Picts, Saxons and Celtic metalwork by Lloyd R Laing PICTISH BEASTS collectioe th n I f theno . National Museu f Antiquitiemo Edinburgn si h ther bronza s ei e buckle from Orkne 157 C 1/1g y(F fi ;, pi lla) bucklee .Th , evewhicn a ns hgreeha n patinas ha , confronted animal heads at each end of the loop, on either side of the strap bar. On the hoop is impresse dt ornament lineado d distinctive ran Th . e typ terminalf eo t oncsa e suggest affinities sit s well-knowe th o t n serie zoomorphif so c buckles from late Roma earld nan y post-Roman Britain, which have been discusse t somda e lengt catalogued han Hawkey db Dunnind san g (1961)e Th . type of buckle to which the Orkney example most closely corresponds is Mrs Hawkes' Class IIIA, which she distinguishes as having 'semicircular loops terminating in open-jawed animal heads confronted across the hinge bars' (Hawkes and Dunning 1961, 59). These buckles are rare in Britai n- Hawke Dunnind an s g record only eight examples l apparentlal , y fro e soutmth f o h England and all (with the exception of three chance finds without proper association) from late Roman contexts (1961, 59-60). These, and related buckles, were produced on the Continent in late Roman workshops, and Hawkes suggested that they were brought to Britain by Germanic federate fourte th earl d n si han y fifth centurie werd san e subsequently copie Britainn di , believing them to be evidence for Germanic soldiers and settlers in the late Roman period (Hawkes and Dunning 1961, 40-1). Althoughelw dno thas i t hi t such metalwork enjoye dwida e circulation ni the late Roman period, and was not merely used by Germanic soldiers and their families, the IIIA buckles were without doubt of Continental manufacture, as is attested by the finds from Haillot and elsewhere (cf Evison 1965, fig 4/8). Although the Orkney buckle might at first sight appear to be a Continental product, there are reasons for believing it to be a copy. Without exception, the IIIA buckles have animals with ope ne animalOrknee th jawsth d n o san y, piece have closed mouths. Secondly, the prototypes appear to have had metal attachment plates, while the wear striations on the bar of the Orkney buckle suggests it was attached to a leather belt which caused considerable friction. Thirdly, the shape of the animal heads, with snouts and receding chins, is difficult to parallel anywhere among the prototypes, but can be matched in the later but remark- ably similar chapes t Ninians froS e mth ' Isle treasure (O'Del t ale l 1959, pi XXXII). Whae th t Orkney brooch represents is a Pictish product probably directly inspired by a IIIA brooch. Wer confrontee eth d animal Orknee th f so y buckle uniqu Pictiso et h metalwork similarite th , y betweet Ninian'S e buckle th nth sd Islan e e chapes migh regardee b t fortuitouss da . Ther, eis however, a further series of objects from Scotland which have the same device of confronted animals. These are 'swivel rings' - rings with a perforated expansion or ring at right angles to the hoop, through which originally passe riveda t which allowe rine pivodth o g t t freely through 180°. On these swivel ring swivee sth grippes i l paia confrontef y dro b d animal survivinheadse th f O . g examples one unprovenanced ring differs slightly from the others (FC 128) in that a penannular | PROCEEDING 0 19 S OF THE SOCIETY, 1972-4 ring with animal terminals is threaded through a loop on the top of an openwork polyhedron (pi lla). This polyhedron, which has 20 facets, has projecting knobs or bosses at the angles of the facets e overalth , l effec tf strapworkbeino e gon t firsA . t sigh objece th t t closely resemblee sth openwork dodecahedrons from Romano-British contexts, suc thahas t from Fishguard, Pembroke (Collingwoo Richmond dan d 1969, pi XXId). These curious objects have been variously interpreted as candle holders, ornaments and surveyor's instruments, as the opposing pairs of holes are not equaf o proportionalle lar sizt ebu y relate could d an use e db sightins da g instruments (Collingwood and Richmond 1969, 316). This explanation can hardly apply to the Scottish piece, for there is no apparent relationship of this nature between the opposing facets and the object was clearly meant for suspension from its ring in one position only. It is also much smaller than its Roman counterparts. Most probabl harnesa s i t yi s pendanremainine th d an t g swivel rings shoule db FIG 1 1, bronze buckle from Orkney (FC 157); 2, bronze buckle from Haillot grave XI (after Breuer and Roosens and Evison) (2: 3) interprete horss da e fittings which would enabl echaia leather no r thon swiveo gt l freely without snarin g- suc h coursf o ring e esar usesimilaa n di r fashion today examplr leadsg fo , do .n eo Openwork consistin straa f g o domicap e patterth seee b n no n l nca moun t attache pivoa o dt t ring fro Hile mf Fortrith o l f Balnooneo , Inverkeithing, Banf 127C (F f) (pi lla 2/6)g ,fi . Thers ei ample evidenc Pictisr efo variouf o he horsemanshius se typeth r horsf so fo d e pan harnes s from Pictish sculpture - horsemen appear, for example, on the Aberlemno Churchyard cross, on a stone from Invergowrie Hiltoe th n Cadbolf no ,o Meigle somln stono th f d eo e ean crossese Th . Britons of S W Scotland too were horsemen, as is shown by the fragment of a bridle from Mote of Mark, Kirkcudbright (publication pending) poinnote,a e b o tdt since three swivel rings come from Glenluce Sands ,remainin e Wigtownshir fiTh threl e . gal se example2/25) o r e- , gfo 4 ,tw H e(B s come from the Hebrides, one from Vallay, North Uist (FC 267) and one from A Chrois, Tiree (FC 266) (fig 2/3, 1). LAING: PICTS, SAXON CELTID SAN C METALWORK 191 It is noteworthy that the style of the beast on the Luce Sands rings differs from those on the more northerly finds e LucTh .e animal-head e extremelar s y n reasonablstyliseca d an de yb interprete devolves da probabld dan y later versions remaindee th s .A r come from areas whice har either undoubtedly Pictish or probably Pictish in the fifth to seventh centuries (the period during whic swivee hth l rings were most probably made) reasonabls i t ,i assumo et e that the producte yar s of Pictish workshops, and that the Luce examples are British copies. FIG 2 1, ring from A Chrois, Tiree (FC 266); 2, Luce Sands (BH), Vallay3 ; Uis 267)N , ,anC 4 , Luc(F td;5 e Sands (BH); , rin6 f openworo g k swivel from Fortri f Balnoono e , Banff (FC 127) (2 : 3) There is no need to evoke a settlement in Pictland of Germanic soldiers of fortune or even wanderinw fe a g late Roman soldier accounswive e o st buckle th th d r l ean ringsfo t . Thee yar products of a general tradition of animal ornamented metalwork current throughout the non- Germanic areas of Britain in the period under review, of which one source of inspiration may have been late Roman metalwork of Vermand type. There are without doubt other sources of inspira- tion and these can now be considered. Within the limits of historical Pictland there were two categories of fifth-sixth century | PROCEEDING 2 19 THF SO E SOCIETY, 1972-4 animal-ornamented metalwork firse tTh . class comprises hanging-bowl escutcheons with animal- head attachments, a class represented by the well-known find from Castle Tioram, Inverness (Kilbride-Jones 1937, 208) (fi gimmediateln 3/1)ca t I . seee yb n tha treatmene th t animae th f o t l head on the Castle Tioram escutcheon is identical to that on certain of the swivel rings - the ears on the Castle Tioram beast are similar in technique to those on the swivel ring from A Chrois. questioe Th whethef no Castle rth e Tioram bow mads lwa takePictlans n i ei t belop nu no r dwo FI G3 1, escutcheon loop, Castle Tioram, Inverness (after Kilbride-Jones) , ox-hea2 ; d bucket escutcheon, Mount Sorrel, Lines (after Hawkes); 3, cruciform brooch, Malton, Cambs (after Aberg); 4, Anglo-Saxon finger-ring, Guildown, Surrey (after Hawkes) , termina5 ; f braceleto l , Freestone Hillo C , Kilkenny (after Raftery) (various scales) (Section II) where it is argued that it was, although it is not necessary to argue a Pictish origin bowe foth r suggeso lt t fro mt thai t such hanging-bowl animals were know Pictlandn i sucr fo , h beasts were not peculiar to the Castle Tioram bowl but appear as a recurring feature in the hanging-bowl series as a whole (for a convenient series of illustrations, see Kilbride-Jones 1937, fig 4). The second series of animal heads that were available for Pictish copyists comprises the penannular brooches with zoomorphic terminals, which repea idee f confronteth tao d heads. The most relevant brooches are those of Fowler's Classes E and F (1963, passim), the Scottish LAING: PICTS, SAXONS AND CELTIC METALWORK | 193 examples of which have been conveniently figured by Kilbride-Jones (1936, 124-38). There is no reaso supposo nbroochet F d Pictisf o an ee E tha sar h e originth t , though they almost certainly develope Nortn di h Britain durin Romae gth n period broochesF .