Evaluation of an area of land close to Glocester Docks was undertaken by Cotswold Archaeology in November 2002 on behalf of South West Regional Development Agency. The site lies outside of the walls of the colonia, some 100m from the site of the Roman south gate, in an area of back plots within the southern Roman suburb and bordering the road to Sea Mills. Archaeological deposits encountered included pits of medieval date and somewhat truncated Roman deposits, predominantly of third and fourth century date. A ditch, which contained a small quantity of broadly datable Roman pottery and ceramic building material, also yielded a clasp-knife handle of unusual form.
Clasp-knife handle of red deer antler in the form of a seated horse. The blade and folding mechanism are absent, and there is damage to the head end of the handle and to the animal’s tail. Additionally there is damage to the muzzle area, with some suggestion that this was joined to the body to form a suspension loop. A rebate at the head end of the handle was clearly the means of attaching the blade, together with a ferrule and pin. An absence of copper staining to the area of the rebate suggests that the ferrule and fixing pin were of iron. The pivot holes are, characteristically for this class of object, offset from the centre. A groove to accept the folded blade runs the length of the back portion of the handle.
The subject is carved in the round with prominent mane, small forelegs, stubby ‘docked’ tail and hind legs crossed at the base. Although the pose is stylised, as dictated by the functionality of the object, aspects of the design are naturalistic, particularly treatment of the head and hind quarters. Details such as the strands of the mane, foreleg hooves, nostrils, ?tendons to the lower hind legs and folds defining the hind legs and face are picked out by cut strokes or deeper grooves. The mane is sharply defined, serving to frame the face. Deep-cut grooves, depicting the fall of strands within the mane are carefully executed and paired at the rear. It has been suggested (pers. comm. R. Jackson), that the handle viewed from behind mimics the form of a horse’s hoof and is an example of an artistic device sometimes employed by Roman craftsmen. The level of craftsmanship is undoubtedly high, the portrayal lively or even humorous and some familiarity with the equine subject is evident.
Clasp knives of Roman and post-Roman date are reasonably well known finds from Britain. The most simple Roman-dated examples are made from antler tines, with the iron blade attached by means of a ferrule and pin and folding into a groove cut into the side. More elaborately decorated examples, with handles of bone, ivory, antler or cast bronze take the form of anthropomorphic or zoomorphic subjects. The commonest representational designs are openwork hound and hare types which occur in metal and bone.
The most unusual forms occurring in Britain are those of bone, ivory or antler which are carved in the round and depict animal or human subjects often in very elaborate fashion. A later second to mid third-century date is most frequently applied to such objects (pers. comm. N. Crummy). The number of the more elaborately representational knife handles from Britain has in recent years significantly increased, with a number of new discoveries. Representational clasp-knife handles appear to be widely distributed in Britain, though they derive mainly from urban, villa or military sites. Western British sites which have yielded clasp knives include Wanborough, Wilts (Vaughan 2001), Cleeve Hill (Cheltenham), Glos, (Rawes 1986) and Shakenoak, Oxon (Brodribb et al. 1973).
The continental distribution of the elaborate representational clasp-knife handles would seem to indicate an origin in the Rhineland, with workshops most likely situated in Trier or Cologne (Von Mercklin 1940). Von Mercklin illustrates 45 examples of bone, ivory, jet and bronze depicting a variety of subjects including male and female busts, deities, gladiators, wrestlers and erotic scenes. Animal subjects include hare and hound groups, lions, leopards, dogs, dolphins and birds. The sole reference to an equine subject is a horse head scratched into the side of a plain handle.
In most respects the Gloucester knife handle fits well with the artefact class grouped by Von Mercklin. The means of execution and the high level of craftsmanship are certainly comparable. Significantly, the form of the rebate and method of blade attachment would seem to compare well with the illustrated examples (ibid., Tafel XXXVIII, nos 1-2). The subject matter is unusual, although the horse is in other media a not uncommon theme in classical art. In respect of the material used, red deer antler, the Gloucester knife is exceptional, although it is not clear whether Von Mercklin’s methods of analysis would have been sufficient to discern antler from bone.
Conservation treatment and identification of material was carried out by Esther Cameron of the Institute of Archaeology, Oxford. I am indebted to Nina Crummy for supplying a copy of Eugen Von Mercklin’s article and to Christel Inder for the translation. I would also like to thank Nina Crummy and Ralph Jackson for viewing images of the handle soon after discovery and giving me their opinions.
Ed McSloy, Cotswold Archaeology,
ed.mcsloy@cotswoldarchaeology.co.uk
References
Brodribb, A C C, Hands, A R and Walker, D R, 1973, Excavations at Shakenoak Farm, near Wilcote, Oxfordshire, pt IV, Site C (Oxford)
Mercklin, E von, 1940, ‘Romische Klappmessergriffe’ in Festchrifte Victoru Hoffileru (Zagreb), 339-52
Rawes, B, 1986, ‘The Romano-British settlement at Haymes, Cleeve Hill, near Cheltenham’, Trans Bristol and Glos Arch Soc 104, 61-93
Vaughan, S, 2001 ‘Objects of Bone and Related Materials’ in Anderson, A S, Wacher, J S & Fitzpatrick, A P, The Romano-British ‘Small Town’ at Wanborough, Wiltshire Britannia Monograph Series 19 (London)