The New Order

How would someone who had been brought up in a roundhouse adapt to life in a rectangular world? The experience of a servant working for a family in Malawi shows how difficult it could be. Her predicament is described in a book entitled Women’s Work in Heathen Lands, published in 1886. Jan Deregowski quotes the following extract:… In laying the table there is trouble for the girl. At home her house is round; a straight line and the right angle are unknown to her or her parents before her. Day after day therefore she will lay the cloth with the folds anything but parallel with one edge of the table. Plates, knives and forks are set down in a confusing manner, and it is only after lessons often repeated and much annoyance that she begins to see how things might be done (Laws 1886, quoted by Deregowski 1973: 180–1)… That simple story introduces a larger issue. Under what circumstances did people make the transition from a world of circular structures to one of squares and rectangles, and how were their lives affected by that process? It is surprising how much attention had been paid to structural changes among ancient buildings and how little to the political and social circumstances in which they happened. One way of approaching this topic is not only studying the advantages offered by new styles of architecture, but also asking which important features might be lost. That is too rarely considered. Many of the approaches described in Chapter 2 emphasized the possibilities offered by the change from circular to rectangular buildings. Houses could be larger and could accommodate more people; they would be easier to maintain; they could be expanded as the number of inhabitants increased and space was subdivided; in many cases rectilinear dwellings could be inhabited over longer periods than roundhouses. None of those arguments is unsatisfactory in itself, but all are incomplete because they do not take into account the motives of the people who chose to live there. Chapter 2 also showed how houses could be used to emphasize subtle distinctions among their inhabitants: differences that were based on age, gender, and social standing.

The Attraction of Opposites

One of the best known accounts of the psychology of perception is Richard Gregory’s book Eye and Brain (Gregory 1998). It is relevant to this chapter because it uses an example from archaeology to illustrate the way in which the mind creates visual patterns. The author considers the methods by which excavators distinguish between the remains of rectangular and circular buildings. He considers the Middle Bronze Age settlement of Thorny Down in southern England, where different scholars have inferred the existence of different types of buildings on the basis of the same field evidence. The original excavator was uncertain of the precise form of the settlement (Stone 1941), but, in later years, Piggott identified the site of a large rectangular house there (1965: Figure 87) and Musson recognized circular structures (1970: 267; Figure 57). Gregory’s summary of their method is as follows:… Science and perception work by knowledge and rules, and by analogy . . . [In the case of Thorny Down] some of the holes in the ground might be ancient post holes; others might be rabbit holes, to be ignored. One group of archaeologists accepted close-together large holes as evidence of a grand entrance. They were altogether rejected by other archaeologists. One group constructed a large rectangular hut; the other, a small rectangular hut, and a circular building. ‘Bottomup’ rules—holes being close together and forming straight lines or smooth curves, and ‘top-down’ knowledge or assumptions of which kinds of buildings were likely—affected the ‘perceptions’. Both could have been wrong (1998: 11–12)…. The identification of a rectangular building at Thorny Down took place at a time when it was believed that the Netherlands had been settled from England during the Bronze Age. The argument was based on pottery styles and the distribution of metalwork (Theunissen 2009). Most likely there were contacts in both directions. As the Low Countries were characterized by a tradition of rectilinear architecture, what could be more natural than the construction of a longhouse at a site on the Wessex chalk? Dutch prehistorians attempted to find similar links between domestic architecture on both sides of the North Sea and soon they identified roundhouses of British type in their excavations.

Significant Forms

Most of the problems investigated by prehistorians involve material that has been known for a long time. Thus, artefacts, settlements, and monuments have been placed in chronological order and their distributions have been mapped. Taken together, the results of this work constitute an ‘archaeological record’ which can be analyzed and interpreted with the aid of different theories. Only occasionally is the nature of that record entirely transformed. It has happened in some regions where large-scale excavations are necessitated by commercial development, but it is in Central and south-eastern Europe that the transformation has been especially dramatic. This has happened for two reasons. The first is a consequence of political changes which have allowed archaeologists to employ aerial photography for the first time. The second has been the use of large-scale geophysical survey which has often been inspired by the discoveries made from the air. Taken together, these new methods have identified largely new classes of prehistoric monument and have shed fresh light on those that were already known. An unexpected result of these developments has been the discovery of enclosures dating from the Neolithic period. Until twenty years ago their distribution was largely confined to West Germany and Austria. Now that restrictions on flying have been lifted, it has been possible to undertake aerial reconnaissance in other areas and similar monuments have been found in considerable numbers in East Germany, Slovakia and Hungary (Andersen 1997: Chapter 5). At the same time, the rapid development of geophysical survey has had dramatic consequences. It has shown that some of the prominent tells that were already recorded were bounded by considerable earthworks. The reason that these discoveries were so surprising is that the enclosures were circular, while the houses of the same date were rectangular. The same situation arises in other regions and periods. What was the significance of curvilinear structures in a right-angled world? Part Three of this book investigates this relationship. The oldest circular enclosures probably originated during the earlier fifth millennium BC. There are some striking contrasts between them.

The Enormous Room

It is ironic that anyone studying the domestic architecture of Copper Age Sardinia must turn to underground tombs as a source of information, whilst the monumental ‘houses’ of the Bronze Age are among the most conspicuous structures anywhere in Europe. The contrast between these two periods introduces a new theme. The first part of this chapter will study stone towers and related structures in the West Mediterranean. In every case they were associated with settlement sites and some may have been domestic buildings in their own right. Over time, they came to favour a circular plan. This section considers monuments in Sardinia, Corsica, and the Balearics and compares them with sites in Spain. It also asks whether they represent a single phenomenon. The second part considers three groups of timber buildings in Britain and Ireland and the enclosures with which they are associated. In this case they date from different periods. Although they resemble one another on the ground, there were few direct connections between them and, in at least two instances, they are thought to have been ceremonial centres. These sites in the Mediterranean and the North Atlantic are considered together in the light of a recent analysis of monumental architecture in the eastern USA. Both sections are designed to complement one another. The first considers the development of Great Houses in a series of well-populated landscapes. The second works at a larger geographical scale and studies the ways in which such buildings, and the earthworks associated with them, were related to a still more extensive area, until the most elaborate of all could be considered as microcosms of the wider world. Chapter Four discussed the domus de janas of Sardinia, the surprisingly realistic copies of domestic interiors provided for the dead during the Ozieri Culture. They were lavishly decorated and have produced radiocarbon dates which suggest that they were contemporary with the later phases of megalithic art in Western Europe. At the same time, most of the Sardinian structures were roughly square or rectangular and only a minority were designed as depictions of roundhouses. That reflects the results of excavation at the comparatively few surviving settlements of the same date.

Life and Art

Not many prehistoric houses survive above their foundations. The three dimensions of the buildings are collapsed (sometimes literally) into the two dimensions of the site plan. That may be all that can be discovered by archaeology, and yet the missing component could have been all-important. The change of perspective is revealing, for the treatment of the walls and roof may be just as significant as the layout of the floor. Few excavated houses are as well preserved as those in the Near East, and there are many parts of Europe in which the question cannot be investigated directly. Here, the existence of ceramic models suggests an alternative approach. During 2010, two exhibitions featuring the arts of the first farmers took place in Britain. They ran simultaneously, one in Oxford and the other in Norwich. They also complemented one another geographically and thematically. The Lost World of Old Europe was organized by The Institute for the Study of the Ancient World at New York University (Anthony 2010), and Unearthed by the Sainsbury Centre for Visual Arts of the University of East Anglia (Bailey et al 2010). The display at the Ashmolean Museum in Oxford featured artefacts from Romania, Moldova, and Bulgaria, whilst that in Norwich was restricted to finds of figurines from Romania, Albania, and Macedonia, although they were compared with others from the Jomon Culture of Japan. Not surprisingly, the Neolithic and Chalcolithic objects spanned a long period of time and were associated with several regional groups. Some were elaborately decorated, while others were entirely plain. The artefacts shown in Norwich were all depictions of the human form, but those in Oxford also included pottery vessels, stone artefacts, and early metalwork. One small group of objects was especially striking, for it consisted of ceramic models of domestic buildings. In one case, from the Cucuteni Culture of Romania, a group of figurines had been discovered inside a miniature house of this kind. The evidence of such models is revealing. There were examples in which the outer wall was highlighted by angular designs, as if to emphasize the rectilinear outline of the building, but there was also a model in the Oxford exhibition which showed a structure with a similar ground plan whose exterior was covered by curvilinear motifs.

Conceptions and Perceptions

The first academic conference I attended was held in 1970. Its proceedings were published two years later with the title Man, Settlement and Urbanism. The book was notable for including contributions from both archaeologists and anthropologists. The meeting itself had an unusual format—part of the audience was hidden from view. Instead of participating in the discussion, we gathered in another room and watched the proceedings on closed circuit television. Only at the final reception was the barrier removed so that we were able to see the protagonists in the flesh. As this chapter will suggest, obstacles to communication still remain. At the time, one paper had a special impact—Kent Flannery’s discussion of ‘The origins of the village as a settlement type in Mesoamerica and the Near East’ (Flannery 1972). Four decades after it appeared in print it is still being quoted. Other contributors to the meeting covered some of the same themes, but, true to the spirit of the conference, the distinguished social anthropologist Mary Douglas considered the perils of archaeological interpretation. Her paper ‘Symbolic orders in the use of domestic space’ presented a series of cautionary tales which compared the approaches of prehistorians with those of contemporary ethnographers (Douglas 1972). If Flannery’s paper had an immediate impact, Douglas’s was rarely cited, perhaps because it dismayed so many of those at the conference. There could have been other reasons why it was overlooked. The article was short and lacked much direct reference to archaeological research. That is ironic, for a decade later her work was to exert a major influence on theorists in archaeology. Indeed, it played a growing role in their thinking through to her death in 2007 (Gosden 2004). Although she appreciated the attention, it was a role that she was reluctant to assume. The differences between Flannery’s paper and Douglas’s are not those between the disciplines of archaeology and anthropology. Flannery was well aware of ethnographic accounts of settlements in traditional societies. Rather, their articles reflect two different strands in twentieth-century thought. Flannery’s approach was influenced by functionalist anthropology and Douglas’s by structuralism. That is why her work provided a source of inspiration for those who became disenchanted with processual archaeology.

The Circular Ruins

This is not a book about a period or a place; it is about an idea. Why did so many people in prehistoric Europe build circular monuments? Why did they choose to live in circular houses, when other communities rejected them? Why was it that those who preferred to inhabit a world of rectangular dwellings so often buried their dead in round mounds and worshipped their gods in circular temples? The best way of introducing such questions is through a specific example. Certain monuments exert a special fascination. Beside the road at Uisneach in the Irish Republic is a signboard which makes some remarkable claims. This was the ‘site of the Celtic festival of Beltane’ and ‘an ancient place of assembly’. It was associated with ‘the Druidic fire cult’ and the ‘seat of Irish kings’. The notice makes a still more intriguing assertion, for the Hill of Uisneach was also the ‘sacred centre of Ireland in pagan times’. The archaeology of the hill is hardly less remarkable, and it is easy to see how it has suggested such ideas. Some interpretations of the site are based on its distinctive topography, and others on literary evidence (Schot 2006, 2011). The hill is an irregular plateau which rises out of an extensive plain. It is also at the junction of two different landscapes. To the east, there have been many discoveries dating from the prehistoric and early medieval periods. To the west, where the soil is less fertile, they are comparatively rare. Uisneach dominates the view from all directions. It also commands an extensive vista on every side. Indeed, Macalister and Praeger (1928) who studied its archaeology over eighty years ago published a map showing the land that can be seen from the hilltop. Although it is claimed that twenty Irish counties are represented, the area does not extend as far as the coast (Figure 1). Not all those regions can be observed from a single point. In order to appreciate the full extent of the view, it is necessary to move between a series of ancient monuments.

Turning to Stone

The starting point for this chapter is a work by the German artist Joseph Beuys. ‘7000 oaks’ is an installation which he inaugurated at Kassel, a city that had been damaged during the Second World War (Scholz 1986). Each tree was paired with a basalt stele which was quarried locally. In Beuys’s conception, the installation would change its character over time. For the first few years the standing stones would be the dominant feature, but they would become less conspicuous as the oaks grew to maturity. After that, there might be two very different outcomes. Either new trees would be planted as the old ones died— that was the artist’s plan—or a setting of monoliths would be all that remained with the stones themselves marking the positions of oaks that had disappeared. Beuys was concerned with regeneration in a way that was entirely appropriate in a war-damaged city where the oak trees would gradually replace a setting of rocks. His work was informed by his interest in ecology and played on a contrast between wood and stone which is equally relevant to archaeology. They are very different materials from one another, but both were used in prehistoric structures and employed in distinctive ways. Wood is an organic substance and eventually decays. Stone, on the other hand, is inorganic and for that reason it lasts a long time. The distinction is important in considering ancient architecture (Parker Pearson and Ramilsonina 1998). Of course, there were places in which only one of these materials was available, but there were others where the distinctive ways in which stone and wood were used are especially informative. Two examples illustrate the point. Neolithic houses in Northern Europe were timber constructions, but most of the tombs that accompanied them were made of local stone. In this case, the choice of building material suggests that these dwellings were thought to have a finite lifespan, whilst the tombs of their occupants would have a longer history. Similarly, the Neolithic longhouse at La Haute Mée in north-west France was built of wood but was accompanied by a granite menhir (Cassen et al. 1998).

Houses into Tombs

At the French site of Aillevans, not far from the border with Switzerland, there is a group of megalithic tombs (Pétrequin and Pinigre 1976). At first sight, these monuments conform to a wider tradition which is best represented at Sion on the Swiss side of the frontier, and at Aosta in Italy. In each case they feature massive stone cists associated with unburnt human bones (Mezzena 1998). These structures were sometimes located at one end of a low rubble platform or cairn, which could be either triangular or trapezoidal in plan. At Aosta and Sion they incorporated the remains of a series of anthropomorphic sculptures and, for that reason, the excavated evidence has played an important role in studies of statue menhirs. Dolmen 1 at Aillevans is equally remarkable but, in this case, the results of excavation have not attracted the attention they deserve (Pétrequin and Pinigre 1976: 325–49; Figure 25). In its original form, this structure consisted of a round mound six metres in diameter with a stone chamber and an antechamber. Again, it was associated with a quantity of disarticulated human bones. In a subsequent phase that construction was encased within a much larger trapezoidal cairn, seventeen metres in length. Although the circular monument was no longer a freestanding element, both its chamber and antechamber were retained. This was one of the latest megaliths in Europe, but sequences of this kind can be recognized at older tombs distributed across a much larger area. In its final phase, Dolmen 1 changed its character again. The chambered tomb was enclosed within a large wooden structure which had a similar outline to the cairn. The excavators concluded that it had been a roofed building. The stone chamber was located inside its eastern end, but the antechamber was left uncovered and acted as a kind of porch. Seen from a distance, the monument might have looked like a domestic dwelling. Indeed, Pétrequin and Pinigre (1976) specifically compare it with the well preserved buildings in the waterlogged Late Neolithic settlement at Clairvaux. According to their account, a megalithic tomb at Aillevans was almost completely concealed inside what appeared to be a house.

From Centre to Circumference

This book began with one site in Ireland and closes with another. The Loughcrew Hills in County Meath include at least twenty-five megalithic tombs, located on three summits along a prominent ridge. Many of them were investigated in the nineteenth century when Neolithic artefacts were found there. More recent work has been less extensive but features an analysis of the carved decoration inside these monuments, for the Loughcrew complex is one of the main concentrations of megalithic art in Europe (Shee Twohig 1981: 205–20). Early excavation in the westernmost group of monuments had an unexpected result, for Cairn H contained a remarkable collection of artefacts which must have been deposited three thousand years after the tomb was built. They included bronze and iron rings, glass beads, and over four thousand bone flakes (Conwell 1873). A new excavation took place in 1943, but its results only added to the confusion and, perhaps for that reason, they were not published for more than six decades (Raftery 2009). They seemed to show that the artefacts, which obviously date from the Iron Age, were directly associated with the construction of the monument; today it seems more likely that they were a secondary deposit. When they were introduced to the site, the tomb may have been rebuilt. One reason why the bone flakes attracted so much attention is that a small number of them—about a hundred and fifty in all— were decorated in the same style as Iron Age metalwork. Most of the patterns are curvilinear and show the special emphasis on circles and arcs that characterize ‘Celtic’ art (Raftery 1984: 251–63). This discovery illustrates a problem in Irish archaeology. A few stone tombs in other regions were decorated in a style that has been identified as either Neolithic or Iron Age (Shee Twohig 1981: 235–6), but in the case of the flakes from Loughcrew there is no such ambiguity. Not only do the incised patterns compare closely with those on metalwork, the decorated artefacts were associated with beads and rings dating to the end of the first millennium BC. Even so, two problems remain.