Romano-British brooches have been subject to more detailed study than the majority of the period's small finds. This is due to their large number, potential for dating and, more recently, because new approaches to the study of 'identity' within Roman archaeology have placed a greater emphasis on the study of objects which have close associations with people and their day-to-day practices.
In over a century of scholarly attention most brooch studies have been site specific. Donald Mackreth's Brooches in Late Iron Age and Roman Britain is a notable exception to this trend, being the first ever comprehensive classification of Romano-British brooches to be published in full. However, it is not the only overarching brooch typology in existence. Hull's major corpus compiled during the 1950s and 60s may still await publication (Simpson et al. forthcoming) but it has, over the years, been disseminated through various site reports (e.g. Dudley 1967, 28-63; Cunliffe 1968, 79-93; Crummy 1983), the books of Richard Hattatt (1982; 1985; 1987; 1989), and Bayley and Butcher's Richborough study (2004). As the basis for these standard reference works, it can be said to be the foundation of all brooch studies in Britain to date. Mackreth's classification offers the first viable alternative to Hull's, and is significant for doing so.
The study is based on some 15,000 brooches from England and, to a lesser extent, Wales and Scotland. These have been collated from a variety of sources, including museum collections, site reports, studies published by others (e.g. Hattatt 1985; 1987; Snape 1994) and metal detecting – although no use has been made of the important and ever-increasing records from the Portable Antiquities Scheme.
The book comprises two volumes - one for text and the other for illustrations. The introductory chapter (ch. 1) provides a succinct outline of the technique for manufacturing brooches and sets out the study's methodology. In this it is clear that Mackreth's primary aims are to redress the two major problems of site-specific classification and dating, the latter of which was still heavily reliant on continental parallels when he began the project in the 1960s. He succeeds in both cases, although, given that a high number of brooches are unprovenanced, contextual dating does in some cases remain weak and in others is still unknown.
The greater part of the book - chapters 2 – 10 - examines the variety of brooches which are known to have been in circulation between 200BC and 400AD. The major different family groupings are discussed in separate chapters – Late La Tène (ch. 2), The Colchester Derivative (ch. 3), The Headstud (ch. 4), The Trumpet (ch. 5), Continental Imports (ch. 6), Plate (ch. 7), Knee (ch. 8), Crossbow (ch. 9), and finally Penannulars (ch. 10) - each of which contains a further breakdown of the major types and their varieties, along with information on their dating and distribution.
Mackreth does not adopt a formal typological framework in which the major groups are numbered and then subdivided and numbered again as required. Instead, acknowledging the polythetic nature of the brooch families and hence the difficulty of dividing them into groups, he names the major groups and subgroups using a mixture of the place names where they were found, and the diagnostic features of the type e.g. British Plate, Colchester Derivative, etc. Thereafter further subdivision is denoted by a variable system of letters and/or numbers. As the number and letter attributes are duplicated throughout his classification an abbreviation of the major family name is assigned to them, hence the Colchester Derivative Harlow type becomes CD Ha followed by 1.a, 1.b, etc. This does not make for a memorable classification but a full index of the abbreviations may be found on the CD-ROM which accompanies the book.
Throughout, oddities and hybrids are noted and situated, or even grouped together, within the classification according to best fit. Their inclusion is obviously useful as a point of reference. However, they also serve to remind the reader that any impression of discrete groups ordered by a linear progression either of devolvement or increasing complexity, as given by most typologies, is false, and that in truth the development of the various brooch types was multifarious and overlapped chronologically.
In terms of the end product, it is safe to say that Mackreth tells no word of a lie when he confesses to having "devised a monster of a classification system" (2011: vi). In all, he identifies over one thousand different types and varieties of brooches. Inevitably daunting, his detailed approach does not make for an easy and immediate framework of reference. However, there are a number of advantages to his system. One of these is the fact that by the nature of their construction the divisions in Mackreth's scheme are designed to better encompass the classification of fragmentary brooches than the more commonly utilised categories in Hull's typology.
The exacting division of this classification system should also be welcomed by all those who are aware of, and have commented on, the variation encompassed within Hull's definitions of types (e.g Hingley 2006: 2; Olivier 1988: 44). A good case in point is the well-known Colchester Derivative with Harlow spring type (aka double pierced lug). Hull subdivided these into a total of six varieties (Type 92; Type 93a-d; and Type 121(a)). By contrast Mackreth distinguishes a total of 22 (CD Ha 1.a (1-3)-1.d; CD Ha 2.a – 2.e; and CD Ha 3.a (1-2) – 4.b.).
The benefits of Mackreth's specificity are especially manifest in his recognition of individual workshop groups and regional distributions, both of which have been suggested for some time, but have previously been the subject of only limited systematic study. These identifications are important when considering the usage and wider social implications of brooches and it is these which Mackreth discusses in his final chapter (ch. 11).
As is to be expected, Mackreth's concluding observations are based largely on the geographical and chronological distributions that have emerged through his brooch classification. In the absence of an explicit theoretical framework his straightforward equation of these groupings with particular tribes and military units can be said to verge on the cultural-historical, especially when contrasted with more recent theoretical and contextual artefact analyses. However, it would be unfair to criticise him for this, as a more detailed analysis is beyond the already ambitious scope of the book.
Instead Mackreth's work deserves to be recognised as a strong platform upon which future research can be built. For example, the regional patterning he has identified within brooches has been recognised in other artefact types during the Late Iron Age to Roman period. This diversity and its decline in the later Roman period cannot simply be documented; it must be accounted for. In addition, smaller-scale and cross-cutting patterns need to be explored through the contextual analysis of brooches and the wider suites of artefacts in which they may be found. Enquiry such as this will improve our understanding of cultural change in the Roman period, the social context in which adornment was used, and the nature and function of different sites.
It will also be interesting to see if, and how, future metallurgical examination of brooches may affect our understanding of Mackreth's classification. Certainly, Bayley and Butcher's (2004) analysis of the brooches from Richborough proved beneficial in assessing the validity of categories within Hull's typology and for identifying some inter and intra-type workshops. In light of the limited physical evidence for the manufacturing and trading of personal adornment, the pursuit of this type of analysis is useful to our understanding of production and the economy in Roman Britain.
The second volume of the book contains the plates and figures referenced in the text. A total of 2093 brooches are illustrated, all deliberately selected from those not already published elsewhere. Expediently those illustrations which represent the more common forms of brooches are noted with an asterisk. As has been demonstrated by the separate reprinting of the visual catalogue from Hattatt's Ancient Brooches and Other Artefacts (1989), Mackreth's synthesis will be greatly appreciated by many, even if the complete absence of brooch illustrations throughout the text makes it more difficult to follow. Other types of graphics are low in number in both volumes, and those few which do intersperse the text are not easy to comprehend. In particular, there is a noticeable omission of any distribution maps, which is odd as they would have been highly beneficial and were certainly originally envisaged as a component of the publication (Mackreth, 1993).
The CD-ROM which accompanies the book contains PDF, CSV and MS Access versions of the massive database that Mackreth has compiled. It includes the serial numbers of the brooches, their location and details of publication, along with a sparse and problematic semblance of a concordance with the Hull corpus. Unfortunately the Access edition of the database appears to be of limited functionality, but the CSV file can be transferred into an Excel spreadsheet or an Access database. At a time when the compilation of large datasets is less feasible due to restraints of time, money, etc, it is good to see the importance of data re-use being appreciated by the author and the inclusion of this data in digital format is a great boon to current and future researchers.
This book, the end result of forty years of research, is an impressive achievement. Its content and format are not very accessible, but persistence does pay dividends. Mackreth has succeeded in producing a valuable comparative and analytical device that will greatly facilitate the future investigation of the dating, distribution, production, and use of Romano-British brooches. As such, it will be of relevance to anyone with an interest in Roman small finds, although given its intricacy and cost (£70) it will definitely find greater currency amongst specialists than the more general finds enthusiast.
1 Please note that all comparisons are with the original Hull manuscript as viewed by myself at the courtesy of Colchester Museum. They therefore do not take into account any potential alterations implemented by the late Grace Simpson, nor more recently by Nina Crummy who is continuing the mammoth task of preparing the corpus for publication.
Hingley, R 2006 Review: Roman Brooches in Britain: a Technological and Typological Study based on the Richborough Collection. By Bayley, J. and Butcher, S. 2004. American Journal of Archaeology 110(1) .