Three types of sources were searched in order to compile a database of figurines from Britain: publications, the Portable Antiquities Scheme database and museum collections. As well as the catalogues by Green, Pitts, Hutchinson, Lindgren, and Foster and the VCH, a thorough search of monographs and national and county journals was undertaken. Since the catalogue authors had themselves conducted literature searches, and any references they cite were consulted, it was thought that little could be gained by searching complete runs of journals published before the 1970s; thus only journals published from 1970 onwards were examined.
The Portable Antiquities Scheme database was searched for metal figurines of Iron Age and Roman date. Following the search of publications and the PAS database, museum collections were visited in order to see pieces first-hand and record any previously unpublished figurines. Major museum collections throughout Britain were examined (Table 1), although several were inaccessible owing to refurbishment (Ashmolean, Exeter and Somerset County Museum). Wherever possible the objects were examined, recorded and photographed, although occasionally it was not possible to gain direct access to figurines on display. All of the information collected was then entered onto an Access database. Photographs taken by the author have also been attached to these records.
| Museum | No. of figurines |
|---|---|
| British Museum | 107 |
| National Museum of Wales, Cardiff | 13 |
| Ashmolean Museum | 3 |
| Roman Baths Museum, Bath | 4 |
| Bristol City Museum | 3 (plus 6 copies) |
| National Roman Legion Museum, Caerleon | 6 |
| University of Cambridge Museum | 16 |
| Tullie House Museum, Carlisle | 14 |
| Chesters Museum | 3 |
| Colchester Castle Museum | 43 |
| Corbridge Museum | 5 |
| Corinium Museum, Cirencester | 12 |
| Gloucester City Museum | 2 |
| Guildford Museum | 9 |
| Ipswich Museum | 10 |
| Jewry Wall Museum, Leicester | 8 |
| Museum of London | 63 |
| Great North Museum, Newcastle | 7 |
| Newport Museum, Isle of Wight | 1 |
| Newport Museum | 9 |
| Norwich Castle Museum | 19 |
| Oxfordshire Museum | 2 |
| Reading Museum | 15 |
| Verulamium Museum, St Albans | 22 |
| Shrewsbury Museum | 7 |
| West Stow | 5 |
| Wiltshire Heritage Museum, Devizes | 10 |
| Winchester City Museum | 3 |
| Yorkshire Museum | 14 |
| Total | 441 |
During this process the details of each object were recorded on a pro-forma sheet with a series of fields covering details such as material, identification, size, condition, attributes, date and provenance. Full details of all the fields used can be found in database section. A description of the object was also written and an assessment of its quality made. This was obviously more difficult for pieces which were not seen first-hand. However, wherever possible each piece was assigned a 'quality value'. One issue that has been raised by authors such as Henig (2007) is whether figurines were made in Britain or imported, and in many cases this judgement has been made with the assumption that well-executed, high-quality pieces were imported, while clumsy, crude pieces were locally produced. The quality values assigned to figurines ranged from Classical 1 and 2 (highly classical in style, often thought to have been imported) to Stylised 2 and 3 (stylised pieces which were almost certainly locally produced).
A number of fields were also devised to describe the provenance of the figurines. These include the circumstances of the find (chance, metal detector, field walking, or excavation), the type of site, the location type within the site, and specific context details. Where possible the site type was allocated to one of eight broad categories (colonia, civitas capital, large town (which came to mean London), small town, villa, rural, military and religious), which were designed to make a preliminary contextual sort. Obviously these categories are not without problems; for instance some sites have military origins but then develop into civilian settlements. On such sites finds from within the military establishment were labelled 'military', while those from external settlements or contexts dating after military activity ceased were assigned to the appropriate settlement category. In the discussion of the distribution of figurines among the various site types (Section 4), figurines from unknown contexts are labelled on graphs either 'unknown' or 'urban/military'. The next field, location type, also helped to define the area from which a find came, such as barracks or house. The site type 'religious' was used for those sites with a primarily religious function, such as the temples at Lydney or Uley in Gloucestershire. Finds from shrines or temples associated with a military or civilian settlement were identified in the field 'location type'.
The context quality was given a value between 0 (uncertain or regional context) and 3 (sealed excavation context). Unfortunately, the vast majority of metal-detector and antiquarian finds fall into categories 0 and 1, sometimes because false or inaccurate findspots were given by metal-detectorists and collectors. In some cases museum objects were assigned a provenance based on where the piece was purchased or donated from, not where it was actually found. For instance a group of figurines in the British Museum were said to be found in Blandford Forum, Dorset (1923, 0306.1-10) but two are Egyptian and the rest probably of Greek or Etruscan origin, thus it seems unlikely that they were found in Britain (Rigby et al. 1995). This issue is discussed by Robinson (2003) for three figures in the Wiltshire Heritage Museum, Devizes, which were said to have been found at Avebury but were part of a collection donated by Joshua Brooke, a collector who could not always be relied upon to provide an accurate provenance. While Cupid 570 may indeed be from Avebury, it seems unlikely that Worshipper 565 and Mars 566 are. For the purposes of this study, museum finds with no provenance were not recorded, while a note of the uncertainty was made in the record for those with probable provenances.
Metal-detector finds may have a known provenance, but more often than not no details of the site type are available. This means that these pieces cannot contribute to any analysis of the specific provenances from which figurines are being recovered. However, they can be used in a broad examination of the national distribution of material. In addition, many excavated finds from the late 19th-mid 20th centuries also have little contextual information about their specific location within the site, and even today it can be extremely difficult to extract the context details of finds from modern publications. However, it is hoped that sufficient data have been collected to allow some analysis of the contexts in which figurines were used in Britain.
© Internet Archaeology/Author(s)
University of York legal statements | Terms and Conditions
| File last updated: Tue Mar 20 2012