THE GOSFORTH MASTER

Six of the sculptures (nos. 1–3, 5–7) at Gosforth seem to be the work of the same master-craftsman (Bailey and Lang 1975). The remaining carving, Gosforth 4, expresses a similar interest in Scandinavian mythology, but its incised style of carving and its presentation of the human figure sharply differentiate it from the other material at this site. These six sculptures, which represent at least four separate monuments, stand apart from the rest of the work on the peninsula in the high quality of their carving and in the originality of their iconographic programmes. Within the area there is no other trace of the work of this hand; the mason seems to have worked solely for a patron at Gosforth.

The case for treating all of these sculptures as the work of one hand is both a negative and a positive one. The negative case has already been stated – nowhere else is there work of this quality and nowhere else do the stylistic details discussed below recur in the peninsula.

The positive case involves the recognition of stylistic details common to the six carvings: the figural carving on nos. 1, 5, and 6 is identical, with a pear-shaped head sunk deep into the shoulders; the Crucifixions of nos. 1 and 5 share a series of rare iconographical details; the tightly-packed, fleshy strands of plait, terminating in a tight spiral, which are so marked a feature of the main cross, recur again on nos. 2–3 and 5–7; a thin modelled, angular closely-knotted interlace is common to nos. 1 and 6; the drill is used on nos. 1 and 2. Similar zoomorphic details further emphasize how closely knit is the entire set of sculptures, the work of a master craftsman whose patron was a man of radical theological insights.

It is difficult to measure the impact of this artist, but it should perhaps be noted that the 'Saint's Tomb' hogback (Gosforth 3) exhibits two characteristic Cumbrian traits: tall slim proportions and the theme of serpent combat. It is not impossible that the origins of these regional forms lie in this technically impressive monument.

THE SPIRAL-SCROLL SCHOOL (Fig. 8)

Almost a quarter of the surviving Viking-age sculpture in Cumbria is the work of the spiral-scroll school. This distinctive group embraces twenty-seven carvings which are spread across sixteen sites. With the exception of Addingham 1 in the Eden valley, the school's work is confined to a thirty mile stretch of the coastal plain between the river Ellen and the river Irt (Bailey 1980, fig. 65). Most of the sculptures are worked in a Carboniferous sandstone which varies in colour between a pale yellow and a yellow/grey. This geological uniformity deserves notice, since other carvings in the same area draw equally on the red sandstones of the Triassic series; indeed it is particularly noticeable that at sites like Cross Canonby and Aspatria, which are on a geological boundary (see Figs. 4 and 8), there is a marked preference on the part of the school's sculptors for Carboniferous sandstones of the Triassic series. Such unity either implies that there was a common aesthetic agreement among the various masons as to the desirability of a yellow/grey surface or (since the carvings would presumably have been painted) it argues that many of the sculptors drew on a common source of supply for their stone.

The work of the school can be identified by a persistent combination of three motifs. The first is a cruciform pattern used on the cross-head which consists of a combination of central boss and ring(s) linked by a raised moulding to an encircled boss in each arm (Fig. 7). This was the pattern which Collingwood termed 'spine and boss' or 'lorgnette' (Collingwood 1913a). In variant forms it had long existed in Insular sculpture but the distinctive feature of the spiral-scroll school's usage is that, whatever the form of the central combination of boss and ring, there is only a single 'spine' moulding leading out to the arms and this consistently terminates in a small boss encircled by a single ring. In terms of the diagram and variants on Fig. 7, the distinctive cruciform head-pattern of the spiral-scroll school is type 2.

The second motif is the one labelled by Collingwood as 'stopped-plait' in which the strands of the interlace, instead of giving the appearance of passing over and under each other, are 'stopped' short of the crossing. The result is a patterning of short lengths of strand, the separate nature of each section being emphasized by its rounded end and by the fact that the ubiquitous median-incised line is entirely contained within the strand (see Fig. 6b).

The final characteristic motif, which gives its name to the school, is a form of scroll ornament; in this volume, this stylized vegetal ornament is designated as 'spiral-scroll'. It can be divided into two closely connected types. The most frequent rendering involves the transformation of a traditional tree, bush or spiral scroll (G.I., fig. 10) into an all-covering pattern of thin branches. The naturalistic basis of the scroll is deliberately confused by underplaying the distinction between the main and subsidiary branches, by detaching parts of the scroll from the stem, and by linking up shoots in a manner which is botanically impossible. Foliate elements are reduced to a minimum and great stress is laid on the way in which spiralling and trailing branches can take on the forms of symbols and line patterns. Haile 1, Beckermet St John 2, and Dearham 2 are fine examples of this treatment. Closely associated with it is the second type which consists of a single-stemmed scroll, emaciated and stripped of its foliage, which often adopts line-pattern forms. This single-branch type is particularly appropriate for use on narrow panels or for compositions involving parallel vertical strips of decoration: it can be seen on the cross-head Bridekirk 1, the edge of the cross-shaft St Bees 5, and on the central strip of both Aspatria 2 and St Bees 4.

The chart below emphasizes how frequently these three motifs are found on the same carving. Inevitably the chart under-represents the original extent of this co-occurrence since most of the crosses survive fragmentarily. Nevertheless the statistics are impressive. Of the nine surviving cross-heads in the group, seven carried the cruciform head-motif and it may have been present on an eighth. Twenty-four out of the twenty-seven carvings in the school carried spiral-scroll and in seventeen cases this is combined with stopped-plait; most of the other seven have only one face surviving or now exist only as very small fragments.

All three motifs have a background in Anglian art (Bailey 1974a, i, 49–57; and above, pp. 10; 15; 18). Stopped-plait, for example, though often attributed to imitation of metalworking effects (Collingwood 1927a, 64; Radford and Donaldson 1953, 14) is best seen as a development of interlace carved in the grooved technique (G.I., fig. 8d). This technique was already established in Anglian England (Peers and Radford 1943, pl. XXVc) and its popularity in the Viking period can only have encouraged the particular treatment which characterizes the spiral-scroll school. Similarly, grave-markers at Hartlepool and Lindisfarne and crosses at Ripon, Northallerton, Heysham, and Hartlepool, provide an Anglian ancestry for the cruciform head-motif used in the school (Collingwood 1927a, figs. 30, 104, 116(3), 128). Even elements of the eccentric spiral-scroll were already present in earlier sculpture, although most Anglian artists kept a rigid control over such tendencies. Yet as early as the scroll on Hexham 1 (Cramp 1984, pl. 172, 909) there is a deliberate juxtaposition of branches to form a potentially symbolic swastika shape as well as a linking of shoots which, botanically, should be distinct; the scroll of Kendal 1 takes on a triskele shape and another at Dewsbury even has a severed branch (Collingwood 1915a, 170, u). It is therefore hardly surprising to find that carvings outside Cumbria, such as Falstone 1, Northumberland, should display analogous developments to those seen in the spiral-scroll school (Cramp 1984, pl. 165, 881).

Whilst recognizing that spiral-scroll could represent an exaggeration of scroll treatments found across a wide area of Northumbria, it is possible that the Cumbrian school derived its main motif from a single local Anglian monument. The larger of the two round shafts at Beckermet St Bridget (no. 1) was clearly, to judge by its lengthy inscription and massive proportions, an ambitious monument. The presence of an inscription almost certainly points to its Anglian date. Three of its faces are ornamented with a spiralling scroll whose effect is very close (though not completely identical) to that of spiral-scroll. Since we can show elsewhere that Viking-age carvers copied and re-worked the ornamental repertoire of earlier crosses in their area, it is highly probable that the whole concept of spiral-scroll was taken over from this pioneering Anglian carving.

Given that the three basic motifs of the school are Anglian-based, it is pertinent to ask why the group is here placed in the Viking period. This question becomes all the more pressing when we note that eight of the nine surviving cross-heads in the school are of free-armed type, and thus in an Anglian tradition.

The dating relies on a combination of factors (Bailey 1974a, i, 57–69). First there is the evidence of Addingham 1, a carving which is geographically and geologically set apart from the other sculptures in the set. Though remote from the rest of the group it is clearly influenced by the coastal sculptures, and combines spiral-scroll and stopped-plait with a cross-head which is both ringed and hammer-headed, two clear indications of Viking-age date (see pp. 31–2). Whilst the other cross-heads are all free-armed, many of them have the square arm-pits and short arms which represent late developments. Supporting this is the evidence of the type of interlace on Beckermet St John 1 and St Bees 2 which seems to be a rendering of ring-chain, whose Borre-style associations have already been noted (p. 24). If this interlace is not based upon ring-chain then it is certainly arranged in two parallel vertical strips and this type of composition, where it occurs with diagnostic ornament, seems to be a popular one in the north-west in the Viking period. Among the line patterns adopted by the single-branch spiral-scrolls are T-patterns and battlement patterns which elsewhere occur in combination with Scandinavian-derived ornament. Swastikas, triskeles and the taste for incised cruciform shapes are also indicative of a tenth or eleventh-century date (Bailey 1974a, i, 63, 255). The accumulation of evidence thus points to the Viking-period centuries as the date of the school's floruit, but a closer dating is impossible; certainly the persistence of Anglian motifs need not necessarily imply that the carvings are work of the tenth rather than the eleventh century.

Within Cumbria there are two further crosses which seem to be linked to the main group. The first is Carlisle 4 which is the only cross-head outside the spiral-scroll group to use a cruciform head-pattern type 2a (see Fig. 7); its presence may reflect the usage of the coastal set or might possibly even have inspired the particular variant used in the school. More interesting is the case of Burton in Kendal 1, securely tied to the Viking period by its ring-head, which seems to ally the stopped-plait of the spiral-scroll school with the contoured interlace of the Beckermet school (see below, p. 38).

Outside the peninsula there are other sculptures which seem to be linked to this Cumbrian school. The most numerous form a set in Galloway and there is a further connected group around Govan, Lanarkshire; all employ stopped-plait (Collingwood 1922–3). Clearly there is a connection between Cumbria and these Scottish areas in the use of this distinctive motif, but the differences are also significant: nowhere in Cumbria do we find the encircled plaits of the Whithorn school nor the characteristic Galloway disk-head. Conversely the Whithorn group does not employ the parallel strips of vertical ornament so popular in Cumbria, nor does it use ring-chain or plain plaits, or intersperse its interlace with pellets and symbolic forms. The two areas may have shared a common inspiration but they developed that impulse in very different directions.

Only one Whithorn sculpture combines a stopped-plait with a spiral-scroll and intersperses its plait with pellets in a Cumbrian manner (Ills. 688–91). Collingwood placed this slab at the head of his Whithorn series, though its general incoherence is in marked contrast to the rest of the work from the Galloway area. Whilst not therefore accepting its typological primacy within the Whithorn series it does seem to mark an example of Cumbrian influence north of the Solway. Other traces of the school's impact are found in Peebles (Stevenson 1958–9, pl. XI(3)) and in Ayrshire, where a shaft from Colmanell closely echoes the ornamental organization of Dearham 2 and Aspatria 2 (Anderson 1925–6, 268). The Maughold stone in the Isle of Man combines spiral-scroll with stopped-plait and symbolic forms in a manner which immediately recalls Cumbrian fashions (Kermode 1907, no. 3) and is thus presumably a reflex of the English coastal group. Finally, a stone at Prestbury, Cheshire, is decorated with a form of spiral-scroll which is unique within its area (Croston 1884): it also appears to be another far-flung imitation of this populous Cumbrian school.

THE BECKERMET GROUP (Fig. 9)

The Beckermet group is made up of nine sculptures from the four sites of Beckermet St John, Haile, Workington, and Brigham. The group is defined by its use of interlace types which rarely, or never, appear elsewhere, together with the presence of certain decorative details which are never combined on other carvings. The following chart summarizes these combinations, an asterisk marking features which do not occur outside this south Cumbrian group.

The dominant interlace motif is a continuous run of Stafford knots (simple pattern E). Though this type of interlace frequently appears elsewhere on cross-heads it only occurs twice on Northumbrian cross-shafts outside this south Cumbrian area (Collingwood 1911a, 272(h); idem 1915a, 201(c)). The two asterisked interlace patterns in the chart above occur nowhere else. Double-incised strands are extremely rare in sculpture and the binding of parallel strands of interlace seems to be unparalleled. The bird-head terminal, with its long jaws and cabled round head, which is found on Beckermet 4 and 6 and Haile 2, is only the most distinctive of several zoomorphic forms which further help to isolate work of this school.

The Viking-period dating of the group is assured by the vestiges of a ring-head on Beckermet 6 and by the use of ring-twist and ring-knots. In addition, both contoured interlace and strands formed by three parallel strands (when they do occur on English sculpture elsewhere) seem to be of the Viking period. Finally the **bombé plan of Haile 2, a very ambitious grave-cover, is best paralleled in work of the tenth or eleventh century.

Despite this dating there is little in the ornamental repertoire of the group which owes anything to the art of Scandinavia, apart from some elements of the zoomorphic forms of Beckermet 4 and 6 which recall Jellinge details. There is one exception to this statement: the ring-knot on Beckermet 4. Its diamond shaped 'tie' reflects the general popularity of this form of ornament in the Viking colonies which stems from its use in the Oseberg and Borre styles. More significant is the trilobed extension at the top of the pattern, for these lobes are a persistent element in the art of Scandinavia from the pre-Viking phase represented by Style E through to its elegant apogee in the Ringerike and Urnes metalwork of Ireland (Salin 1935, figs. 615–38; Wilson and Klindt-Jensen 1980, figs. 7, 19, 42, pls. XXI–XXIIIa, XXV, XXXV, XLVIII). In these details the Beckermet sculpture stands alone in the English series and its presence may consequently betray either direct Scandinavian influence, or, more likely, an indirect impact via Manx carvings where it is frequently employed (Kermode 1907, nos. 93, 99, 101, 105, 108–10).

The single wing-like extensions to the same Beckermet 4 knot, as on the equivalent knot on Beckermet 3, probably had a similar (ultimate) source in Scandinavian art, though, in this case, there are analogous treatments in English manuscripts of the eighth and ninth centuries. Once more the detail is frequent on Viking-age sculptures on Man though it does recur on other Northumbrian carvings (Kermode 1907, pls. 8–9).

To a very minor extent, therefore, there is a Scandinavian contribution to the ornamental vocabulary of the school. But, like the work of the spiral-scroll group, this is essentially native art of the Viking period. Its main motif, the Stafford knot, indeed may well have been suggested by one of the border motifs of Irton 1; it is surely more than mere coincidence that this fine Anglian carving should (alone among known sculpture of the eighth and ninth centuries) employ the dominant interlace motif of the sculptural school which later developed in its vicinity.

The Beckermet group seems to have had some impact outside its immediate area. Kermode long ago recognized that a stone from Braddan on Man used a combination of knot motifs which were unique to the island but familiar in Cumbria; his comparisons were with work of this group (Kermode 1921, fig. 3). Across the Solway the Stafford knots at Boghouse, Mochrum, Wigtown, and at Old Kirconnel, Dumfriesshire, are probably also a response to the same Cumbrian school (Charleson 1928–9, pl. facing 129; Anderson 1929–30, fig. 3).