File:Iron Age, Detail of Bracelet 3 terminals (FindID 820219).jpg

English: From the Treasure Report, Dr Julia Farley, British Museum:

The hoard consists of three complete neck torcs and one bracelet. Torc 4 was initially found incomplete but the second half was found at a latter date and has been included below.

1. Complete gold torc with thistle-shaped terminals

A gold torc with solid circular-section neck ring, expanding towards the terminals. The terminals are made in one piece with the neck ring, and are thistle-shaped. Each consists of a small 'bead', narrowing to a thinner waist, then flaring out to a wider trumpet-shaped terminal. There are three slightly raised ridges on either side of the 'beads', and similar concentric ridges on the outer edge of the 'trumpets'. The circular inner face of each 'trumpet' head is slightly concave, and surrounded by a lipped rim. The interior is slightly scratched, but very smooth.

On each terminal, sets of small triple-circle motifs have been stamped around the circumference of the neck-ring where it swells to meet the bead of the terminal. On one side (which would have been on the wearer's left based on wear patterns, see below) the terminal has four equally-spaced sets of motifs. On the other terminal there are only three sets, with one wider gap, and the circles are slightly larger. The circles are not arranged identically within each trio, suggesting that each circle was stamped individually, rather than as sets of three.

The torc has been bent out of shape, but is generally in very good condition. There is some light scratching and abrasion over the surface. This seems to be a mixture of the small-scale cracking and striation expected from the original hammering out of the gold during manufacture, and more recent abrasions and scratches. The more recent scratches are concentrated on one face of the torc, running around the circumference and along the exposed edges of the flared terminals. This may have been the underside of the torc as originally deposited, with the abrasions caused by dragging during ploughing or movement down the hillside over time. The other side of the torc is much less abraded.

Small areas of wear and polishing can be seen on the raised parts of the terminals, showing where the torc would have been worn against the body. These are concentrated on one face (most likely the underside as deposited), suggesting that the torc was normally worn the same way up. The location of these worn areas shows that the terminals have been bent up and back from the position in which they would have lain when the torc was in use. The neck-ring is also bent at the sides and slightly at about the centre of the back, almost as if it had been 'folded' across the sides. This could have been intentionally carried out before the torc was deposited, but it is more likely that this occurred later, due to ploughing, especially given the nature of the damage to the other objects (see below).

Measurements:
Maximum external diameter (side to side): 146mm
External diameter (from terminals to back): 134mm
Internal circumference: approx. 410mm, plus gap between terminals
Thickness of neck-ring in cross-section: 5.7-5.8mm, widening to approximately 9mm before the terminal
Terminal 'bead' diameter: 11.9-12.4mm

Terminal 'waist' diameter: 7.5mm
Terminal end diameter: 20-21mm
Weight: 230.6g

2. A Complete gold torc made from a pair of twisted wires

Gold torc made from a pair of wires twisted around one another. At each end, the two wires are hammered and soldered together, continuing as a single length which flares into a small conical terminal with concave circular face at the tip. Each terminal is bent back on itself. The two bent-back ends are hooked together to form a clasp, which closes the circle of the torc. The clasp was probably worn at the back of the torc. One of the clasp terminals is heavily worn along one side, perhaps where it rested, and rubbed, against the neck of the wearer.

The torc has been bent into an irregular shape, but is generally in good condition. There is some light wear where the strands have rubbed against each other, and light scratching and abrasion over the surface. This seems to be a mixture of the small-scale cracking and striation expected from the original hammering out of the gold wires during manufacture, and more recent abrasions and scratches. Like torc no. 1, most of the more recent abrasions are concentrated on one face, probably the underside as deposited, and overall the form of the torc is as if it has been bent in half. The irregular bending of torc no. 2 can be better understood if it is placed within torc no. 1. It seems likely that torcs were buried in a nested arrangement, and later hit by the plough, causing bending and warping, and perhaps at least some of the abrasions on the underside.

Measurements:
Maximum external diameter (diagonal): 144mm
External diameter (side to side): 88mm
Internal circumference: approx. 420mm
Thickness of individual wires in cross-section: 1.9-2.1mm
Terminal diameter: 5.3-5.7mm
Length of terminal from head to bend: 9.0-9.5mm
Weight: 42.1g

3. Complete gold bracelet

Gold bracelet with flaring buffer terminals and twisted body. The terminals are thick circular discs, with slightly concave ends. From the weight of the object, they are possibly hollow. Light hammer marks can be seen in the dished inner surfaces, suggesting the piece was probably raised rather than cast.

Behind the bead-like terminals is a slight indentation around their circumference, and beyond this begins the body of the bracelet. Each end narrows gradually towards where the twisted wire region begins and, on both sides, this flared conical region is decorated on the external face of the bracelet. Three flat loops of wire, each forming a pelta shape (two adjoining circles with an arc over the top), have been ?soldered on, over chased or engraved decoration. The peltas nearest each end meet back-to-back, and the other pelta runs perpendicular across the top. These wires enclose, within or between them, positive and negative shapes: six circles, three on either side, run down the sides, and in the central space between them are created four fans and a concave-sided kite shape. Within each of these shapes, the surface of the bracelet is decorated with chased or engraved lines which fan out from the centre of the motif towards the sides. The lines in the central kite and fan shapes are perpendicular to those in the circles.

As the flared terminals narrow, they have been sliced into two and slightly flattened at the join to create a pair of hollow 'wires', which are twisted around each other to form the body of the bracelet. The spaces between these two thick 'wires' are filled by two thinner wires, which coil around the thicker pair. Each thin wire is square-sectioned, and has been tightly twisted on itself to give a hatched, rope-like effect. These two thin wires are joined to the terminals on the inner and outer surface of each end, at the gap where each terminal splits to form the two thick 'wires'.

The construction technique is most visible at a point about a third of the way around the circumference, where the bracelet is bent back and has untwisted slightly. At this point one of the thin wires is broken and one of the thicker 'wires' is partially cracked, revealing its hollow, folded interior. One of the terminals is bent at an unusually sharp angle to the body of the bracelet, with several more recent abrasions around this area. This bending, untwisting and cracking probably occurred as a result of the bracelet being caught by the plough. Otherwise the surface is in very good condition, with the normal small cracks and striations associated with Iron Age jewellery manufacture.

Measurements:
Maximum external diameter (side to side): 105mm
External width (from terminals to back): 63mm
Terminal diameter: 14.4-14.9mm
Thickness of terminals: 5.2-5.5mm
Length of terminal from head to start of wires: approx. 22-24mm
Thick wire diameter: 2.5-4mm
Thin wire diameter: approx. 0.8mm
Weight: 35.1g

4. Fragmentary gold torc with small thistle-shaped terminals

A gold torc with solid circular-section ring, widest at the back, expanding towards the terminals. Broken into two pieces. The terminals are made in one piece with the neck ring and are thistle-shaped, similar to (though simpler than) those of torc no. 1. Each consists of a small 'bead', narrowing to a thinner waist, then flaring out gradually to a wider trumpet-shaped terminal, with disc head. The circular inner face of each 'trumpet' head has an irregular polygonal indentation. Each terminal is more worn on one side (probably the lower side as deposited), suggesting that the torc was consistently worn one way up.

The torc is broken and distorted, but is otherwise in good condition. The break and bending most likely happened along an existing stress line in the metal when the group was hit and dragged by the plough in the 1980s. There is some light scratching and abrasion over the surface. This seems to be a mixture of the small-scale cracking and striation expected from the original hammering out of the gold during manufacture, and more recent abrasions and scratches. The latter are concentrated on one face of the torc, probably the underside as deposited, caused by being dragged by the plough. Like the others, it appears to have been placed in the ground the 'right' way up, based on wear patterns. The plough damage has also caused irregular bending which can best be understood if the torc was at the bottom of the nested arrangement of the hoard, and has been bent against the larger, heavier torc (no. 1).

Measurements:

Total internal diameter of torc, approx. 38cm

Larger fragment:

Maximum length in plan: approx. 150 mm
Thickness of wire in cross-section: 2.5-3.5mm
Terminal 'bead' diameter: 6.5mm
Terminal 'waist' diameter: 4.5mm
Terminal end diameter: 10.0mm
Length of terminal: 18-19mm
Weight: 41.0g

Smaller fragment:

Maximum length: 126mm
Thickness of wire in cross-section: 2.6-3.2mm
Terminal 'bead' diameter: 6.9mm
Terminal 'waist' diameter: 4.6mm
Terminal end diameter: 10.6-10.9mm
Length of terminal: 18-19mm
Weight: 22.2g

Metals Analysis

Non-destructive XRF analysis of the four objects was carried out by Pieta Greaves (BMAG). This revealed surface compositions of approximately 74-78% gold, 18-22% silver, 1-3% copper, with small traces of iron, mercury and tin. These values are consistent with Iron Age gold work, and the alloy is close to some natural European golds (Northover 1992, 241). The average values from each torc are below. The detailed analysis is available on request.

<tbody></tbody>

Torc no.	Au/%	Ag/%	Cu/%	Zn/%	Fe/%	Hg/%	Sn%	Pb/%
1.	77.44	18.54	1.79	0.00	1.48	0.37	0.38	0.00
2.	77.13	18.81	1.91	0.00	1.44	0.34	0.38	0.00
3.	76.77	19.39	2.06	0.00	1.12	0.29	0.36	0.00
4.	74.23	22.10	2.85	0.00	0.30	0.22	0.31	0.00

Parallels

Torcs 1 and 4 both have plain gold loop bodies with thistle-shaped terminals. This is generally a continental style of torc, and there are only two parallels from Britain: a gold torc terminal from Caistor in Lincolnshire (NLM-605352; 2013 T130), and a bronze torc from Medway, Kent (Jope 2000, Pl. 32a-b). These finds belong to Stead and Rigby's (1999, 67-8) 'torques with thistle-head terminals' type, Eluère's (1987) group of 'Torcs with Buffer Terminals and Intermediate Collars', and Hautenauve's (2005) Type IIIa ('Torcs with a Large Body and Terminals without a Clasp'). Hautenauve calls the thistle-terminal group the 'variante de Mitteleuropa' of her Type IIIa. On the continent the distribution of these torc types centres on central and eastern France, western Germany and Austria, although one find was made in a grave at Filottrano, in central Italy (Jacobsthal 1944, No. 44), and another is reportedly from Clonmacnois in Ireland (Jacobsthal 1944, No. 49; Raftery 1987, no.451). Most torcs of this type are made of bronze (see for example ML.1430, 1611, 1715, 1767, 1810, 1811, 1812, 1818, 1899 from the Morel collection at the British Museum, and Jacobsthal 1944 no.s 206-221). Gold examples are usually more highly decorated, such as the torc from the Waldalgesheim grave (Jacobsthal 1944 no. 43, Joachim 1995, 61-63), and finds from Filottrano, the Czech Republic, Austria, and Clonmacnois (Jacobsthal 1944 no. 44, 47-9). However, in addition to the Caistor example, there are two similarly plain gold thistle-terminal torcs. One (BM GR1867,0508.477) is believed to be from France, possibly in the Department of Loiret (Echt et al. 2011, 45). The second is from Heerlen in the Netherlands (Echt et al. 2011). The decoration on these torcs is different is slightly different to the Staffordshire find however, with rows of beading on the Loiret torc, and more elaborate engraving on the Heerlen example, although it should be noted that the decoration on the Heerlen piece also includes triple dot motifs behind the terminal 'beads', as does the decoration of several torcs from the Morel collection (ML.1762, ML.1767). This motif may have held a particular meaning or significance to the makers or wearers of these objects.

No close parallel could be found for torc no. 2 from the Staffordshire find. It most closely resembles simple twisted-wire torc types from southern Britain, but these have large terminals, probably worn at the front, rather than simple clasps. The simple hook-clasp mechanism is paralleled on ribbon torcs (the distribution of which centres on Scotland and Ireland), but on ribbon torcs the ends of the clasps are either plain hooks, flat discs, or have conical knobs (Eogan 1983). The Staffordshire torc, in contrast, has dished conical clasp terminals.

The twisted bracelet, no. 3, is perhaps the most important find from the group. The form of the terminals is a variant on the thistle-shape, but with a simpler, rounded buffer form, without the bead-and-trumpet which typifies the thistle. The form of the terminals is quite closely paralleled by the pair of bracelets from the burial at Waldalgesheim (Jacobsthal 1944 no. 55, Joachim 1995). The form of the body of the bracelets, with the double pairs of twisted wire, is likewise very similar to the annular arm ring from Waldalgesheim (Jacobsthal 1944 no. 54, Joachim 1995). Whilst the form of the Staffordshire bracelet is paralleled at Waldalgesheim, the decoration is not. The site at Waldalgesheim gave its name to the sinuous, often three-dimensional 'vegetal' style of decoration which adorns the many of the finds from the grave. In contrast, the design on bracelet no. 3 from the Staffordshire assemblage is two-dimensional, and much stricter, indeed almost geometric, in form. Each design is made from three adjoining pelta motifs that individually seem to derive from a palmette or lyre pattern (compare Fox 1958, Fig. 82.B: 'palmette derivatives'), and as a whole also give a sense of referencing Greek-inspired palmette designs. The use of this strict style of Celtic art, referencing the Greek palmette and known as 'Early Style', is relatively rare on objects found in Britain. Examples include the Wisbech scabbard, the Cerrig-y-Drudion crown and a bucket fitting from Minster in Kent Jope (2000, Pl.10c, 28, 29 and 30). The designs on the Staffordshire bracelet are unlike most Early Style Celtic art, in that the decorative shapes are formed from adjoining circles with an arc over the top. Circles joined by an S-shaped curve are much more common in Celtic art, appearing, for example, on the jewellery from Waldalgesheim and many bronze torcs from north-eastern France (ML.1611, 1711, 1715, 1746). However, there are also examples of other thistle-terminal torcs with pelta-like decoration on the terminals, such as ML.1810 from Saint-Rémy-sur-Bussy in the Marne region on north-east France. The pelta motif itself is also widely seen on other (probably later) Celtic art objects from Britain, such as the Battersea shield (BM 1857,0715.1, Stead 1985). The unusual nature of the designs on bracelet 3 makes them hard to date stylistically, but given the context it is most likely that they are connected to Early Style traditions of Celtic art, rather than later developments in Britain.

Date range

Thistle-terminal types are a varied group of torcs, and dating varies between authors from the late 5th to the early 3rd century BC (Stead and Rigby 1999, 67-8; Eluère, 1987, 28; Hautenauve 2005, 67-70; Echt et al. 2011, 44). The Waldalgesheim grave, which provides some of the strongest parallels in form to bracelet no. 3 from find, has been dated to around 340-300 BC (Joachim 1995). These parallels suggest that a 4th to early 3rd century BC (400-250 BC) date for the manufacturing of the Staffordshire finds is likely. If the decoration on bracelet no. 3 is indeed a rare example of Early Style Celtic art in Britain, then a 4th century BC date is most likely. That would make this hoard some of the earliest Iron Age gold (as well as some of the earliest Celtic art) from Britain.

Regional context

There are two main distributions of precious metal torcs in Britain: an eastern group centred around East Anglia, and a smaller western group centred around Staffordshire. The latter include a type with a distinctive form of ring-terminals, known as 'cushion terminals'. Two complete 'cushion terminal' torcs have been found in Staffordshire: one from Glascote (BMAG, Painter 1971), and one from Needwood Forest (Royal Collection, Hawkes 1936). A group of entangled, twisted torc fragments from Alrewas (declared Treasure Trove in 1996 and subsequently acquired by The Potteries Museum and Art Gallery) are closer in style to the East Anglian group, with simple ring terminals. These torcs are conventionally dated to the 2nd or 1st century BC (c. 150-50 BC), significantly later than the likely production date of this most recent find.

Discussion

Whilst the objects were not recovered in situ, it seems likely that they were buried as a single deposit, near the top of a hill, with the bracelet (no. 3) and smallest torc (no. 2) probably nested inside the heaviest torc (no. 1), and the slimmer torc (no. 4) on the bottom. The torcs could have been buried any time after their production around 400-250 BC. Despite the moderate wear on some of the pieces, there is no reason to think that they were particularly ancient when they were buried, so a Middle to Late Iron Age date for deposition is likely.

The damage to the finds most likely happened when the group was hit by the plough sometime in the 1980s, bending the two complete torcs across the middle, distorting and breaking torc no 4, and bending back one of the terminals of the bracelet. It was probably at this time that the objects were dragged down the slope, separating the deposit.

The torcs have internal diameters of 38-42cm (approx. 15.5-16.5 inches), which would fit a woman or a slender man, but based on attributions from grave finds (see e.g. Stead and Rigby 1999), these objects were probably female ornaments. From the wear patterns it seems that all the torcs were predominantly worn one favoured way up. Since they are almost entirely symmetrical, and could in theory have been worn either way up, it is tempting to wonder if this represents long wear over time by particular individuals, who either did not often remove their torcs, or had a preferred way of wearing them. Alternatively, the torcs could have been worn by several individuals who had largely been inducted as to the 'correct' way to wear them, suggesting they changed hands within a particular family or community. It is interesting to note that (assuming inferences based on plough damage are correct) all three torcs seem to have been laid in the ground the 'right' way up. This may suggest that they were carefully buried by someone who knew these objects, rather than as a hasty deposit.

All of the objects in the hoard are most likely continental imports into Britain. Continental imports from the 4th to 3rd centuries BC are rare, and gold objects even more so (La Niece et al., forthcoming). These torcs could have arrived as gifts, or as trade goods, but the most likely scenario is perhaps that they were carried to the region around the necks and wrists of their owners, perhaps continental women who married into the local community.

The discovery of this early group of continental torcs in Staffordshire has the potential to transform our understanding of this region in the Iron Age. The survival of these objects reminds us what a fractional picture we have of Iron Age societies in most areas of Britain. This tantalising hint of early continental connections re-casts the West Midlands as a dynamic centre with its own links to Europe. Perhaps the later development of a local style of 'cushion terminal' torcs represents a long tradition of gold ornaments in this area, rather than being merely a reflection of developments in East Anglia.

References:

Echt, R., Marx, M., Megaw, V., Thiele, W.-R., Van Impe, L. and Verhart, L. 2011. An Iron Age torc from Heerlen (Prov. Limburg/NL). Archäologisches Korrespondenzblatt 41: 31-50.

Eogan, G. 1983. Ribbon torcs in Britain and Ireland. In Clarke, D.V. & O'Connor, A. (eds.) From the Stone Age to the 'Forty-five: studies presented to R B K Stevenson. Edinburgh: j. Donald. 87-126

Eluère, C. 1987. Celtic Gold Torcs. Gold Bulletin 20(1/2): 22-37.

Fox, C. 1958. Pattern and Purpose: A Survey of Early Celtic Art in Britain. National Museum Wales

Hautenauve, H. 2005. Les torques d'or du second âge du fer en Europe: techniques, typologies et symbolique. Travaux du Laboratoire anthropologie, préhistoire, protohistoire, quaternaire armoricains 44: 7-340.

Hawkes, C. 1936. The Needwood Forest Torc. The British Museum Quarterly, 11(1), 3-4.

Jacobsthal, P. 1944. Early Celtic Art. Oxford: Clarendon Press.

Joachim, H.-E. 1995. Waldalgesheim: Das Grab einer Keltischen Fürstin. Köln: Rheinland-Verlag GmbH.

Jope, E. M. 2000. Early Celtic Art in the British Isles. Oxford: Clarendon Press.

La Niece, S., Farley, J., Meeks, N. and Joy, J. Forthcoming. 'Gold in Iron Age Britain', In Iron Age gold in Celtic Europe: Society, technology and archaeometry. Proceedings of the International conference in Toulouse, 11-14 March 2015. Forschungen zur Archäometrie und Altertumswissenschaft, Verlag Marie Leidorf.

Northover, J.P. 1992, Materials issues in the Celtic coinage, in M. Mays (ed.) Celtic coinage: Britain and Beyond, British Archaeological Reports British Series 222, Oxford, 235-299.

Painter, K. S. 1971. An Iron Age gold-alloy torc from Glascote, Tamworth, Staffs. Transactions of the South Staffordshire Archaeological and Historical Society 11, 1969-70, 1-6.

Raftery, B. 1983. A catalogue of Irish Iron Age antiquities. Marburg: Vorgeschichtliches Seminar der Philipps Universität

Stead, I.M. 1985. The Battersea Shield. London: British Museum Press

Stead, I. M. and Rigby, V. 1999. The Morel Collection: Iron Age Antiquities from Champagne in the British Museum. London: British Museum Press.