Great Wilbraham causewayed enclosure

Site

The Great Wilbraham enclosure is in Cambridgeshire, in the valley of the Little Wilbraham River, near the village of Great Wilbraham, east of Cambridge. Two concentric ditches, about 20 metres (66 ft) apart, each interrupted by multiple gaps, lie on a flat area on the north-facing side of the river valley, though the river has been diverted and no longer runs on its former path. The site is about 170 metres (560 ft) across, and encloses about 2 hectares (4.9 acres). Most causewayed enclosures are oval-shaped, but the two circuits at Great Wilbraham are almost perfectly circular, though many of the ditch sections (between the gaps) are themselves straight lines. Many enclosures were built on valley floors, usually positioned not far above the high water mark of the river, and Great Wilbraham fits this pattern. The site is slightly sloped towards the old river course, with the upper edge of the site at the edge of the higher ground. The enclosure is located on a sandy terrace, in Little Wilbraham Fen, a small fenland area connected to the larger Fens to the north by a stream.

Great Wilbraham was identified by J. K. St Joseph, who ran the aerial photography program at Cambridge University for many years, from cropmarks on an aerial photograph taken in July 1972, and included on a list of sixteen possible causewayed enclosures based on observed cropmarks published in 1975. The site was listed as a scheduled monument in 1976.

A prehistoric henge was tentatively identified in an adjacent field in the 1970s, but a more recent review considered it to be a naturally disturbed area of gravel.

Archaeological investigations

Clarke, Alexander & Kinnes, 1975–1976

David Clarke was one of the leading figures in New Archaeology, a heterogeneous movement to bring archaeology closer to the natural sciences, and conduct archaeological excavations in a scientific manner. He had some experience of fieldwork, but had not directed an excavation until 1975, when he started a five-year project to investigate the Great Wilbraham site. John Alexander, the other leader of the project, was a very experienced excavator, and joined Clarke on the project after the plan included using it as a training ground for students.

Great Wilbraham had several advantages as a project from Clarke's point of view: it was close to Cambridge, where he worked; it was expected to be rich in Neolithic material, which was the recent focus of Clarke's teaching work; and it was planned as a way to put into practice some of the theoretical ideas he had propounded during the previous decade. Great Wilbraham was the only known causewayed enclosure to include peat deposits, and Clarke hoped that Neolithic wood might be found as a result, since peat is one of the few natural environments in which wood can be preserved. According to a later review, the excavation "was intended to be an experiment in what Clarke called total archaeology" (italics in the original). "Total archaeology" is an approach which attempts to integrate information from "all of the disciplines capable of bringing understanding" to a site; accordingly, Clarke planned to include interdisciplinary analyses and an evaluation of the surrounding landscape and environment in the project.

A grant from the British Museum funded two weeks of excavation in September 1975; the prospect of finding preserved organic material was the reason for the museum's interest, and the grant came from money allocated for purchasing artefacts for their collections. Clarke assumed that the site was a settlement, despite the uncertainty about the function of causewayed enclosures, and did not mention the possibility of ritual uses in an application he made to the British Museum for funding for the 1976 season. Clarke died unexpectedly in the summer of 1976. He and Alexander never published their work, but the British Museum grant proposal records some of the details of the two weeks. A magnetometer survey was performed, and fieldwalking to recover surface finds, and a trench 80 by 2 m (262 by 7 ft) (GW I in the diagram) was excavated. Finds mentioned in the proposal included bones of cattle, sheep, pig, deer and wolf; worked wood; seeds; pottery sherds; and flints. A pollen column was taken—a column of soil, from a hole bored at the site, examined for changes in the frequency of pollen species, with each stratum of soil corresponding to a different period in the past. The column covered 5000 years of the peat, and demonstrated changes in the environment over the life of the site, such as the clearance of the site in Neolithic times and later pasture development. Two more trenches were dug the following year, directed by Alexander and Ian Kinnes.

Clarke's long-term plan for the site was ambitious, and could not have been achieved with the money requested in his proposal; a later review of the project identified shortcomings in the plan and execution as well. Excavation proceeded by horizontal spits (layers of fixed depth), rather than the accepted method of stratigraphic excavation (removing the material in each identifiable layer of soil as a unit). Although excavation by spits is considered an unsound approach, the work was in other ways advanced for its time: for example, a computer program was written to render the contour topography of the site in a 3D perspective.

Reanalysis by Evans et al. 2006

Some of the material from Clarke and Alexander's two seasons of excavation was preserved in an archive, which was stored in Cambridge and London. In 2006, the archive (less than half of which could be located by that time) was reviewed by a group led by Christopher Evans, a Cambridge University archaeologist, resulting in a paper that presented the results of the original dig, as well as a review of Clarke's ideas and how they were implemented in the field.

The cropmark plot used by Clarke to determine where to dig the trenches proved to be inaccurate, and this meant the trenches were not ideally positioned. The first trench (GW I in the diagram) was intended to be bisected by the circuit's ditch, but only just overlapped it at the southern end. The plot of the cropmarks was altered, probably by Clarke, to allow for the new information, but it was still incorrect in the east, and the trench in the northeast of the site, dug in 1976, was started (as trenches GW II a & d) further outside the line of the circuit than was expected, and the trench had to be extended (GW II c, e, & f) before the ditch was encountered. In both seasons this led to more investigation of the fenland peat deposits and less of the interior of the enclosure. The trench at the east side of the site (GW II in the diagram) was U-shaped, with two 45 m-long (148 ft) arms, connected at the southwestern end.

For all three trenches, the initial soil removal was done by machine until the base of the ploughsoil was reached, and the spoil heaps were examined for finds. Recording of finds below this level was poor within the trenches and inconsistent between them, so it was not always possible for the later analysis to be sure of the original stratigraphic position of the finds.

Findings

Two radiocarbon dates were obtained at the time of Clarke's investigation: an oak sample from the peat to the north of the enclosure was dated at 410–190 BC, and a sample of the peat itself to 7000 BC. Evans obtained two more dates from the outer ditch, submitting samples from the archive. A charred seed gave a date after 1950 AD, and was assumed to have been a later intrusion; the other, a charcoal fragment, was dated to 1750–1520 BC, in the Bronze Age. This is significantly later than the known dates for causewayed enclosures, and was also considered to be intrusive; these dates did not alter the conclusion that the enclosure was of mid-4th millennium BC date. In 2011, Great Wilbraham was included in Gathering Time, a project to reanalyse the radiocarbon dates of nearly 40 causewayed enclosures, but no further samples could be found from the excavations that were suitable for radiocarbon dating. The magnetometry survey performed for the 2006 reanalysis recorded evidence of the Neolithic ditches, along with many traces of later activity.

Fieldwalking recovered nearly 900 pieces of worked flint, and the various trenches brought the total to over 4,200. Items included a flint axe (which had been lost by the time the archive was reviewed in 2006), scrapers, blades, arrowheads, and hammerstones. The earliest estimated date was for a retouched blade which may have been from the Upper Palaeolithic (from about 40,000 to about 12,000 years ago). Other material came from many different periods, including some from as late as the first millennium BC. The majority of the flints came from gravel sources, and it is likely that the site's location was chosen by its Neolithic occupants because of the ready availability of flints from the nearby river-gravel terrace. A few artefacts made from non-local stone were identified, including fragments of stone axes from Cornwall (greenstone) and Cumberland (volcanic tuff). There was evidence to show that flints had been worked on site, rather than having been knapped elsewhere and brought to the area of the enclosure, but it appeared that the earliest stages of work on the flints had been done elsewhere, perhaps on the gravel terrace closer to the river, rather than near where the ditches were dug.

Of the 2,590 pottery sherds found, nearly half were Mildenhall ware, a form of Neolithic pottery found in southern England that dates from the mid-fourth millennium BC. The sherds were stored in the 1970s in labeled envelopes, identified by spit depth, trench, and grid square; since the excavation was by spits this meant that each envelope could contain material from more than one stratigraphic layer. The upper four layers included some Mildenhall material, but also pottery from the Late Bronze Age through to the present-day. All layers below this, which were all probably from within the various prehistoric ditches, contained Neolithic pieces, with a few Iron Age and Roman fragments. The outermost ditch found in trench GW II contained Roman material in all layers, indicating that this ditch dated from the Romano-British era (the mid-first century to the early fifth century AD).

Most of the animal bones from the Neolithic levels of the excavation were from cattle, with a few bones from sheep and pigs, and a single bone each from red deer, dog and horse. The cattle were probably killed on site as the finds included bones from every part of the skeleton. Clarke's proposal to the British Museum after the first year's excavation had mentioned finding a wolf bone, but in the reanalysis the only canid mentioned was a dog. A dog was apparently present at the enclosure during the Neolithic occupation, as one of the cattle bones showed clear signs of gnawing. An analysis of snail shells taken from the Neolithic levels found 36 Cepea nemoralis shells, most of which had visible bands. The banded form of this species is thought to be more common in grassland than woodland, implying that the area around the site had been cleared at the time the shells were deposited in the ditches. It may also be that the banded form was becoming more common over time during the Neolithic, and if this is the case the Great Wilbraham data would be anomalous, as more of the banded forms were found there than at other sites of a similar age.

Multiple cores were taken to the north of the site. Pollen analysis showed a change from a landscape dominated by pine to a grassland environment. Radiocarbon dates were obtained on organic material from the boreholes. The lowest (and hence oldest) sample, from the peat, was dated to 7940–7580 BC, a period in the early Mesolithic when birch and pine predominated. The upper sample was dated to 2440–2030 BC, at the boundary between the Neolithic and the Bronze Age, when woodland was declining and being replaced by agricultural land. The pollen spectra from these samples were in agreement with the expected environmental conditions for these dates.

The later review of the records from the 1970s concluded that both the ditches in GW I had been backfilled and recut, and that the dig had not found evidence of a bank associated with the ditch. Overall the site was found to be remarkably rich in finds, in comparison to the results of excavations at similar Neolithic enclosures.