Research Notes - Roman

RESEARCH NOTES 240 a roman silver crossbow brooch from the vicinity of st radigund ’s abbey , near dover A silver crossbow brooch was recovered from the plough soil in the late 1980s near to St Radigund’s Abbey. The exact find spot near a footpath has been lost to the passage of time; it was in an area of level ground on the North Downs at an elevation of 135m aod, between the Alkham Valley and Coombe Valley (NGR TR 26883 42235), 0.76km north-west of St Radigund’s Abbey farm. Unfortunately, the finder had not recognised the importance of the find which was completely covered in green and black patina, and thought to be bronze. The item was viewed by the author during 2013 – during the conservation process, small slivers of patina cleanly lifted from the surface revealing a decorated silver brooch. The early types of these heavy crossbow cloak fasteners/ brooches had terminals that screwed into the arms of the brooch and first appeared around ad 200. They occur in significant numbers in the Limes forts along the Rhine and Danube, and also in Oudenburg, although silver and gold examples are rare. Around ad 250, an even heavier type with fixed onion shaped terminals evolved, some of the earliest examples being found in graves at Koln in association with objects assigned to 270-330 (Bushe-Fox 1949). Crossbow brooches, an important variation of fibulae, were worn as status symbols by senior ranking officers in the Roman army as well as eminent officials of the Roman state (Johns, C. 2006). These brooches were an effective and robust design of safety pin, the most prestigious examples being made in silver or gold. A number of other bronze and gilt examples have been found at military sites around Britain including several examples from Kent. A published study records 15 silver and 8 gold examples of the period found in the UK and Continental Europe (Swift 2000). Parallels have been found in Belgium and the Netherlands in burials 27, 57, 115 and 172 from the cemetery of Oudenburg, dated to the second half of the fourth century (Mertens, J. and van Impe, L. 1971). Other examples are: Sc110, with the catchplate missing, from Tongeren, Gallo-Roman Museum, Belgium; RMOL05 from Maasdiel; RMOL25 from Wijk bij Duurstede, a Roman rural settlement in the Lower Rhine (where many military objects have been found). A recent study of the crossbow brooches from the Netherlands and Belgium indicates that the average length of Type 3/4b is 75mm, width 50mm and height 27mm. The St Radigund’s example is slightly bigger and uncommon and probably related to the military reorganisation and changing fashions in the 4th century (Van Thienen 2014, pers. comm.). RESEARCH NOTES 241 Fig. 1 shows the fully developed Type 3/4b (in Swift 2000), or Type 4a (Keller 1971), dated to the second half of the 4th century (although this late form does not appear in the Limes forts). During the excavation at the Roman fort at Richborough, similar (albeit bronze) examples were recovered of which the eight relevant ones are: no. 9 (Bushe-Fox 1926, plate XII), nos 16, 17 and 18 (Bushe-Fox 1928, plates XVII-XVIII), no. 64 (Bushe-Fox 1949) and no. 65 (being the fully developed 4th-century crossbow with exact parallels having been found in Mainz (Lindenschmit, A.V.H.V. iii,2, iv, fig.2), and in Totis, Hungary (Roder, Fahrbuch des Provinzial Museum zu Hannover, N.F.V., fig. I, p. 14). nos 80 and 81 (Cunliffe 1968, plate XXXIII). (For a more recent review of the brooches from Richborough and composition analysis, see Bayley, J. and Butcher, S. 2004.) Fig. 1 The St Radigund’s Roman crossbow brooch. (Illustration by William Laing.) RESEARCH NOTES 242 In detail the St Radigund’s example was cast with three fixed ‘onion’ terminals with beading below the knobs; the arms are decorated with three raised ribs on either side. Inset into an aperture just below the head terminal on the bow, the brooch is embellished with in situ dark blue niello setting, rarely still present on most examples. The ornamentation down the stem is formed by linear double vertical groves on opposing sides, closed just above the bow an d foot juncture. The foot decoration is almost the same as the Richborough brooch no. 65, consisting of an undecorated central faceted border separating two rows of incised twin annulet patterns on the upper foot and two rows of three incised annulets on the lower part. However, the Richborough top row of annulets is notably uneven in comparison to the St Radigund’s example. Both types also differ in that the St Radigund’s brooch has three incised linear grooves separating the two sides of the lower foot’s geometric pattern. The grooves on the St Radigund’s example also terminate between the top juncture of the faceted border and the upper annulet decorations on the bow-stem. The Richborough brooch also terminates between the top juncture and faceted border although the bow-stem is decorated with chevrons enclosed by twin vertical borders. From the brooch examples found at Richborough, the foot design of the bronze brooch no. 65 is geometrically parallel. The overall length of the St Radigund’s example is 8.2cm, marginally greater than the 8.1cm of example no. 65. The individual finish and decoration of these brooches were very personal although similar motifs and styles are found on most crossbow brooches in the UK and Continental Europe. The foot decoration, Type b2, is very common, although less so in Britain. The distribution of examples with this foot pattern shows many in the Danube provinces (Swift, E. 2000). Imperfections in the casting process of the St Radigund example can been seen in the twisted left angle of the head terminal together with two small holes present on the bow-stem and a further indentation on the left lateral side of the bow. A peculiarity in the functioning of the brooch is the complete absence of the solid catchplate and pin socket which had been vertically sawn off from the top to the bottom of the foot. During the cleaning process, on removal of the thick layer of patina, a series of grooved hacking marks were exposed running down the length of the foot suggesting removal of the catch-plate in antiquity. The sawn off catch-plate provides good evidence of intentional removal, possibly related to a phenomenon from brooches found in Nijmegen, Netherlands. Here many sole arms with attached knobs were found intentionally severed from the rest of the brooch. They all have approximately the same size and weight. Two possible options for this occurrence are postulated: one related to weights used for balances, the other to paying solders when no money was available or where precious metal objects were cut up to pay off the barbarian hordes. Further research is being undertaken (van Thienen, 2014, pers. comm.). The St Radigund brooch silver content has been non-invasively examined and authenticated as solid silver by F. Hind, jeweller, Dover. The silver weight is 54.14g; the missing catch-plate and pin is estimated at about 6.0g, suggesting an overall weight of around 60g. In the mid 4th century, the brooch represented a sizeable investment for a single dress accessory with the average pay of a common solider a little over one denarius per day. Based on the average weight of four RESEARCH NOTES 243 specimens of late 4th-century Roman silver denarius at 3.5grams, the intrinsic ancient value in silver of this brooch equates to about 17 days of a soldiers pay. This figure does not take into account the artisan’s time in creating the mould and producing the brooch for sale. Theoretically, it can be estimated that the final net value of the brooch could represent in time and material around 20-25 days pay of a soldier. The findspot of the brooch is perplexing in that no significant Roman archaeology has been found in the area despite a field inspection and previous archaeology undertaken in and around St Radigund’s Abbey by Dover Archaeological Group (Parfitt, K. 1993) suggesting that there is no evidence for a late Romano-British settlement or other buildings at this location (Parfitt, K. 2014, pers. comm.). Google Earth satellite coverage of the location shows no tangible evidence for contemporary structures, pits, large enclosures, tracks or roads. Situated 5km nnw of the Roman Fort at Dover (Dubris), the location has no obvious military strategic significance and is some 3.5 km west of the main Roman road leading from Dover to Canterbury (Margary 1973). Previous discoveries twenty years apart by Mr Phipps, detectorist, at or near the same location, comprise two rarely found Roman gold coins: Magnus Maximus ad 383-388 Solidus, Trier Ric 76; and Honorius ad 395-402 Solidus, Milan Ric x 1206 (Holman, D. 2009 pers. comm.; information extracted from Bland, R. and Loriot, X., 2010.) The landowner has recovered samian bowls, two damaged and one partially reconstructed from sherds, whilst ploughing the fields in the general area (Moynan, W.J. 2014, pers. comm.). The casting defects and the hacked catch-plate may indicate that the brooch, although finished, was rejected and retained for bullion and may indicate a potential small founder-hoard or offering site dating around the mid 4th century; thus, despite the physical absence of Roman structural material some kind of settlement could lie undetected in the vicinity of the findspot. It can be speculated that this item, if not rejected by the maker, may have been in the possession of a Roman military official of some standing, perhaps associated with the fort at Dover. Geophysical fieldwork in the St Radigund’s area is planned by the author that may help in answering this mystery. Acknowledgements The writer thanks Mrs Susan Lees for bringing the brooch to his attention (and donating the find for display in Dover District); Mr Moynan for his helpful local background knowledge; Mrs Rebecca and Miss Elissia Burrows for their time cleaning the brooch and William Laing who kindly produced the illustration. He is also grateful to Vince van Thienen (ph.d. student, Ghent University Belgium), Drs Steve Willis and Ellen Swift (of Kent and Canterbury University), and Keith Parfitt (Canterbury Archaeological Trust) for their kind assistance during the research for this report. vince burrows Bayley, J. and Butcher, S., 2004, Roman Brooches in Britain: a technological and typological study based on the Richborough collection. RESEARCH NOTES 244 Bushe-Fox, J.P., 1926, First Report on the Excavations of the Roman Fort at Richborough, Kent, p. 44, plate XII. Bushe-Fox, J.P., 1928, Second Report on the Excavations of the Roman Fort at Richborough, Kent, p. 44, plate XVIII. Bushe-Fox, J.P., 1949, Fourth Report on the Excavations of the Roman Fort at Richborough, Kent, pp. 121-122, plate XXXII. Bland, R. and Loriot, X., 2010, Roman and Early Byzantine Gold Coins found in Britain and Ireland. Cunliffe, B.W., 1968, Fifth Report on the Excavations of the Roman Fort at Richborough, Kent, p. 91, plate XXXIII. Johns, C., 2006, The Jewellery of Roman Britain: Celtic and Classical Traditions, Routledge. Keller, E., 1971, Die spatromische Grabfunde in Sudbayern. Margery, I.D., 1973, Roman Roads in Britain (3rd edition 1997). Mertens, J. and van Impe, L., 1971, ‘Het Laat-Romeinse grafveld te Oudenburg’, Archaeologia Belgica, p. 135. Parfitt. K., 1993, unpubl. site report. Swift, E., 2000, Regionality in Dress Accessories in the Late Roman West, Monographies Instrumentum II, Monique Mergoil, Montagnac, pp. 14-15, 24-27, fig. 44.