Fieldwalking: the 2000 season
General website directory
Twelve fields within the project area were walked over a period of three weeks in September and October 2000. These lie variously in the parishes of Deanshanger, Leckhampstead, Lillingstone Dayrell, Lillingstone Lovell, and Wicken. In addition a further two fields were walked in the parish of Stowe. Of these fourteen fields, the results from thirteen, relating to the pottery finds are reported below. The results from the settlement site of Lillingstone Dayrell are presented separately and can be access by clicking the hyperlink in the lefthand column.
The fields were sampled using the line walking method. Lines were set 15m apart and all finds collected and bagged in 20m stints. A narrow band, approximately 1.5m wide, along each line was scanned for surface finds. The sample thus represents 10% of the surface area of each field. The two fields in Stowe were walked using a line spacing of 20m and a stint length of 20m. Lines were orientated to fit the field shape. Lines were not orientated to the cardinal compass points. Whilst total collection would have been ideal, certain artefact types, such as modern brick, tile and field drain were not collected from each field. However, pottery of all periods, worked flint (tools and wasters) and all other cultural artefacts (hones, quernstones, architectural fragments etc.) were collected throughout.
All pottery collected from these fields was washed and marked and sent to Paul Blinkhorn for identification. All sherds were counted and weighed and organised into fabric types based on the Northamptonshire pottery type series and cross referenced to the Milton Keynes pottery type series. The assemblage has been subdivided into twenty-one categories based on fabric type. No attempt has been made at this stage, however, to identify all types of Romano-British wares – which is known to include Samian Ware together with coarsewares – or 18th and 19th century wares. These have been grouped together as single types under the codes F1000 and F1001. Most of the medieval sandy wares have also been grouped together as they are all broadly dateable to 1100-1400, and have very similar petrological compositions. They were also made at numerous unknown sources in the region and so it was considered that individual identification of such wares, which would be extremely time-consuming, would impart little information which would be of use at this stage of the project.
A description of these pottery types is represented in the table below, together with their Northamptonshire County type-series and Milton Keynes type-series codes where they exist. A date range for the production of this pottery is also included.
Results from Individual Fields
Each field is identified by its parish two-letter code followed by the field number. The grid reference is the approximate centre of each field and should allow identification from Ordnance Survey maps. Location maps and distribution maps will be added in due course.
An attempt has been made to describe the makeup of each assemblage and point towards historic landuse which might have produced such distributions. The empirical data – number of sherds, weight in grams – and the percentage within the assemblage for each fabric type is given in the table.
DE 1: [SP 738 413]
Field immediately north of Forest Farm. A field created in the mid-nineteenth century from woodland (pers. comm. landowner). The pottery scatter appears to suggest that there was no medieval activity on this parcel pointing to use as meadow or, more probably given location, woodland. Romano-British wares, however, might indicate that this land was formerly part of a more extensive field system at this earlier date.
DE 2: [SP 742 413]
Field immediately to the east of DE 1 exhibiting similar origins. Once again the assemblage appears to suggest a break in activity during the medieval period. Romano-British wares dominate this assemblage but their density would suggest agricultural activity rather than settlement.
LE 1: [SP 737 381]
Western part of this field only was walked focusing on an area which had previously produced Romano-British metalwork through metal detecting. The range and density of medieval fabrics appears to indicate manuring on open fields. The concentration of Romano-British wares in the western salient of the field strongly suggests the site of a minor settlement, possibly an isolated farmstead. A halo of Romano-British wares around round the major spread might indicate infields.
LE 2: [SP 732 383]
The field occupies a marginal site with steep drop off to a stream to the south-west. The field was known to have produced Romano-British metalwork from metal detectorist finds. The field is the current meeting point of four footpaths. Fieldwalking identified an artificial depression, possibly a dew pond or gravel quarry which may have acted as the focus for these paths. Earlier maps show a building to the south of the field whose position could be established by fieldwalking. The pottery assemblage strongly suggests that this field was formerly under the Leckhampstead open fields. Romano-British finds, all concentrated to the south-east corner of the field suggest the proximity of a settlement site, but this is likely to lie in the adjoining fields which are under pasture.
LL 1: [SP 711 403]
Field lies south of Lillingstone Lovell church and immediately south-west of the village earthworks. Fishponds can be identified immedately outside the field to the north-east. The field appears to lie on light soil with a high limestone content. This field produced a large range of medieval wares indicative, perhaps, of its proximity to the medieval settlement centre. A drop off of fabric ranges can be detected if this field (100m from settlement centre) is compared with the assemblages from LL 2 (800m), LL 6 (900m), and LL 7 (700m). Low density Romano-British wares are again present and attest to earlier agricultural activity.
LL 2: [SP 705 402]
This field lies at the parish boundary with Lillingstone Lovell. Nevertheless, the presence of medieval pottery indicates that this was taken into the open field system. While no strong concentration of Romano-British wares was identified in the field, their strong representation in the assemblage might suggest proximity to a settlement site.
LL 6: [SP 721 406]
The field lies to the eastern part of the parish towards Leckhampstead Wood. The assemblage is dominated by the local Potterspury wares suggesting that the strips of the open field were laid out post 1250. Romano-British wares present.
LL 7: [SP 709 409]
Lying to the north-west of the village centre, the low density of medieval pottery scatters across the field suggest open field agriculture at this date. Nearly half of the assemblage (by number) is Romano-British in date. A clear nucleus for this material was identified on the field suggesting a settlement site, possibly an isolated farmstead. The field lies close to an outcrop of gravel providing lighter soils than the predominant glacial boulder clays.
ST 1: [SP 678 387]
The small assemblage suggests Romano-British agricultural activity with a hiatus in the medieval period, suggesting pasture, arable for woodland use at this date. The latter, given its location away from any known medieval settlement centres and close to surviving stands of woodland would appear to be the more probable
ST 2: [SP 682 371]
Two areas of this field were walked, the first area along the south-western fence line and the second in the northernmost part of the field. The major concentration of medieval pottery is associated with a possible house platform in the north part of the field which may be associated with Lamport DMV. Elsewhere traces of ridge and furrow were visible and this strip farming is borne out by the low density pottery scatters found. Romano-British pottery present also.
WI 3: [752 370]
This field lies immediately to the west of Little Hill Farm. The field slopes south to the valley bottom of the Great Ouse. Surviving ridge and furrow to the south of the field proves that this field once lay within the complex field systems of Wicken and its dependent hamlets. The assemblage is dominated by Potterspury ware.
WI 4: [SP 749372]
This field lies immediately north of WI 3 and its origins and use can be considered to be the same. The field, however, lies along the line of the Buckingham road (A422) which appears to follow an historically established alignment. The discovery of a single sherd of St Neots Ware (1000-1200) may relate to the juxtaposition of field and road, or relate to the lighter alluvial terrace soils found in this part of the parish.
WI 5: [SP 745 392]
This field lies in the centre of the road loop immediately south of Wicken village. House platforms and ridge and furrow are visible in the unploughed north-eastern section of this loop. The strip alignment would continue over the walked area suggesting that this was formally part of the open field system. The medieval pottery assemblage and spread bares this out. The assemblage is restricted to the major pottery types and would not suggest close proximity to settlement alone. Romano-British finds might be part of a halo around a clearly important settlement site (located by metal detectorists) lying close to SP 744 390.
Whilst it would be dangerous to draw any firm conclusions from such a small sample, a number of trends appear to emerge from the work undertaken and might usefully be identified here. They should, however, be seen as provisional and liable to change. Nevertheless they form a working hypothesis which future research will seek to corroborate or dismiss.
Of great significance is the ubiquitous presence of Romano-British material. The identification of at least three farmstead sites and the clear opportunity to discover others in subsequent sorties will add greatly to our understanding of the Roman rural population away from the villa sites that are already known. Understanding how these isolated farmsteads relate to the villa estates may provide clues about the division and use of land into the early medieval period. Of no less significance has been the discovery of low-density Roman pottery scatters suggesting agricultural activity away from the settlement centres. Roman evidence has been found, for instance in the Deanshanger fields, in areas which are thought to have been tree-covered in the medieval period. Is this evidence for extensive Roman field systems over a larger area than was ever brought within the medieval open field system? Certainly the evidence points to post-Roman abandonment of some areas and subsequent woodland regeneration, woodland which was never subsequently cleared in the medieval period.
Of the medieval pottery, very little pre-conquest pottery has been recovered. The sample remains minute, and it would be dangerous to conclude simply that it is not present. Much more work will be required before a picture of early medieval activity will emerge. But it may be signficiant that the single sherd of St Neots Ware was found off the glacial boulder clays that dominate the geology of the area north of the River Great Ouse floodplain. Models elsewhere, notably in Northamptonshire, have suggested a withdrawal of the population after 400AD, away from the heavy clay soils of the Midland watersheds in favour of the lighter soils of the valley bottoms. Too much can be made of a single sherd, however, and it remains a possibility that further work on the clays will reveal evidence for early medieval occupation.
The post-conquest pottery is, by contrast, abundant. The assemblage is dominated by locally produced wares, notably Potterspury ware, and the not too distant Brill and Boarstall wares. The absence of medieval pottery on some fields points to landuses other than arable, since on those fields known to have formerly lain within the open field systems, a characteristic low density spread of pottery, deriving in all probability from the manuring of the land, can be readily identified. This early quantitative analysis is essential. Working from the known to the unknown, models establishing the relationship between former landuses and the artefacts recovered through fieldwalking, should allow fields of unknown former use to be categorized by comparison. Finally, it can be noted that there appears to be a correlation between the artefact range and proximity to settlement. In the case of the Lillingstone Lovell fields, it can be shown that the field lying closest to the settlement produced a far greater range of pottery than those fields lying some distance away. If this hypothesis can be shown to work elsewhere – the Stowe evidence exhibits similar traits, the Wicken evidence is less compelling – comparative study of number of types of pottery present may indicate proximity to, or distance from, settlement sites which await identification.