THE ROGIET HOARD AND THE COINAGE OF ALLECTUS EDWARD BESLY Introduction THE three-and-a-half centuries of Roman rule in Britain have left an extensive legacy, not least in the form of coinage. The system of coinage introduced under Augustus (27 BC-AD 14) - gold aurei, silver denarii and a range of copper alloy denominations - provided the money used in the province for over two hundred years until its final collapse in debasement and inflation around AD 270. There followed a quarter of a century in which a partial reform by the emperor Aurelian (270-5) appears to have had relatively little impact in Britain: hoards and individual finds point to continued circulation of the basest 'radiates' (double-denarii)1 of the recently-suppressed 'Romano-Gallic' state (260-74), contemporary issues of legitimate emperors such as Gallienus (260-8) and Claudius II (268-70) and widespread production of unofficial imitations, notably of coins of the Gallic usurpers Victorinus (269-71) and Tetricus I and II (271-4); with gold and silver nowhere to be seen. Around 286-7, Britain and parts of northern Gaul became the scene of another usurpation, by the fleet commander Carausius, who had fallen out with the legitimate emperors Diocletian and Maximian. Carausius was in turn murdered and succeeded by his minister Allectus, in 293. Carausius issued the first fine-silver denarii for nearly a century and both he and Allectus produced gold coinage, the latter apparently in some quantity. At the everyday level, both issued billon coinage on the pattern of Aurelian's reformed issues (these are known today as aureliani); the numerous British hoards of the time, however, are mostly dominated by the debased 'Gallic' issues. The 'British' state was reincorporated into the Empire in 295 or 296. In the meantime, in 294-5, Diocletian undertook a fundamental reform, introducing a uniform coinage empire-wide that set the pattern for the fourth century. After reconquest, the coinage of Carausius and Allectus was suppressed and the new currency imposed in Britain. In the absence of reliable historical evidence for the 'British' empire of Carausius and Allectus, interpretation of its coinage has assumed considerable significance.2 Understanding of this coinage has unfortunately been hampered by the rarity of well-preserved specimens and hoard groups and some published work has been highly speculative, from the time of Stukeley, in the eighteenth century, to the present. In September 1998, however, an unusual and significant hoard dating from the reign of Allectus was found in south-east Wales and this provides a rare opportunity to examine a well-preserved sample of one element of the coinage of the 'British' empire - the Q-radiates, or 'quinarii' of Allectus. The purpose of this paper is to place on record this remarkable assemblage and its context in the monetary circulation of the north-western part of the Empire; new evidence on the minting of aureliani; and to discuss the Q-radiates of Allectus and their place in Romano-British currency on the eve of reconquest and monetary reform. Acknowledgements. This report is published with the aid of a substantial grant from Amgueddfa Cymru — National Museum Wales. I am grateful to my colleagues Mary Davis. Penny Hill and Louise Mumford, for their superb conservation of the hoard; to Mary Davis for her analytical work; Jim Wild, photography; Sally Carter, documentation; Jackie Chadwick, for the splendid drawings that form Figs 1, 5, 6, 8 and 9; and Richard Brewer, for his support and encouragement; Cathy King and David Algar for information on the Ewelme and East Harnham hoards, respectively; Richard Abdy, who shared his catalogue of the recent Gilmorton hoard and his work on the new Gloucester list; Volker Heuchert, Ashmolean Museum; Adrian Poptscu, Fitzwilliam Museum; Donal Bateson, Hunterian Museum. Michel Amandry and Sylviane Estiot kindly answered questions on some coins in the Bibliotheque nationale de France; and I am particularly indebted to Sylviane Estiot for her comments on the coinage of Probus. Roger Bland commented helpfully on an early draft. Especial thanks to Stewart Lyon for his patience in explaining things statistical. Thanks also to Classical Numismatic Group, Inc, for permission to reproduce an image from their sales and to the Ashmolean Museum, the Trustees of the British Museum and the Hunterian Museum, University of Glasgow, for the use of images of coins in their collections. The text and all other images are Copyright ©National Museum of Wales. 1 The term 'radiate' is descriptive, from the rayed crown of the emperor's effigy that distinguished this double-denomination from single denarii, which bore laureate or bare-headed portraits. 'Radiates' of the empresses in fact depict a crescent behind the portrait bust. 2 See Casey 1994, Chapters 5, 6, 10 and Shiel 1977; see now also Williams 2004. 46 THE ROGIET HOARD In terms of design and production, the coinage of the later third century lacks in general the quality of that from earlier imperial periods, though amongst the aureliani are to be found some very handsome specimens. However, the rapid turnover of the 'soldier-emperors' of the third century, the debasements and attempted reforms of the coinage and the complications of several significant usurpations have given the Rogiet deposit an unusually varied composition amongst Romano-British coin hoards. Its forty-year span covers coins from eighteen reigns, in the names of twenty-six emperors or members of their families, with over 1,050 individual varieties. THE HOARD AND ITS CONTEXT Discovery The hoard was discovered on 10 September 1998 by Colin Roberts, who was using a metal detector on farmland between Llanfihangel Rogiet and Rogiet (ST 4587), approximately 2 km west of Caldicot, Monmouthshire.3 The find was promptly reported to the National Museum of Wales (NMW) and to HM Coroner for Gwent, as required by the Treasure Act 1996, which had come into force in England and Wales on 24 September 1997, replacing the former common law of Treasure Trove. The site was visited on 11 September by staff of NMW and the Glamorgan- Gwent Archaeological Trust to record the findspot and take delivery of the coins. The hoard, which at this stage numbered 3,778 coins, was declared treasure at an inquest in Newport on 10 December 1998 and was acquired in May 1999 by NMW (accession number 99.31H). It transpired subsequently that small numbers of coins had been found at the site for some years previously and the total number of coins certainly attributable to the deposit now stands at 3,813.4 Under previous treasure trove practice, the Rogiet hoard would have received no legal protection, since it consists of copper alloy coins containing very little precious metal. The hoard can claim a special place in the history of treasure law and practice in England and Wales as the first significant hoard of base metal coins to be declared treasure under the terms of the 1996 Act.5 Archaeological Context The hoard was found by metal detecting on recently-seeded grassland, in a field that had been under cultivation for many years. According to the finder, the first coins were located in plough soil and the bulk of the hoard at a depth between 14 and 20 inches (0.35-0.50 m). As examined on 11 September 1998, the find spot comprised a roughly oval hole 0.82 m by 0.34 m; some 0.3 m of plough soil overlay a subsoil c.0.2 m deep which in turn rested on an orange/red sandy gravel with some larger rounded pebbles in it. No trace survived of the original depositor's cut, but this appears not to have penetrated the natural gravel. There was no sign of a container, but the finder reported several small iron nails which, with the general shape of his excavation, might suggest that the coins had been deposited in a rectangular wooden box, though this is not certain. No traces of mineralised fabric were observed on any of the coins. The hole also yielded a few sherds of worn pottery; pottery and stone scatters and a number of late third- and fourth-century coins have been found elsewhere in the field. The hoard site is near the shore of the Severn Estuary on a slightly elevated area of land (c.10 m O.D.) between the Caldicot Levels to the south and hills rising to 82 m O.D. to the north. (In terms of the modern landscape this lies between the M4 motorway/Great Western Railway to the south and the M48 to the north.) Excavations in 1996 in the adjacent field to the east, ahead of residential development, located a Roman building in stone, of probable second-century date.6 About 3 km to the north, over the hills, lies the important Roman town and 'tribal capital' of 3 'Rogiet' is pronounced with a hard G: 'Rog - it'. 4 Of the supplementary coins, six have subsequently been acquired by NMW (2000.7H/1-5; 2002.14H). 5 See Treasure Annual Report 1998-1999 (London, 2000), pp. 121-2, no. 306; a preliminary report, which shared the title of the present paper, was delivered to the British Numismatic Society's meeting on 25 June 2002. 6 Marvell 1996. THE ROGIET HOARD Caerwent (Venta Silurum)', to the west, just under 12 km away, is the legionary fortress of Isca at Caerleon. The 'shore fort' at Cardiff, further to the west, was built towards the end of the third century (Fig. 1). Composition The Rogiet hoard comprises 3,813 coins of the middle and later years of the third century AD, summarised by reign and by mint in Table 1.