Is the Piercebridge assemblage a military votive deposit? Philippa Walton Over the past twenty years, an assemblage of approxim­ ately 5,400 Roman objects has been recovered from the bed of the River Tees at Piercebridge, County Durham. The objects were discovered during campaigns of amateur ar­ chaeological survey by two divers from the Northern Archaeological Group, a local archaeological society. They were reported to the Portable Antiquities Scheme in 20031 and are currently being processed as potential Treasure un­ der the Treasure Act 1997. THE SITE The riverine findspot for the assemblage is situated in a landscape rich in Roman archaeology. Dere Street crosses the River Tees at Piercebridge where a second century Ro­ mano­British settlement and fort dating to the later third century AD are known through a combination of survey and excavation.2 The deposit itself comprises more than 1,300 coins, 4,000 individual artefacts, 40kgs of pottery and a quantity of animal bone. When the assemblage was first discovered, there was some suggestion that it repres­ ented refuse washed from the river banks. However, both the location and composition of the assemblage point to­ wards an intentional votive act or acts. The distribution of artefacts found by the divers was relatively limited, being concentrated in an area measuring 5 metres by 5 metres, towards the centre of the river. Most objects were coated in a concretion comprising burned organic material and iron corrosion products which secured them to the riverbed. Although the deposition of coins and objects in water is characteristic of Roman religious practice throughout the western Roman Empire,3 the range of material deposited at Piercebridge makes the assemblage unique, with almost every functional category of Romano­British small find represented. For the sake of brevity, only a small propor­ tion of object types will be considered here.4 THE NUMISMATIC ASSEMBLAGE The numismatic assemblage comprises 1,319 coins, dating from the Republican period to the early fifth century.5 Al­ though there are numerous interesting rare or unpublished issues, as well as exotic coin finds including an issue of Juba II of Numidia and a worn Hellenistic piece, it is as an assemblage that the coins can provide the most informa­ tion. They provide an indication of the chronology of votive deposition at Piercebridge, whilst also potentially shedding some light on the sectors of Romano­British so­ ciety involved. Using Applied Numismatic techniques pioneered by Richard Reece,6 average values for coin loss in each period have been calculated using Reece periods (Table 1). These periods separate the Roman period in Bri­ tain into 21 divisions, based primarily on emperor’s reigns and enable a coin assemblage to be displayed graphically (Fig. 1). It is immediately clear that the profile is dominated by early Roman coin loss, with particular peaks in issues dating to Period 7, (AD 138–161), Period 10 (AD 196–222) and Period 11 (AD 222–238). It is interesting that these peaks correspond with particular episodes of campaigning and intense activity on the northern frontier and it is there­ fore possible that the deposition of coinage was associated with troop movements from the legionary fortress at York northwards. More generally, the emphasis on early Roman coin loss is characteristic of military installations or early urban foundations.7 The numismatic evidence may therefore confirm an extensive military presence in the area prior to the construction of the Roman fort which is argued to date to the late 3rd century AD.8 The denominational composition of the coin assemblage is quite different to that of other votive deposits with large coin assemblages in Roman Britain. The assemblages from Coventina’s Well and the Sacred Spring at Bath are almost entirely composed of bronze denominations, suggesting that low value coinage was deliberately selected for depos­ ition. However, at Piercebridge, the silver denarius is the dominant denomination. This dominance could be accoun­ ted for in a variety of ways. Firstly, it may be a reflection of the value attached to the cult worshipped at Piercebridge. The relatively high number of gold and silver items within the assemblage does demonstrate the willingness of de­ votees to deposit valuable artefacts. Secondly, military provinces, such as Britain, Upper and Lower Germany and Pannonia, appear to have received more silver coinage than civilian ones, such as Gaul and Italy and that individual sites with a military character tend to have greater propor­ tions of high value coins than established civil sites.9 JRMES 17 2016, 191–194 192 Journal of Roman Military Equipment Studies 17 2016 Therefore a military element to votive deposition might also be responsible for this high proportion of silver. Thirdly, however, the apparent wealth exhibited in the assemblage may in some part be illusory, with the dominance of the den­ arius instead reflecting the composition of the coinage pool from which the coins were selected for deposition. In the early third century AD, bronze denominations were not sup­ plied to Britain in any quantity and therefore the denarius for a time represented the lowest value denomination in circula­ tion. As there are peaks in coin issues from this period, it may be that deposition of coinage occurred most frequently in the early third century AD. Approximately 10% of the coins in the assemblage have been subject to some form of defacement, piercing, mutila­ tion or bending. More than 30 of the denarii, ranging in date from Neronian to Severan issues have been deliber­ ately cut and a further 30 have small semi­circular cut outs on their circumference (Fig 2). Again, interpretation of this treatment is difficult. Although halved Roman bronzes are known from several Roman sites throughout the Empire, including the Roman military spa of Bourbonne­les­ Bains,10 they have tended to have been interpreted within an economic context as improvised small change, whereas the clipping of denarii, where recorded, tend to be ex­ plained away as tests for silver content. However, it is also possible here that the modification of a small percentage of the coins from Piercebridge has a ritual element, either to mark them out as specifically votive objects or to render them useless should anyone attempt to retrieve them from their watery resting place.11 The abusive treatment of offi­ cial coinage and the imperial image may seem unusual, when it appears that at least some of the devotees making offerings at the site were soldiers. What might this say about their attitude to Roman authority? Table 1: The Piercebridge coin assemblage MILITARIA The large number of finds of a military nature also reinforces the theory that the army were involved in votive deposition at Piercebridge. Finds include 27 apron pendants, 11 scab­ bard slides, two sword chapes and six fragments of shield edging. There are also 12 portions of lorica squamata com­ prising rows of between two and six scales, a lobate hinge from lorica segmentata accompanied by two possible frag­ ments of lorica segmentata. These objects, although essential elements of military equipment, would all have been relatively easy to replace. There are sword fittings, but no swords; there are portions of armour, but no whole suits or helmets. These explicitly military objects are also accom­ panied by more than 40 elements of horse harness and equipment, including equine pendants, strap slides and stud mounts. As the horse was an extremely valuable commodity in the ancient world, these fittings are likely to have fur­ nished the horses of soldiers, rather than civilians. A hint regarding the identity of at least some devotees may be provided by the small assemblage of lead sealings Fig. 1: Histogram illustrating the Piercebridge coin assemblage by Reece period. Fig. 2: A denarius of Septimius Severus (FASAM­21D833), which has been deliberately cut. Photo: the author. Journal of Roman Military Equipment Studies 17 2016 193 Fig. 3: A copper­alloy Knee brooch (FAPJW­FD5267) from the assemblage. Photo: the author. Fig. 4: A copper­alloy ram figurine (NCL­DACBB3) from the assemblage. Photo: the author. recovered from the river. They include eight which read ‘LVI’ (Legio VI Victrix) and a further three which read ‘OVA’. The letters ‘OVA’ have been interpreted as a retro­ grade abbreviation for ‘Ala Vocontiorum’ a cavalry unit from Lower Germany, known from an inscription at Newstead and lead sealings from Leicester12 and South Shields.13 Tène III, through to third century Divided Bow types. Des­ pite the relatively wide date range, late second and third century brooches are most common. There are 23 examples of the Knee brooch (Fig. 3) which is not a common form in Britain, more frequently found in the forts of the German limes where they are known as ‘Soldatenfibeln’. They are generally regarded to be a military brooch type.14 OBJECTS OF PERSONAL ADORNMENT OR DRESS There are also numerous objects which have no obvious link with the Roman army. Indeed, items of personal ad­ ornment such as brooches, finger rings and pins form a large proportion of the material recovered from the river. For example, there are more than 70 jewellery elements in gold and silver as well as two gold and six silver finger rings. The presence of a relatively large quantity of pre­ cious metal comments on the status, wealth and significance of the votive deposit and the individuals mak­ ing offerings, and accords well with the numismatic evidence. However, the brooch is the most common item of personal adornment, with 111 examples recovered. The majority of the brooches survive intact, complete with the pin and spring mechanism, suggesting deliberate depos­ ition rather than discard and range in date from a single La OBJECTS ASSOCIATED WITH RELIGIOUS BELIEFS AND PRACTICES There are also a significant number of objects in the as­ semblage which have a direct association with religious beliefs and practices. These include two copper­alloy Cu­ pid figurines, two fragments of pipe­clay Dea Nutrix figurines and a copper­alloy ram (Fig. 4) and tortoise, two attributes of Mercury.15 Other smaller votive objects are represented including three miniature brooches and two minature axes, a silver plaque depicting Jupiter as an eagle abducting Ganymede and a lead votive plaque of a type more commonly found on the Danubian limes. There are also more than 150 rolled lead sheets which may or may not represent ‘curse tablets’.16 194 Journal of Roman Military Equipment Studies 17 2016 MOTIVATION FOR VOTIVE DEPOSITION This paper has briefly discussed some aspects of the as­ semblage from the River Tees at Piercebridge and has suggested a military element to the votive activity which oc­ curred at the site. However, the discovery of these objects has raised many questions. For example, why was the de­ posit placed in the river at Piercebridge? Was it a response to a specific event or a particular military campaign? Does the presence of some Late Iron Age objects17 in the deposit in­ dicate that the practice evolved from an earlier Iron Age focus? Answering these questions effectively is difficult, if not impossible. However, the location of the deposit at a ma­ jor strategic point on the way to the frontier may be significant. It is possible that the assemblage represent offer­ ings made by the military community at large, on their way from civilisation to Barbaricum. The geographical boundary of the river may have acted as a sort of trigger, reminding all who passed that they were entering the frontier zone, where discourse with the gods was deemed particularly appropriate or indeed necessary. NOTES 1 See www.finds.org.uk for more details about the Portable Antiquities Scheme and to search their Finds Database. 2 COOL & MASON, 2008. 3 Horace commemorates a sacrifice he made to the spring at Bandusia (Horace Odes III, 13) and Pliny the Younger dis­ cusses the offerings made at the source of the river Clitumnus (Pliny the Younger, Letters LXXX, 8), whilst deposits of coins and finds are known from numerous European rivers in­ cluding the River Liri in Italy (FRIER & PARKER, 1970; METCALF, 1974; HOUGHTALIN, 1985) and the Moselle and the Rhine in Germany (DERKS, 2014, 139–40). 4 See WALTON, 2008 for a broader summary of the assemblage. 5 See CASEY, 1989, 37–42 for a catalogue of 166 of the coins. A more detailed numismatic analysis of 1021 coins is pub­ lished in WALTON, 2012, 152–66. All coins are now recorded on the Portable Antiquities Scheme database access­ ible at www.finds.org.uk. 6 REECE, 1995. 7 LOCKYEAR, 2000, 403–13. 8 COOL & MASON, 2008, 311. 9 HOBLEY, 1998, 128. 10 SAUER, 2005, 58–68. 11 See KIERNAN, 2001 for a preliminary discussion of the mutilation of coinage from Romano­British sites. 12 RIB 2411.90. 13 ALLASON JONES & MIKET, 1985, 328. 14 BAYLEY & BUTCHER, 2004, 179. 15 These figurines find a parallel with the ram & tortoise which accompany a statue of Mercury found during the excavation of a cemetery ditch just outside the city walls of Verulami­ um (HENIG, 1995, 41). 16 At the time of writing, Sara Brown, a Conservation MSc student at Cardiff University, has opened six of the lead 17 rolls. Only one was inscribed. Although awaiting confirma­ tion from an epigrapher, it appears the inscription comprises repeated numbers rather than a conventional curse (Brown pers comm.). These include a copper­alloy La Tène III brooch, an iron mirror handle and a copper­alloy cosmetic grinder. BIBLIOGRAPHY ALLASON JONES, L. & MIKET, R. 1985: The Catalogue of Small Finds from South Shields Roman Fort, Mono­ graph II of the Society of Antiquaries of Newcastle upon Tyne, Newcastle upon Tyne BAYLEY, J. & BUTCHER, S. 2004: Roman Brooches in Britain: a Technological and Typological Study Based on the Richborough Collection, Reports of the Research Committee of the Society of Antiquaries of London, London CASEY, J. 1989: ‘A votive deposit from the River Tees at Piercebridge’, Durham Archaeological Journal 5, 37–42 COOL, H. & MASON, D. (eds): Roman Piercebridge: Ex­ cavations by D. W. Harding and Peter Scott 1969–1981, Newcastle DERKS, T. 2014: Gods, Temples and Ritual Practices: The Transformation of Religious Ideas and Values in Roman Gaul, Amsterdam Archaeological Studies, Amsterdam FRIER, B. W. & PARKER, A. 1970: ‘Roman coins from the River Liri’, Numismatic Chronicle 10, 89–109 KIERNAN, P. 2001: ‘The Ritual Mutilation of Coins on Romano­British Sites’, British Numismatic Journal 71, 18–33 METCALF, W. E. 1974: ‘Roman coins from the River Liri II’, Numismatic Chronicle 14, 42–53 HENIG, M. 1995: Religion in Roman Britain, London HOBLEY, A. S. 1998: An Examination of Roman Bronze Coin Distribution in the Western Empire AD 81–92, BAR International Series 688, Oxford HOUGHTALIN, L. 1985: ‘Roman coins from the River Liri III’, Numismatic Chronicle 145, 67–81 LOCKYEAR, K. 2000: ‘Site­finds in Roman Britain: A comparison of techniques’, Oxford Journal of Archae­ ology 19 (4), 397–423 REECE, R. 1995: ‘Site­finds in Roman Britain’, Britannia 26, 179–206 SAUER, E. 2005: Coins, Cult and Cultural Identity: Au­ gustan Coins, Hot Springs and the Early Roman Baths at Bourbonne­les­Bains, Leicester Archaeology Mono­ graphs No. 10, Leicester WALTON, P. 2008: ‘Finds from the River Tees at Pier­ cebridge, County Durham’, in H. Cool & D. Mason (eds), Roman Piercebridge: Excavations by D. W. Hard­ ing and Peter Scott 1969–1981, Newcastle, 286–93 WALTON, P. J. 2012: Rethinking Roman Britain: Coinage and Archaeology, Collection Moneta 137, Wetteren