing evidence of a centuries-long architectural heritage exposed to frequent fires, climatic influences and rot, subtle methods must be applied. Some interesting exchanges of methods and experience have therefore been taking place across borders. Arnt Magne Haugen has been physically working on the challenges posed by one of Russia’s most complicated and spectacular structures: the summer Church of the Transfiguration at Kizhi Pogost, Karelia. It is said to have been built during the reign of Peter the Great and is possibly the sole survivor of the multi-domed churches built in that period. The church is a 37-metre high building constructed with pine logs, and crowned by twenty-two onion-shaped domes clad with aspen shingles. Haugen applied his experience with log constructions to develop a restoration method that involved neither the restorers working out in the open, nor the building having to be dismantled and moved to another site. So how was it possible to introduce new elements to replace dilapidated or rotten parts in the middle of the walls of this complicated structure? In this case an advanced and specially designed lifting system was created. This made it possible to lift the building horizontally at any level, so that parts could be removed and taken away for repair work without dismantling the whole edifice. The advantages are obvious. The risk of losing exact information about joints and details when a building is deconstructed can be greatly reduced by this method. Although it is not possible to transfer this approach to stave churches, due to their completely different structural qualities, I will let this very practical, but also scientifically based procedure end our survey. We have seen revelations of a fascinating common heritage, but also of problems, a glance into complex research fields and discussions on proper and responsible management. The purpose of this anthology has been to give an updated insight into contemporary research activity on medieval wooden architecture across geographical and disciplinary borders, which is quite new but absolutely necessary. In Norway, the Directorate of Cultural Heritage recently invited researchers at the main universities dealing with architectural heritage to a seminar discussing current research on stave churches. This dialogue revealed that there was, in fact, hardly any research taking place within these academic institutions. During the first two-thirds of the twentieth century, a whole generation of researchers devoted a substantial part of their efforts to this aspect of our national heritage. Now it is most important that new generations pose new questions about both the empirical material itself and the findings of former researchers. A lot of buildings or parts of buildings are still standing, in addition we have archaeological records, written sources and endless discussions of how to manage this heritage in a proper way. We hope that the texts collected here will nurture curiosity and inspire the reader to put forward new questions, which may in turn inspire contemporary researchers to embark on further investigations.

RZ_wooden-architecture.indd 176

29.09.15 21:23

RZ_wooden-architecture.indd 177

29.09.15 21:23

RZ_wooden-architecture.indd 178

29.09.15 21:23

Author Information Appendix / Imprint

RZ_wooden-architecture.indd 179

29.09.15 21:23

180

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

Siri S. Lexau b. 1951. Professor, Dr. Art. at the Department of Linguistic, Literary and Aesthetic Studies, University of Bergen. Her primary research activity is linked to the history of architecture and urban studies. She has published widely on the history and transformation of sites and buildings.

Karel Kuča b. 1961, Architect, Czech Technical University Prague, National Heritage Institute Prague, The Society for the Research of Wooden Churches and Belltowers. His field of interest is linked to the history of architecture (especially of wooden churches and bell towers) and urban studies, including the principles of heritage protection. He publishes actively on the history and preservation of historic towns, villages and wooden architecture.

Ilya V. Antipov b. 1976. Assistant professor, PhD at the Department of the History of Russian Art, St Petersburg State University. The main topic of his research is the history of Old Russian architecture, building materials and technology. He has provided the archaeological investigations on many monuments of Novgorodian architecture of 12th to 17th centuries.

Nadezhda N.Tochilova b. 1987. PhD.  Her primary research activity is linked to the history of Viking Age art, medieval Scandinavian and medieval Russian applied art. 

Bode Andrei b. 1967. Architect-restorer, PhD in architecture. Researcher at the Institute of Theory and History of Architecture and Urban Planning, Moscow. Research and creative interests: history of wooden architecture, ancient traditions. Author of books and articles on the wooden architecture of Russia and problems of restoration.

Viktor Popov b. 1953. Leading architect, adviser of The State Museum of National Wooden Architecture Vitoslavlitsy, Novgorod the Great (Russia). His research and practice are linked to civil and church architecture of the Novgorod Region of 12th to 17th centuries, and to the preservation of cultural heritage.

RZ_wooden-architecture.indd 180

29.09.15 21:23

181

Marina A. Rodionova b. 1958. Archeologist, Senior Research Assistant of the Center of Archeological Research Organization and Implementation of Novgorod State Museum. Her sphere of academic interests includes history and the archeology of medieval Novgorod. She is involved in the research of Novgorod Detinets and Vladichn Yard.

Arina G. Noskova b. 1988. Research fellow at the Scientific-Research Institute of the Theory and History of Architecture and Urban Planning, branch of the Federal State Budget Institution Central Scientific-Research and Project Institute of the Construction Ministry of Russia. She specialises in wooden ecclesiastical architecture of northwestern Russia and archival research.

Evgeny V. Khodakovsky b. 1974. Head of the Department of Russian Art History, St. Petersburg State University, and Associate Professor in Art History. His fields of interest are Russian and Northern European medieval art and the wooden architecture of Russia and Scandinavia.

Mark F. Gardiner b. 1958. Senior Lecturer in the School of Geography, Archaeology and Palaeoecology, Queen’s University Belfast. His research interests lie in the archaeology of medieval buildings, trading systems in the North Atlantic and medieval landscapes. He has published on the development of English medieval houses.

Jørgen H. Jensenius b. 1946. Architect, Dr.ing. specialises in  surveying  and researching medieval wooden churches. His credentials include employment with the Central Office of Historic Monuments and Sites, and the Norwegian Institute for Cultural Heritage Research in Oslo. http://www.stavkirke.info/

Leif Anker b. 1956. Art historian. Senior adviser, Directorate of Cultural Heritage, Norway, where he has participated on the Stave Church Programme. His main interest and published work is on stave churches and architectural history.

Arnt M. Haugen b. 1972. Senior adviser at the Directorate for Cultural Heritage, Norway. Master of Architectural Conservation, Oslo School of Architecture and Design with extensive practical experience as a carpenter, involved with the restoration of log buildings. His main field is the restoration and conservation of traditional wooden constructions, but he is also interested in questions concerning repair or modernisation of protected buildings in general. He has attended several ICOMOS missions to Kizhi Pogost.

RZ_wooden-architecture.indd 181

29.09.15 21:23

182

INDEX

Belarus 134

Frimley 36

Woodeaton 35

Goltho 32

Worcester 73

Greensted  13, 29, 32, 33, 35, 38, 94, 173

Worcestershire 35

Hampshire  35, 36, 37

Yorkshire 30

Czech Republic  8, 10, 13, 134, 138,

Hereford  69, 73

Čáslav 139

Hertfordshire 35

Český Brod  141

Kensworth  69, 70

Chlumec nad Cidlinou  141, 142

Kent  32, 69, 73

France 134

Hradec Králové  139

Leighton Buzzard  71

Rouen 34

Finland 134

Kočí 134

Lincolnshire  30, 32, 74

Mesnil-Mauger 34

Mělník  137, 138

Little Oakley  29

Val-de-la-Haye 34

Nový Bydžov  138, 139

London  13, 69, 72, 172

Gisors 34

Rakovník 137, 138, 175

Mattingley 38

Rtyně v Podkrkonoší  134

Melverley  36, 38

Slaný  138, 139

Navestock  71, 72

Bremen 10

Sudoměř  138, 139

Nazeingbury 30

Cologne 53

Turnov  139, 143, 141

Norfolk  31, 32

Künzing 8

Úpice 134

Northamptonshire  31, 32

Mecklenburg-

Norwich  30, 31, 34, 35 Denmark  11, 44, 134, 136

Oseney Abbey  72

Germany  51, 134, 136

Western Pommerania  136 Schleswig-Holstein 136

Oxford  30, 72 England  8, 10, 13, 15, 28, 29, 34, 35, 36, 38, 68, 71, 74, 94, 135, 170, 172, 173

Oxfordshire 35

Iceland  15, 60, 61, 62, 63

Pembridge 73

Skálholt  60, 62

Peterborough Abbey  68, 70

Arborfield 73

Plaistow 36

Ireland

Ardleigh 70

Potterne  31, 32

Dublin 63

Asheldham 29

Raunds  31, 32

Glendalough 10

Aylesbury 71

Rotherwick 37

Kells 10

Berkshire 73

Ruscombe 73

Bury St. Edmund’s Abbey  71

Sandon 70

Italy

Canterbury 64

Shropshire  34, 36, 37

Ravenna 53

Cheshire  35, 36, 37

Somerset 30

Rome  10, 29

Colchester  35, 72

Stafford 30

Cumberland 73

Staffordshire 28

Derby 32

Suffolk  30, 32

Dorset 30

Surrey  36, 73

Dover  32, 33, 38

Sussex  36, 68

Durham  69, 70, 73

Thetford  31, 34

Earley 73

Vowchurch 73

21, 22, 26, 44, 58, 60, 61, 62, 63, 65,

East Riding  30

West Bergholt  29

94, 95, 96, 97, 99, 101, 104, 134, 160,

Lithuania 134 Netherlands  134, 136 Norway  8, 9, 10, 11, 12, 13, 14, 15, 20,

Essex  13, 28, 29, 30, 32, 34, 35, 94

Westwick  69, 70

Eynsham 35

Whaplode Drove  74

Årdal 104

Fairfield 73

Wharram Percy  30

Bergen 12

Foxley  30, 34, 35

Wiltshire  31, 73

Bø  22, 23

RZ_wooden-architecture.indd 182

165, 173, 175, 176

29.09.15 21:23

183

Borgund  24, 25, 26, 59, 101, 175

Kushereka  119, 148

Eidsborg 12

Lake Onego area  14, 114

Eidskog 22

Mezen  86, 114

Fantoft  22, 106

Moscow  50, 79, 81, 82, 83, 84, 85, 86,

Feios  99, 104, 105, 106

87, 89, 110, 150

Sweden  11, 13, 44, 94, 134, 135, 136, 137, 138, 139, 140, 141, 142, 173, 175 Bohuslän 142 Dalsland  139, 140 Gästrikland 140

Fortun 106

Muezero Monastery  15

Hopperstad  11, 99, 100, 101, 102,

Murmansk 78

Hälsingland 140

Nënoksa  152, 153

Härjedalen 140

104, 105, 106

Gotland  44, 45

Høre  22, 23, 62

Northern Dvina  86, 114, 117

Hemse  45, 47

Kaupanger  22, 23, 26, 60, 103, 104,

Novgorod (Novgorod the Great) 

Jämtland 140

105, 106, 107

14, 15, 42, 43, 44, 46, 47, 48, 49, 50,

Lake Vättern  137

Kinsarvik 23

52, 53, 54, 78, 81, 82, 88, 90, 110,

Medelpad 140

Kvernes 13

122, 123, 124, 125, 126, 128, 129, 131,

Närke 137

Lardal  44, 45

147, 171

Östergötland  137, 138

Lom  12, 23

Olonets  81, 87

Småland  135, 137

Lomen  25, 26

Olonets guberniia  87

Södermanland 137

Møre og Romsdal  13

Petrozavodsk  82, 87, 90

Uppland  139, 142

Nidaros  23, 60, 61

Pinega  86, 88, 114

Vänern 137

Numedal 96

Povenets  84, 85

Västergötland  137, 138, 139

Oppland 12

Pskov  78, 125, 126, 128, 129, 130

Ringebu  22, 23, 64

Purnema 83

Ukraine  50, 134

Sogn  9, 11, 94, 105

Rostov 50

Galich 50

Telemark 12

Shirkovo 149

Kiev 50

Torpo  60, 99

Siberia  110, 111

Voskresensk 50

Trondheim 12

Smolensk 50

Urnes  8, 9, 23, 26, 44, 94, 97, 98, 99, 100, 101, 102, 103, 105

Soginitsy  152, 153

Wales 73

Solovetskii Monastery  80

Uvdal  23, 95, 96, 97, 98

Staraia Ladoga  14, 17, 122, 125,

Russia  9, 10, 13, 15, 78, 81, 86, 89,

Suzdal 46

129, 130 110, 146, 149, 153, 156, 170

Tot’ma 78

Arkhangelsk  16, 78, 81, 82, 88, 90, 148

Tsypino 151

Borodava  147, 150

Tver 50

Brusenets 118

Ustiug (Velikii Ustiug)  51, 78

Karelia  9, 15, 78, 82, 86, 87, 88, 90, 112,

Vazhiny 52

115, 156, 176

Velikaia Guba  112

Kargopol’  78, 150

Vershina 148

Kholmogory 78

Vladimir  45, 46, 127

Kirillo-Belozerskii Monastery  150

Vologda  78, 81, 82, 88, 90, 117, 119

Kizhi  8, 9, 16, 86, 87, 88, 90, 147, 156,

Vysokii Ostrov  148

158, 160, 162, 163, 166, 176 Kolomenskoe, Moscow  88, 119, 150

White Sea  14, 86, 117 Yakutia 113

Kondopoga  116, 147 Krasnaia Liaga  152

RZ_wooden-architecture.indd 183

Slovakia 134

29.09.15 21:23

Copy editing Julia Dawson Project management Alexander Felix, Lisa Schulze Production Amelie Solbrig Layout, cover design and typography Jenna Gesse Typerfaces Campton, Parable Paper Munken Lynx, 130 g/m2 Printing DZA Druckerei zu Altenburg GmbH The edition is sponsored by SIU – Norwegian Centre for International Cooperation in Education (project CPRU-2011/10073 University of Bergen – St. Petersburg University cooperation programme). Library of Congress Cataloging-in-Publication data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the German National Library The German National Library lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, re-use of illustrations, recitation, broadcasting, reproduction on microfilms or in other ways, and storage in databases. For any kind of use, permission of the copyright owner must be obtained. This publication is also available as an e-book (ISBN PDF 978-3-0356-0542-6; ISBN EPUB 978-3-0356-0550-1).

© 2016 Birkhäuser Verlag GmbH, Basel P.O. Box 44, 4009 Basel, Switzerland Part of Walter de Gruyter GmbH, Berlin/Boston Printed on acid-free paper produced from chlorine-free pulp. TCF ∞

Printed in Germany ISBN 978-3-0356-0566-2 9 8 7 6 5 4 3 2 1

RZ_wooden-architecture.indd 184

www.birkhauser.com

02.10.15 16:30