Archaeology at Settrington, North Yorkshire
An archaeological field evaluation
Acknowledgements
I am indebted to a number of people without whose advice and assistance this project would not have been completed. Dave McLeod of English Heritage originally flagged the Settrington 'squares' as worthy of investigation, and Dr. Jon Finch brought them to my attention - with out them I would not have had the opportunity to work on such a project. Steve Roskams provided guidance and advice on numerous matters; The Birdstall Estate kindly agreed to my undertaking the work and both tenant farmers, Norman Lamb & Sons and David Bower Esq., were extremely supportive. Extra thanks is due to all of those who gave willingly of their free time to assist with the fieldwork itself: Frank Clough, Eric Christiansen and Luke Doyle, who made up the core of the fieldwork team; and to Nick Trustram-Eve, John Walker and Brian Rahn, for additional assistance in the field. With regards to the post-fieldwork processing thanks must go again to Frank Clough, John Walker and Rebecca Briscoe, who all assisted with cataloguing finds, and to John for generously providing materials for processing. Steve Dobson and Ben Gourley gave advice on the use of the geophysical equipment and processing of the data, and James Gerard, Mick Atha, Rob Collins, and Sven Schroeder all lent books and a friendly ear for discussion. Finally, I am extremely grateful to Rod MacKay who gave up an afternoon of his time to identify the pottery - his expertise has proved an invaluable source of knowledge to a pottery novice such as myself.
Index
1.0
Summary
1
2.0
The definition and purpose of a field evaluation
3
3.0
Geology
4
4.0
Aerial Photographic and evidence
4
5.0
Archaeological and historical background
5
6.0
Methodology
6
7.0
Geophysical analysis
7
8.0
Data processing
9
9.0
Results: Field Two
9
10.0
Results: Field Five
14
11.0
Finds
15
12.0
Conclusions
18
13.0
Postscript
20
14.0
Bibliography
20
List of figures
Fig. 1
Location maps
Fig. 2
Detailed map of research area
Fig. 3
Aerial photograph showing features in Field 2
Fig. 4
Plan showing the location of the survey area in field two
Fig. 5
Plan showing the location of the field-walking grid in field five
Fig. 6
A table showing the types of features normally located by gradiometry and resistivity
Fig. 7
Topographical survey results: Field two
Fig. 8
Iron pan staining in underlying gravels of Field two
Fig. 9
Magnetometry plot: Field two
Fig. 10
Resistivity plot: Field two
Fig. 11
Interpretation: Field two
Fig. 12
Topographical survey results: Field five
Fig. 13
Finds plot and underlying cropmarks: Field five
Fig. 14
Finds quantification table
Fig. 15
Numbers of identified pottery types from field five
Fig. 16
Distribution over time of the pottery from field five
Fig. 17
Examples of square temples from Europe
1.0 Summary
1.1 The author, assisted by colleagues from the Archaeology department at the University of York, carried out an archaeological survey of two fields within the parish of Settrington, North Yorkshire (SE810730, Fig. 1) between March and October 2004. The parishes of Settrington and neighbouring Scagglethorpe are rich in Iron age and Roman material but have been investigated very little in comparison with the adjacent Wolds area. Some crop mark features have been mapped by Cathy Stoertz in her publication "Ancient Landscapes of the Yorkshire Wolds" (1997), but the identification and date of many remains unclear. Information passed to the author via English Heritage and the Archaeology department concerning unidentified 'square' crop mark features in one field (SE829714, Fig. 2), originally dictated which area was to be evaluated. However after successful negotiations with landowners and the tenant farmers, it was possible to extend the research area to include all of the land farmed by Norman Lamb and David Bowers. In the case of the evaluation of the second field, the state of the land at the beginning of the project dictated its choice and type of work that could be undertaken. It was proposed that a formal programme of investigation be carried out in two fields involving topographical survey, field-walking and geophysical survey. This report summarizes the topographical, geological, archaeological, and historical setting of the site, the research methodology and presents the results of the evaluation.
Fig. 1 Location maps: (top) the location of sites mentioned in the text, and (bottom) the location of the research area in relation to its surrounding parishes and Malton/Norton.
Fig. 2 Research area: fields scheduled for survey (north) field2 and (east) field five; cropmark features previously mapped by Stoertz; limits of modern village of Settrington; limits of scheduling of shrunken medieval village (Cropmark information after Stoertz 1997, scheduling information from SMR records).
2.0 The definition and purpose of a field evaluation
2.1 An archaeological field evaluation is defined by the Institute of Field Archaeologists (IFA) as;
"...a limited programme of non-intrusive and/or intrusive fieldwork which determines the presence or absence of archaeological features, structures, deposits, artefacts or ecofacts within a specified area or site on land, inter-tidal zone or underwater. If such archaeological remains are present field evaluation defines their character, extent, quality and preservation, and enables an assessment of their worth in a local, regional, national or international context as appropriate" (IFA, 1994, rev.1999)
2.2 The purpose of a field evaluation is similarly defined by the IFA and is;
"...to gain information about the archaeological resource within a given area or site (including presence or absence, character, extent, date integrity, state of preservation and quality), in order to make an assessment of its merit in the appropriate context, leading to one or more of the following:
* the formulation of a strategy to ensure the recording, preservation or management of the resource
* the formulation of a strategy to initiate a threat to the archaeological resource
* the formulation of a proposal for further archaeological investigation within a programme of research (IFA, 1994, rev.1999)
3.0 Geology
3.1 The village of Settrington lies approximately 5km to the east of Malton, in the former district of the East Riding, now North Yorkshire. The two fields evaluated in this report are situated to the north-east and north of the modern village, along the minor Settrington to Scagglethorpe road "Forkers Lane". Both fields occupy relatively flat land, which slopes very gently from the eastern most edge of the Wolds chalk plateau towards the River Derwent. The underlying geology consists of alluvial sands and gravels.
4.0 Aerial photographic and excavation evidence
4.1 The cropmark features in field two (SE829714, Fig. 3) were first noticed by Addyman in the late 1970's. He described them as "insignificant", possibly failing to notice the arrangement and odd internal squares (pers. comm. David McLeod). Although recorded by the NMR and SMR, no archaeological investigation has taken place in this field, and interestingly the features were not mapped by Stoertz (1997) in her final publication.
4.2 The features in field two consist of seven square cropmarks, in two rows on a north-south alignment, four of the squares having at least one other internal concentric square. Although none of them cut each other, they do not appear to have any discernable relationship to each other and are not linked by a trackway or path. Furthermore, so far there is no evidence that they are related to cropmark features existing in adjacent fields, although it seems unlikely that they exist in isolation.
4.3 The features in field five consist of several lines on a n-e/s-e alignment (see Fig. 2). They were first mapped by Stoertz (1997), and are recorded on the SMR, but remain unidentified. They follow the same alignment as other features in adjacent fields to the south-west, so could be the remains of enclosure ditches, or trackways. At the time of going to print,. no aerial photographs for this field could be located for submission in this report.
4.4 The only excavation within the research area recorded by the SMR was in a field just north of field five (SE83717141), although there are numerous chance finds that have been made by the tenant farmers throughout the area. Signs of 4th century Roman occupation, including pottery, a coin of Constantinius II, a rammed chalk floor, and the footings of two walls, were discovered by Corder and Kirk in 1928 (SMR Record 3445, unpublished). The exact location of the excavation is somewhat unclear, but was in the region of SE837712. As it appeared to indicate large-scale occupation, there is every possibility that it is connected to the cropmark features to the south and north, and therefore to field five itself.
Fig. 3 An aerial photograph of field two showing 'square' features (Source: NMR (c) RCHME).
5.0 Archaeological and Historical Background
5.1 Although limited evidence of pre-Roman activity has been discovered within Malton and Norton, the development of an urban centre here was most probably the result of the arrival of the Roman army around 71AD. Although there is still debate as to the exact date of the construction of the first military fort at Malton, Deventio (or Delgovia) had grown into a flourishing town by the mid-2nd century. There is evidence of a substantial vicus outside of the fort, with stone houses, baths and shops; and an industrial area to the south (Norton) where pottery and metalwork were being made (Corder and Kirk 1928; Robinson 1978).
5.2 Within the parishes of Settrington and Scagglethorpe numerous sites dating from the Iron age and Roman periods have so far been identified from crop mark features and subsequently mapped by Stoertz (1997). As there are many other crop mark features which lie outside of the limits of the mapped area but within the two parishes (SMR records, unpublished), only the ones that have been mapped by Stoertz will be discussed here.
5.3 At the Broughs (SE823690) and at Scarlet Bault Bridge (SE82457040) a series of rectilinear enclosures, and rectangular/circular huts appear to be characteristic of Romano-British settlement. At the latter of these two sites Romano-British pottery has been discovered which adds weight to this identification. At Brough Hill (SE828695), the remains of a sub-divided ditched enclosure described by Ramm as an early Iron age farm, exist adjacent to a trackway which joined the road from North Grimston to Malton (1979, 76). Other examples of rectilinear enclosures adjoining trackways exist to the south-west of Settrington Grange (SE8369), at Hall Field (SE848683), and north of Settrington itself at (SE831715) and (SE838715). A number of undated linear features exist between these last two sites, whether or not they relate to trackways or boundaries remains unclear although some do seem to have a relationship with the features to the north of Settrington itself (SE831715). The last four sites listed are shown on the map in Fig. 2.
5.4 The line of one of the Roman roads from Malton (Margary no. 812) can be seen cutting across fields to the west of the village of Settrington (Fig. 2). Its precise route remains unknown but it is assumed that it lead across the Wolds to West Lutton and might have then joined with another route on to Bridlington (Margary 1973, 418-25).
5.5 Within Scagglethorpe parish there are several complexes of square and rectilinear enclosures adjacent to trackways such as at Brow Farm (SE848717), and The Marrs (SE822738). None of these have so far been conclusively dated.
6.0 Methodology
6.1 The archaeological evaluation consisted of topographical survey, field-walking and geophysical survey. All three types of survey were undertaken in field two with the aim of defining the nature and date of the crop mark features within it. In field five, only the first two types of survey were undertaken due to time constraints. Although the crop mark features in field five were not geophysically surveyed, it was envisaged that any finds collected would give an indication of the nature of any underlying archaeology.
6.2 A Lecia 705 TST was used for all survey work. It was not possible to locate the nearest benchmark, so an arbitrary height of 50m above sea level was assigned to each field. Although the land within the research area is relatively flat, the boundaries and topography of both fields were surveyed to provide more information on any micro-topology which might be present, and to provide base data to which grid positions and find positions could be plotted.
6.3 Field-walking and artefact retrieval was identified as one of the main methods of collecting evidence for identification and dating in both fields. A decision was made to conduct the field-walking using a grid system. Even though this method is more time consuming for an individual to undertake, it was chosen because it can help identify scatters of artefacts, and therefore assist in identifying occupation from a simply wider scatter of finds which could have derived from manuring (Roskams 2001, 46) .
6.4 In field five, 10m by 10m grids were plotted using tape measures and ranging rods along the boundary of the field on the south side (Fig. 4).
6.5 It was originally planned to use the same level of resolution in field two, but time pressure during the project and poor ground conditions resulted in the use of the 20 x 20m grid originally laid out for geophysical survey instead (Fig. 5).
6.6 Squares in both fields were identified alphabetically along their east-west axis and numerically along their north-south axis.
6.7 Field five was field-walked between March and September 2004. Until mid-July the field was lying fallow, although the remaining crop did obscure the ground surface.
6.8 Field two was field-walked between September and October 2004. The crop was harvested in early September but the soil was not ploughed, so the remaining stubble and new growth did obscure the ground quite considerably.
Fig. 4 The position of the field-walking grid in field five. The position of the station used for survey is marked in red.
Fig. 5 The position of the field-walking and geophysical survey grid in field two. The position of the stations used for survey are marked in red.
7.0 Geophysical analysis
7.1 The table in Fig. 6 shows the types of features normally identified by gradiometry and resistivity, and the relative strengths and weaknesses of each method.
7.2 As the nature of the features in field two had only been postulated from the aerial photographs, two types of survey were undertaken in order to provide as much data as possible. Gradiometry and resistivity were chosen for several reasons most influential of which being that the equipment was available from the University of York, and that the author had prior experience with both methods. It was also expected that at least one of the methods would provide good results on the alluvial gravels in the survey area (Clarke 2001).
7.3 The grid in field two was located by sight with reference to the only two features visible in both the field and aerial photograph, namely two trees on the north and south sides of field two. The grid was laid out by hand and using a theodolite roughly parallel with the southern most edge. The survey area was divided into 36, 20 x 20m squares, the size of the squares having been chosen after consultation with colleagues and representing what is generally accepted as a standard size for geophysical survey (Banning 2002). The squares were labelled A to F along the southern axis, and 1 to 6 along the western axis. During the last day of the survey the grid was extended by one square in rows B and C.
7.4 During the gradiometry survey each square was surveyed by making parallel traverses 1.0m apart in a south-north direction. The instrument used was a Geoscan FM36 fluxgate gradiometer. A built-in data logger automatically measured magnetic fluctuation along each traverse at 0.25 meter intervals.
7.5 During the resistivity survey each square was surveyed by making zigzag traverses 1.0m apart in a northerly then southerly direction. The instrument used was a Geoscan RM15 resistivity meter. A built-in data logger automatically measured the resistance readings along each traverse at 0.50 meter intervals.
7.6 The resulting data was stored in the internal memory of the gradiometer/resistivity meter and was periodically downloaded to a portable computer in the field for storage and verification.
Survey Type
Features normally identified
Strengths and weaknesses
Resistivity
Walls and rubble spreads
Made surfaces such as yards
Metalled roads and trackways
Stone coffins or cists
Large pits and slots >O.5m
Ditches
Drains and gulleys
Occasionally graves
Best results obtained on well-drained sub-soils such as gravel and chalk
Depth limit is dependant on probe arrangement - greater separation increases depth penetration
Slow
Excessive moisture can disrupt machine
Magnetometry
Ditches >O.5m diameter
Pits >O.5m
Pottery and tile kilns
Hearths and ovens
Ferrous debris including some slags, Briquetage, pottery waster, bricks and tile
Burnt material & fired stones Palaeochannels and other fluvial /geomorphological features
Larger posthole, slots & gulleys
Walls
Large areas can be covered quickly
Anomalies are usually easy to interpret
Operator must be 'metal-free'
Non-ferrous metals (gold, silver, copper etc.,) cannot be detected
Unlikely to identify features below lm deep
Fig. 6 The types of features normally identified by gradiometry and resistivity, and the relative strengths and weaknesses of each method (after Gaffney et. al. 2002).
8.0 Data processing
8.1 Upon completion of the survey, the geophysical data was transferred from the portable computer to a desktop PC and interrogated using Geoplot 3.0. Considerable amounts of data processing were required to remove high levels of background noise and differences in the grayscale between squares. Both sets of data were 'de-spiked', 'edge-matched', and 'interpolated' before being exported as graphics files into Adobe Photoshop. AutoCAD 2004 was used to digitize data for the production of the maps and plans, and Terrain 4.0 and ArchView 3.0 were used to produce the topographical plans. MS Access was used for the production of the finds databases.
9.0 Results - Field two
9.1 The results of the topographical survey (Fig. 7) show that the field is relatively flat.
9.2 The assemblage from field two consists of twenty-six mostly small, abraded shreds of pottery, two fragments of metal and charcoal and one fragment of glass. With the exception of one sherd (29), all of the finds can be dated to the medieval period or later. The finds are quantified in section 4.0.
9.3 Twenty-four squares were surveyed using the gradiometer (A1-2, B1-2, C1-2, D1-6, E1-6 & F1-6) and twenty-six squares using the resistivity meter (A1-5, B1-7 & C1-7). Due to the consistently poor results recorded during the gradiometry survey, a decision was made to cease surveying before the grid was finished. Additionally, the eastern most squares were not surveyed with the resistivity meter due to time constraints.
Fig. 7 The results of the topographical survey of field two.
9.4 It is clear from the geophysical plots that the nature of the features, and/or background geological conditions in field two were not favourable for gradiometry. The resulting plot (Fig. 9) appears to be devoid of archaeological features. A handful of anomalies towards the south are apparent, but it has not been possible to identify them, and they may well be spikes caused by ferrous material.
9.5 It is assumed that staining seen in the underlying gravels (Fig. 8) at the junction of the edge of the field and Settrington Beck, is caused by iron pan and might have caused interference during the gradiometry survey. Although no soil samples were taken, Dr. Alan Hall at the Archaeology Department of the University of York has suggested that these deposits might have been caused by soliflucted chalk and wind-blown sand.
9.6 The resistivity survey was far more successful in locating archaeological features. In order to emphasise the anomalies shown in the resistivity plot (Fig. 10), an overlay was used to mark the features. These appear separately in an interpretation diagram (Fig. 11). Colour coding has been used to distinguish between different types of features. Additionally, individual anomalies are accompanied by a numeric annotation and are discussed below.
9.7 Low resistance linear readings are represented on the greyscale plot in pale grey and are marked on the interpretation diagram in red, green and grey. High resistance readings are represented on the greyscale plot in black, but have not been marked on the interpretation diagram as they are probably a result of differing levels of conductivity in underlying soils.
9.8 A number of low resistance linear anomalies [1] were identified running across the survey area on an east-west axis. They align well with the boundaries of the field and most probably represent ridge and furrow - this identification was confirmed by the tenant farmer. A further low resistance anomaly [2] was identified running southwards from the north edge of the field, intersecting with features [3, 4, 5]. As it appears to cut them, it is assumed that this is a later feature, although what it represents is unclear.
9.9 The largest of the square features measures approximately 20m across [4], and the smallest 10m across [2]. Four of the features clearly contain at least one other internal concentric square [3, 4, 7, 8]. The largest feature [4] appears to contain another feature which could be a smaller square or a pit. Feature [5] contains either a further square disturbed by ploughing or an irregular shaped pit.
9.10 Features [3-9] appear to be formed by ditches, rather than walls. However, they could also be the robbed out foundation ditches of timber or stone buildings. The lack of obvious entrances is puzzling, but could be explained if the features are foundation trenches as suggested.
9.11 With regards to the relationship of the features to each other, there is simply not enough evidence available to make anything more than a suggestion at present, although it must be noted that the features all appear to respect one another whilst having a somewhat unclear relationship to one another. As squares [7 & 8] share a boundary it is possible that they might be contemporary with each other. Therefore is might further be assumed that features [6 & 9] form part of a group with [5 & 6] as they appear to share a similar orientation. If this is the case then the similar alignment of features [4 & 5], along with the adjacent position of [3], might also make these a group.
Fig. 9 Magnetometry plot: field 2. Although a handful of small anomalies are apparent to the south of the plot, they do not appear to be the result of archaeological activity. The dark lines in the fourth and fifth rows are probably the result of operator error.
Fig. 10 Resistivity plot: field 2. The square features seen on the aerial photograph are now visible, as are several other non-archaeological features. The dark area to the bottom of the plot was caused by interference from tree roots.
Fig. 11 Graphical summary of significant anomalies detected through resistivity survey.
10.0 Results - Field five
10.1 The results of the topographical survey (Fig. 12) show that the field is relatively flat.
Fig. 12 The results of the topographical survey of field two.
10.2 1034 finds were recovered from the field consisting of pottery, bone, glass, metal, charcoal, slate and flint fragments. These are discussed in more detail in section 11.0
10.3 During July 2004 the field was manured. Shortly afterwards whilst topographical survey was being undertaken the partially rotted and disarticulated carcasses of two sheep were noticed. It is therefore highly probable that the majority of animal bone recovered prior to this date is of modern origin. Due to the amount of bone that was scattered across the field, a decision was taken not to recover any more bone during the remainder of the field-walking phase.
10.4 The distribution of the finds recovered is shown in Fig. 13. Interestingly, rather than the finds being spread across the field as a result of plough sherd movement, the majority are concentrated in an area approximately 50 x 60m. Although this distribution does not appear to have any strong relationship to the crop mark features, it could well be indicative of further underlying archaeologically activity.
Fig. 13 Schematic diagram showing the distributions of the finds recovered from field five, and crop mark features recorded by Stoertz (crop marks after Stoertz 1997 map 1).
11.0 Finds
11.1 The finds were bagged in the field and annotated with the site code, field number and grid reference. Once returned to York they were washed, numbered, separated into artefact types and quantified by grid reference, weight and count. The material is presented in Fig. 14 and discussed below.
Artefact
Field two
Field five
No
Weight (g)
No
Weight (g)
Pottery
26
367
984
14,916.5
Bone
11
187
Glass
1
42
10
150
Stone
10
3533
Metal
2
9.5
7
287.5
Charcoal
2
2
7
26.5
Slate
3
45.5
Jet
1
7.5
Flint
1
7.5
Totals
32
420.50
1034
19,161
Fig. 14 Quantification of finds from fields two and five.
11.2 Field two
11.3 With the exception of one small sherd of calcite-gritted Huntcliffe ware (29), all of the finds can be dated to the medieval period or later.
11.4 Field five
11.5 The chart in Fig. 15 gives a graphic representation of the quantities of pottery types recovered from field five. The chart in Fig. 16 shows the dating of the classes of pottery so far identified - those where the identification is not confirmed are in italics. 4th century pottery dominates the assemblage although smaller amounts of higher status 2nd century pottery also exist.
11.6 Amphorae
11.7 The amphora sherds are all Dressel 20 (106, 393-4, 953-7, 1044). Although a substantial fragment of handle exists, it has not been possible to identify the form of vessel (or vessels) represented.
11.8 Mortaria
11.9 One sherd has been positively identified as Crambeck ware (575). Of the remaining four sherds, one is possibly Mancetter-Hartshill (1101), one possibly Corbridge (785), and two remain unidentified (3, 382).
11.10 Imported Wares
11.11 The samian sherds consist of six fragments of Central Gaulish (5, 128, 515, 784, 1102-3) and two fragments of Eastern Gaulish (520, 657). Although four of the sherds are rims, they are too small to identify the form of vessels represented.
Fig. 15 Quantities of pottery recovered from field five, by type.
11.12 Romano-British Fine wares
11.13 One fragment (12) and one possible fragment of Nene Valley ware (656) were recovered. The former is probably part of a cup.
11.14 Romano-British Course wares
11.15 There are two forms of Huntcliffe ware present in the assemblage: the majority is calcite-gritted (518 sherds), and the smaller amount is a plainer fabric with no inclusions (20 sherds).
11.16 The Grey wares are mostly reduced Crambeck (159 sherds). There are a few fragments of another as yet unidentified Grey ware (13 sherds) of a different fabric.
11.17 Other unidentified wares
11.18 Two further types of coarse ware exist. Both fabrics are dark grey to black in colour, but one fabric is wheel-thrown and the other hand-made. Neither fabric is similar to black burnished ware, nor similar in style to Huntcliffe ware.
11.19 The oxidised wares represent a group of sherds that may include fragments of Roman material, but otherwise date from the 10th century onwards.
11.20 Stone, Brick and Tile
11.21 Three fragments of rotary quern stone (389, 794, 1004) were recovered along with two pieces of worked stone (891, 997); all probably Iron Age or Roman in date. One piece of tile (986) was recovered, which most probably dates to the post-medieval period.
11.22 Glass
11.23 Of the glass fragments recovered only two may be Roman (138, 687), the rest all being post-medieval or later.
11.24 Animal remains
11.25 As discussed in Part I, the discovery of the partially disarticulated carcasses of two sheep during fieldwalking means that there is a strong possibility that the animal remains recovered are of modern origin. Of the 11 fragments recovered, 2 were teeth and 9 were bone. Of these only one (620) showed evidence of having been butchered.
11.26 Other finds
11.27 The charcoal (9) and slate (3) fragments vary in size, but are few in number and are therefore not significant. One piece of worked flint was found (621), although it was not possible to date it, as it is too small. One piece of jet was also found; it appears to have been worked but is now too damaged to date accurately - it could possibly be Roman or Victorian.
0
50
100
200
300
400
Central Gaulish Samian
Corbridge Mortaria
Crambeck
Dressel 20 Amphorae
Eastern Gaulish Samian
Huntcliffe
Mancetter-Hartshill
Mancetter-Hartshill Mortaria
Nene Valley colour-coated
0
50
100
200
300
400
Fig. 16 Dating of classes of identified pottery from field five (unconfirmed fabrics in italics) (after Tyers 1996)
12.0 Conclusions
12.1 Field Two
12.2 Although it now seems very likely that we are dealing with ditches, or the robbed-out foundation trenches of buildings in field two, we are still lacking vital information which would explain their function, status and date. The relative lack of artefactual material recovered, and the assumption that what was recovered is the result of manuring rather than archaeological activity, means that no firm hypothesis on their identity or dating can be constructed at present. However, based on the study of the morphological features of the squares, it is possible to suggest two theories for consideration.
The first possible theory is that the cropmarks in field two represent the remains of Iron Age Arras barrow burials. The Arras burials identified by Stead (1979) occur across the whole of the former area of the East Riding (including the parish of Settrington). They can occur in large cemeteries, small groups, or in isolation and are usually 6-8m in diameter, although the largest example so far discovered was 10-12m in diameter. Modern disturbance usually means that there is seldom any trace of a mound, but the feature is nearly always represented by a grave at the centre of a square-plan ditch. The burials themselves are nearly always located on the old ground surface, which means that all traces of them can be removed by ploughing, but which in turn usually results in bringing considerable quantities of bone, pottery and metal work to the surface. In terms of the recognition of barrows from aerial photographs, one would expect not to see such a well defined square plan ditch as is evident in field two, as the corners of the barrows are easily eroded (Stead 1979, 7-39). Although the features in field two lie within an area with many Arras barrows, their size and shape, lack of an obvious internal grave-type pit, combined with the lack of artefactual evidence, all seems to indicate that they are not burial barrows. However, without further investigation we cannot totally discount this theory altogether.
The second possible theory is that the squares are the remains of wooden or stone Celtic-Romano temples. Temples dating to the Iron Age and Roman periods are common in the UK and Europe, and are found in both urban and rural contexts (Rodwell 1980, 369). They come in many different shapes and sizes including rectangular, circular, and square. However, they rarely appear in groups of more than 2, which sheds some doubt on the features in field two representing temples, unless they are not all contemporary with each other. There are several examples of groups of square temples from Europe (Fig. 17) which appear to be similar in size, and to a certain extent layout, to the Settrington squares.
There are some obvious, but important flaws in the 'temples' theory that need further investigation. If they are religious sites, then there should be at least some artefactual evidence related to ritual activity (such as votive offerings), but at present there appears to be none. Furthermore, why are there so many 'temples' at one site? Although it is widely accepted that Iron age and Roman cultures worshiped many gods (Henig 1984), it would be unusual to say the least, to find a complex in a rural location dedicated to more than one. Although the numbers of 'temples' could be explained by chronological development, the fact that they appear to respect each other means that we cannot be sure if they were built as one complex, or individually over a period of time.
Fig. 17 Examples of square temples from Europe (not to scale): left to right, Mont-de-Sene; Matagne-la-Petite and Le Catelier-de-Criquebeuf (after Rodwell Figs. 17.20, 17.5 and 17.9, 1981)
However pleasing either of the theories suggested appear to be (and it is tempting to think that they are temples), none of the them can be substantiated without further archaeological work. Until such time as this can be undertaken the nature of the Settrington 'squares' will remain a mystery.
12.3 Field Five
12.4 The pottery recovered form field five indicates a date range of 2nd to 4th century. However, we must remain cautious about attributing too early a date for occupation. Samian is an expensive item during the Roman period and is therefore likely to have been well looked after and given from one generation to the next. Likewise, amphorae are very hard wearing and could have lasted several generations. Therefore the 2nd century material might well be residual, and have little to do with the actual date of occupation in the site. As the greater majority of the pottery is 4th century in date, this swings the balance in favour of this period alone.
The composition of the assemblage is such that it might be associated with a farmstead where food production (quern stones/mortaria), storage and cooking (Huntcliffe and Crambeck wares/amphorae) and consumption (Samian) are taking place. However, the relative lack of worked stone, and total lack of plaster and tesserae indicate that this is unlikely to be a villa site. As the majority of finds are distributed in an area (50 x 60m), which is itself no more than 40m from the line of the Roman road from Malton, we might assume that this relates to a building rather than a pit. Even though the field-walking grid is a 'construction' imposed on the field itself, it does seem unlikely that a feature such as a pit would cause this size and shape of distribution. An excavation undertaken by Corder and Kirk in 1928 revealed 4th century pottery, a coin of Constantinius II, a rammed chalk floor, and the footings of two walls just to the north of this field (SMR Record 3445, unpublished), so it seems acceptable to make the assumption that we are dealing with large-scale Romano-British occupation over this entire area. The activity infield five may well be linked to the rectilinear enclosures across to the west side of Forkers Lane, or as suggested might be evidence of the existence of an undiscovered building. If there is a building (or complex of buildings) in field five, then they could be part of a native farmstead, or even possibly a taverna or mansio. Unfortunately, as none of the cropmarks in the field have a strong relationship to the pottery distributions, it is not possible to do any more than speculate at present.
Further work is required to clarify the nature of the archaeological activity in field five and to identify the pottery fabrics which remain unclassified, so as to confirm the date range of occupation.
13.0 Postscript
13.1 Further work is scheduled for 2005 which will hopefully provide the information needed to substantiate the theories suggested in this report. In field two, field-walking will be undertaken after ploughing, when it is hoped more artefactual material might become apparent. In field five, a geophysical survey of the area where the highest artefact distributions are present will hopefully show if there is any archaeological activity that relates to the finds distributions.
14.0 Bibliography
Banning, E. B. (2002) Archaeological Survey. New York: Kluwer Academic / Plenum Publishers
Clarke, A. (2001) Seeing Beneath the Soil. Prospecting methods in archaeology. Third Edition. London: Routledge.
Corder, P. and Kirk, J. L. (1928) Roman Malton: a Yorkshire fortress and its neighbourhood. Antiquity, 128, II.
Gaffney, C. F., Gater, J. and Ovenden, S. (2002) The use of geophysical techniques in archaeological evaluations. Reading: Institute of Field Archaeologists.
Henig, M (1984) Religion in Roman Britain. London: BT Batsford.
Institute of Field Archaeologists (IFA), 1994 (rev. 1999) Standard and Guidance for Archaeological Field Evaluations. Reading.
Margary, I, D. (1973) Roman Roads in Britain. Third Edition. London: John Baker
Ramm, H. (1978) The Parisi. Gloucester: Duckworth.
Robinson, J. F. (1978) The archaeology of Malton and Norton. Leeds: Yorkshire Archaeological Society.
Rodwell, W. (ed) (1981) Temples, churches and religious buildings in Roman Britain. BAR British Series 77.
Roskams, S. (2001) Excavation. Cambridge University Press: Cambridge.
Stead, I. M. (1979) The Arras Culture. Yorkshire Philosophical Society: York.
Tyers, P. (1996) Pottery in Roman Britain. London: Routledge.
Stoertz, C. (1997) Ancient Landscapes of the Yorkshire Wolds. Swindon: RCHME.
Unpublished sources
SMR, Scheduled Monuments Record, County Hall, Northallerton, North Yorkshire.
NMR, National Monuments Record, English Heritage, Swindon (RCHME Aerial Photographs)
Claire Feldkamp October 2004
i
12
21