14. Snow, Earthquakes, and String Theory

2006 was a bad winter, with record snowfalls that killed many people. In Japan, over 60 people died in January. Many elderly people were killed trying to clear snow from their roofs, while others were crushed to death as their houses collapsed. "It's frightening," said one elderly woman in Asahi. "There were creaking sounds and I couldn't open the doors because of the weight of the snow."


Collapsing houses are not the first things that come to mind when we think about the effects of snow. In the U.K., we have civil engineers and architects who sort that stuff out. However, even modern buildings can be caught unaware by the climate, and snow, being made of water, is heavy.

The collapsed convention hall at Katowice, Poland, January 2006

On the 2nd of January 2006 the roof of a 1970s ice rink at Bad Reichenhall in southern Germany collapsed, killing 15 people and injuring 32. At the end of January the snow-covered roof of a modern convention hall near Katowice in southern Poland collapsed on 500 racing pigeon enthusiasts, killing 67 of them and injuring 150. The number of pigeons killed was not reported. A month later, in Moscow, the reinforced concrete roof of a market collapsed, killing 48 people, in what Mayor Yuri Luzhkov described as a “a technical accident."

These tragic events are surprising because we imagine that gravity has been tamed by mechanics and climate has been modelled by statistics. A modern architect is obliged to consider wind, rain, and snow, using official tables and standards.

The same considerations are implicit in the design of prehistoric buildings. If ancient structures had routinely fallen down in extreme weather, architects might have died out as a profession, and people would have returned the certainties of cave living.

Extreme climatic events are, by definition, rare. The expression “…..worst in living memory” is often employed to describe their severity. This has a certain pertinence in prehistory. Working without written records of the climate, or a detailed knowledge of mechanics, prehistoric builders had to deal with the problem of creating buildings that did not fail in bad weather. The only conceivable strategy to deal with the vagaries of the environment is one of ‘over design’: you make the structure as strong as you can. In terms of the timber buildings of prehistory, this is done by using lots of posts to hold up the roof, which has implications for the layout and utility of the interior.

The massive stone post and lintel architecture of the Hypostyle Hall at Karnac

In this respect, prehistoric timber building was no different from any other form of architecture that is driven by the need to create roofed space that satisfies both gravity and utility. One of the greatest spaces of the ancient world was the Hypostyle Hall, probably built by Seti I (1291-1279 BC), as part of the temple of Amon-Re at Karnac. It illustrates this dilemma at the heart of building.
The Hypostyle Hall had an area of 5000 m2 (103 x 52m). Its flat stone roof was supported by 16 rows of massive pillars, 132 in total, over 20m tall, with 12 exceptionally large ones, 24m tall, at the centre. Each column is 2.8m in diameter, about the same width as the gaps between them, creating a forest of stone with acres of surface decoration.

A plan of the Hypostyle Hall at Karnac, with its 132 pillars

However, the rows of columns slice the interior space into long thin strips, albeit very tall ones, and, even by the standards of Classical spaces, the compromise with gravity is extreme. But you have to start somewhere.

The timber architecture of northern Europe starts in the ‘Neolithic’, the period of the earliest farming peoples in Europe. Starting in about 7000 BC, the first cereal farmers spread northwards and westwards into Europe from the Near East. They sought out light, easily cultivated alluvial soils, spreading round the coast of the Mediterranean and up the Atlantic seaboard, as well as up river systems like the Danube and the Rhine. Agriculture had reached central Europe by 5500 BC and was established in the British Isles about 4000 BC.

The plan of a Neolithic longhouse, Building 32 at Elsloo in the Netherlands, with a part of a 1970s reconstruction of its posts and lintels (tied together) [1]

If you don’t have a weatherproof granary, there is little point in cereal farming, and building is best understood as part of the farming package. Like other aspects of farming, building systems and styles would have adapted to the changing environment as Neolithic culture spread north and west. The balance of domesticated plants and animals had to change, and so did building. The use of mud brick gave way to post-built wooden longhouses.

The Neolithic longhouse is heavily constructed, with dozens of posts in the interior space. This style of structure, where the weight of the roof is carried on posts buried in the ground, persists for most of prehistory. In northern Europe it is gradually replaced in the Iron Age and Roman periods by structures with load-bearing walls. This change reduced the number of posts or columns required to support the roof, increasing the utility of the interior space.

The traditional Chinese timber hall has stayed close to its Neolithic roots

However, something very different happened to the Neolithic longhouse in China, where the traditional timber hall has stayed close to its Neolithic roots, with the weight of the roof still carried on posts rather than on rigid walls. This system had proved better able to survive earthquakes than solid walled structures, an important consideration in the Chinese culture.

Conversely, on the northern edge of Neolithic expansion, timber building had to be abandoned quite early. On the Orkneys, a group of islands off the north coast of Scotland, the natural tree cover of birch, aspen and rowan is not good for building, and, as often happens on islands, trees became scarce, presenting a problem for Neolithic builders.

Flat bedded Devonian sandstone in northern Scotland

However, by a stroke of geological luck, the Orkneys had been at the bottom of a huge fresh water lake between 400 and 360 million years ago in the Devonian Age, and parts of the islands are covered by layer upon layer of thin flat sediments, which solidified to become an easily serviceable building stone. A series of remarkable stone-built Neolithic settlements have been found on the islands, often buried under drifting sand around the coast.

One of the Neolithic houses at Skara Brae, Orkney, with its stone 'furniture'

The eight houses linked by passages found at Skara Brae are famous for having stone ‘furniture’. Two oblong stone-built houses at the Knap o' Howar, on Papay, date from about 3,600 BC and are more typical longhouses, not entirely dissimilar to a traditional croft.

Neolithic stone longhouse at Knap o' Howar, on Papay

In common with Neolithic communities elsewhere, stone circles and subterranean communal tombs were built on Orkney. At Maeshowe the utility of the local sandstone allowed the creation of an exceptional corbelled roof. The same basic technique was being used in the construction of burial chambers inside the pyramids built in Egypt at about the same time. Geographically on the edge of the Neolithic world, the forced use of stone in these Orkney communities has led to the preservation of some of their building culture. As so often happens in archaeology, it is the unusual that is preserved. Neolithic Orkney illustrates yet another aspect of the complex relationship between building and a local environment.

During the Neolithic, where available, stone was used for tombs and monuments. The simple ‘megalithic’ style of architecture, utilising natural slabs of stone with a minimum of working, is very durable, especially for subterranean tombs, where timber would rot. In exceptional cases, such as Stonehenge, stone was transported, but in many areas, such as southeast England, it was simply not available, and timber was the only structural building material.

Some archaeologists have suggested that the use of stone was of itself symbolic of death. If true, such symbolism would have a limited geographical relevance. It has even been suggested that, by contrast, some wooden structures represent life, which would be news to the trees, who pretty much have had their chips when they are cut down for timber. The choice of building material is very much a response local conditions, and is not a straightforward cultural decision.

Knowing how to put these heavy timber components together safely, so they can withstand the local environment without killing their occupants, was a specialist job, even in the Neolithic; someone has to make the call as to what is safe. Even over-design requires a designer.

Neolithic house at Llandygai, Anglesea, a modern CAD style reconstruction;

note the use of 'string' [2]

For four millennia, post and lintel timber building was the dominant system of architecture in Britain, part of a tradition with even older roots in Europe. However, the primitive assumptions made by archaeologists about building in the Iron Age leave little room for development in the 4000 years, leading some to conclude that Neolithic buildings were tied together with some form of rope or string. I don’t think I have seen this written explicitly, but this ‘string theory’ is part of the visual language of prehistory used in illustrations.


Leaving aside the sheer impractically of successfully tying together timbers weighing hundreds of kilograms with honeysuckle or something similar, it implies that Neolithic people could not make a hole through a piece of wood and form a simple tenon joint. A post with a tenon joint at the top to join to a horizontal timber is the reason you use ’post & lintel’ architecture. It's how it works. Not jointing the post and lintel is like using nuts but not bolts!


Neolithic woodworking tools were made of stone. Toolmakers dug mines with antler picks to find the best flint beds, and sought out the hardest stone they could find, then laboriously ground and polished it into tools. They would even bore shaft holes through hard stone axes – and it is not immediately obvious how you drill a hole through stone! (They probably used a bone drill with abrasive sand.) There is something uncompromising about Neolithic tools; they are made to be as good as possible.

A perforated stone axe

The notion that a culture that had the patience to bore a hole through the hardest stone they could find would not be capable of putting a hole through a piece of wood with such a tool is quite perverse. Furthermore, while archaeologists have to accept that some stone tool manufacture was a specialist craft, many would find the idea of specialist builders or architects difficult to justify.

The early antiquarians had to fit all history into the 6000 odd years allowed for it in the Bible, but British prehistorians have made 6000 years of timber building end up with the simple form of mud hut we have imagined for the Iron Age.

Sources & Further Reading:2006 snow [all links accessed 12 October 2009]:
http://www.chinadaily.com.cn/english/doc/2006-01/29/content_516334.htm
http://edition.cnn.com/2006/WORLD/europe/01/29/roof.collapse/index.html
http://english.peopledaily.com.cn/200602/24/eng20060224_245540.html
http://www.redorbit.com/news/international/350344/death_toll_from_record_japan_snowfall_rises_to_63/
Orkney [accessed 12 October 2009]: http://www.orkneyjar.com/index.html
Evidence for string [accessed 12 October 2009]: http://www.telegraph.co.uk/news/newstopics/howaboutthat/3681143/Archaeologists-find-piece-of-string-dating-back-8000-years.html

References:
[1] W. Startin (1978), "Linear Pottery Culture Houses: Reconstruction and Manpower." Proceedings of the Prehistoric Society 44: pp. 143-59.
[2] Neolithic house at Llandygai [accessed 12 October 2009]: http://www.heneb.co.uk/llandegaiweblog/neolithichouse.html