at Yarnton, which I first mentioned in Seahenge, my autobiographical book on Bronze Age religion.35 Yarnton is that rare thing, a large-scale excavation which also happens to be a thoroughgoing landscape project.
Gill recently published her report on the Saxon period at Yarnton.36 Yarnton lies on the very edge of the gravel terrace, on land that would have been just outside the limit of the river’s winter floodplain. This location ‘at the edge’, as it were, was deliberately chosen both in the Bronze and Iron Ages, as in Saxon and medieval times. The Thames floodplain is dressed with a thin layer of flood clay, known as alluvium, every time the river is in spate. This material is very fine, and is rich in natural fertilisers. As a result the grass gets away to a very good start in the spring, and gives young lambs and calves what farmers refer to as ‘a good bite’. In the past, before we learned how to ignore traditional ways of doing things, this land was never ploughed. Today it is, and surprise, surprise, the soil washes away and clogs up streams and drainage dykes.
Good arable land was to be found on the light, well-drained gravel soils around the villages that clustered at the edge of the floodplain, and up the gentle slopes of the valley sides to the north of them. Beyond this arable belt was another landscape of rough pasture, woodland and scrub. This was where most of the building material for houses and fences was grown. It was also a good ‘emergency reserve’ of fodder in wintertime and in very dry summers. Most grazing animals are quite happy to browse (in other words to eat the leaves of trees and shrubs) if grazing is running short – in fact my sheep prefer browsing the young shoots of hawthorn hedges around the fields to the rich Fenland grass at their feet. So Yarnton and the villages around were carefully positioned not just to be safe from flooding, but to exploit their natural surroundings as efficiently as possible.
Yarnton has produced huge quantities of Iron Age and Roman pottery. I can remember being in the finds shed surrounded by trays and trays of pottery stacked up to dry. This is what one would expect of an Iron Age site in the Thames Valley. But archaeology isn’t always predictable. As a general rule of thumb, in areas where pottery is common in the Iron Age, it remains popular in post-Roman times. The converse also applies, so in places like the west Midlands around Cirencester, the Iron Age is almost aceramic – presumably people used basketry and wooden bowls instead – then the usual types of semi-mass-produced Romano-British pottery appear in the Roman period. At the close of the Roman period people revert to their old ways and pottery vanishes from the scene. But at Yarnton, despite the richness of its Iron Age pottery, post-Roman sherds are rare. This is particularly odd given the size and seeming prosperity of the Saxon settlements. These were not pokey, subsistence-style farmsteads clinging onto a blasted hillside somewhere in the mountains, but a thriving and vigorous set of expanding communities in the heart of the pastoral lushness that is the Thames Valley. So what was going on?
Yarnton and other sites around it revealed Early Middle and Late Saxon settlements, yet the total number of pottery fragments found there was just 117, weighing a fraction more than two packets of sugar (actually 2.192 kg). I would have expected something more like a quarter of a tonne, comprising anything from 10–50,000 sherds. We must assume that there isn’t a simple physical reason for the rarity of potsherds, like very acid ground water, which can dissolve shell and other calcareous components of the pottery. So the answer has to be cultural. For some reason the people living in Saxon Yarnton didn’t make or use much pottery. To an archaeologist, and probably only to an archaeologist, that might seem odd. But it isn’t. In fact the decision not to use pottery is perfectly rational, if unusual, because good containers can be made from wood or basketry, and of course birch bark, which we happen to know was used in the Thames Valley in prehistoric times. These organic containers will only survive in waterlogged conditions, where the air needed for fungi and bacteria to break down organic materials is absent. Such containers are durable, they don’t shatter when dropped, and they don’t require complicated technology to make. In fact it’s interesting to note that two probable Iron Age pottery kilns are known from Yarnton. But in Saxon times they wanted none of it, and preferred instead to use local materials such as willow and birch bark that would have grown plentifully nearby. To me, this seems a perfectly reasonable choice.
The pottery that did manage to survive was very interesting, and Gill’s pottery specialist, Paul Blinkhorn, made the most of what little he had to work on.37 Perhaps the most remarkable result actually took very little analysis: the features of the Saxon settlement contained more Iron Age and Roman pottery (3.5 kg) than Saxon. This was material that was lying around on the surface when the Saxon houses were built, ditches were dug and so forth, a strange inversion of what one might normally expect. During the life of the settlement this earlier debris slipped into the various ditches, pits, wells and post-holes, along with fragments of the few contemporary Saxon pots.
The Saxon pottery included nine sherds of Ipswich Ware, and Yarnton is so far the most westerly site to have revealed this early form of mass-produced pottery. Paul Blinkhorn has pioneered a sophisticated form of chemical analysis, known as ICP-AES (or Inductively-Coupled Plasma Atomic Emission Spectroscopy), which has shown that all the Ipswich Ware we currently know about was made from clays occurring in Ipswich. It can therefore be considered a very reliable indicator of Middle Saxon trade. Ipswich Ware was made between about 720 and 850, but was not traded outside East Anglia until about 730. At the height of its popularity it was traded as far north as York and as far south as Kent. Huge quantities have been found in the emporium at London, and it is assumed that Yarnton would have been within its trading area. Recently other sites in the Thames Valley have revealed Ipswich Ware, so it would seem reasonable to suggest that Yarnton was part of this trading network. As at other non-East Anglian sites, the Ipswich Ware from Yarnton included just a few vessels (around seven), the majority of which were large jars that probably originally contained some traded commodity such as salt or oil. It is always a problem, when it comes to pottery, to determine whether the vessels were bought for themselves, or for what they may once have contained.
FIG 6 The Middle Saxon settlement in the Thames Valley at Yarnton, Oxfordshire (AD 700–900).
There have been relatively few recent excavations of Middle Saxon settlements, so Gill Hey’s work at Yarnton is most important. A series of radiocarbon dates has indicated that occupation began in the late seventh century and lasted through the ninth (say 700–900). There was occupation in and around Yarnton in the Early Saxon period, but Gill is keen to emphasise the contrast between that and what followed: ‘The contrast between the Early and Middle Saxon settlements at Yarnton is strong. There are radical differences in the size of the settlement area, in the degree of organisation within it, in building type and in the variety of structural remains and other features … but the coherence of the [Middle Saxon] settlement plan suggests that it was organised on this large scale from the beginning.’38
The settlement involved, for the first time, substantial timber buildings in addition to the more traditional building form of the Early Saxon period, the SFB or sunken feature building, which essentially consisted of a single-storey structure with a wooden floor over a cellar-like space beneath.39 This space probably served to keep the floor dry and would have helped the floor joists resist wet-rot.
The large buildings were a series of post-built rectangular halls and their outbuildings, including an impressive circular poultry house. These buildings were set within enclosures that were defined by ditches, and perhaps by hedges too. Gill notes that from the eighth to the tenth centuries the use of space within the settlement became increasingly formalised. She suggests that this may have been a reflection of two things: greater social control and authority, coupled with a growing shortage of land.
The
