The problems of identifying and dating Viking-age sculpture in northern England have recently been reviewed elsewhere (Bailey 1980, 45–75). In common with the situation in other areas Cumbria has no sculpture discovered in a firmly dated archaeological context. Several fragments have, admittedly, emerged from the foundations or fabric of the Norman period: Appleby and Great Clifton were re-employed as parts of Norman doors; St Bees 3, 4, and 5 came from a wall claimed to be of similar date and Gosforth 4 and 5 were found in twelfth-century foundations. Even if we accept these contextual reports unquestioningly however, they merely provide a terminus ante quem, and do not, of themselves, allow us to distinguish a Viking-age carving from a pre-Viking fragment.
Epigraphic evidence for dating any Cumbrian sculpture to the Viking period does not exist. The only inscriptions showing traces of Scandinavian-derived language in the area are of Norman or later date (Page 1971, 170–4). The lack of inscriptions on these sculptures is worthy of notice, particularly in contrast to the situation on the Isle of Man (Ill. 681; Page 1980), but offers no help in establishing a chronology for Cumbrian sculpture. The only sense in which epigraphic studies are relevant is in demonstrating the occurrence of certain motifs in the Isle of Man in the Viking age; when such themes also appear on Cumbrian sculpture, then there is an argument for assuming a similar chronological horizon for their employment there.
In the absence of sculptures directly dated by context or epigraphy we are forced to turn to stylistic analysis. The first step in so isolating Viking-age carvings in Cumbria must be to identify those sculptures which betray the impact of motifs and styles which were developed in Scandinavia during the period between Halfdan's capture of York and the middle years of the eleventh century. These successive and overlapping styles have long been the subject of intensive study and a widely accepted chronology, based on metalwork finds in coin-dated hoards, is now available (Wilson and Klindt-Jensen 1980; Wilson 1978). With certain reservations, this chronology ought to be relevant to the sculptural renderings of these styles.
The earliest of the Viking-age styles is the Borre, which flourished between the late ninth and mid tenth century (Wilson 1976a; idem 1978, 138). One of its basic motifs is a ring-chain, based upon split bands. Apart from an isolated (and not strictly comparable) example in The Book of Durrow, there is no trace of this motif in pre-Viking Insular work; it developed naturally from earlier art in Scandinavia. It follows that the ring-chain of Gosforth 1 and 2, together with the derivative forms of Bromfield 2, Rockcliffe, and possibly also Beckermet St John 1, imply that these are monuments of the Viking period. The same dating applies to a local Cumbrian development of the motif found on Gosforth 1 and 3, Cross Canonby 5, and Dearham 1, which employ a multiple form of ring-chain; this is clearly based on the single form used in Cumbrian sculpture and is quite distinct from the Manx type (Ill. 681; Bailey 1974a, i, 122, 398). Other varieties of interlace utilizing split bands, such as those seen on Waberthwaite 2, Dearham 1 and Bromfield 2, reflect similar (ultimately Scandinavian) tastes.
The Jellinge style may already have been in use in Scandinavia in the period when Halfdan captured York and the metalwork finds suggest that it remained popular into the second half of the tenth century (Wilson and Klindt-Jensen 1980, 118; Wilson 1978, 138–9). By contrast with York, the style is not represented in its pure Scandinavian form in Cumbria. Nevertheless the contoured ribbon animals of Great Clifton, and the beast with scrolled joints and contoured body on Aspatria 6, are recognizably in this style – their treatments of jaw and lappet are particularly diagnostic features. Jellinge also is a small beast buried deep in the tangled interlace of Gosforth 4. Less marked, but still in the same idiom, are the rhythmic beasts of Plumbland 2 and Cross Canonby 1. Yet further removed from the authentic expression of the style are the animals of Aspatria 1, Brigham 3 and 9, and Cross Canonby 1, where only in odd details (such as contouring, jaws, and spiral hips) are the beasts distinguished from their Anglian forebears.
The Mammen style, so well evidenced on the Isle of Man (Wilson 1970–3), scarcely appears in Cumbria. Claims have been made for Kirkby Stephen 1 as a reflex of this style (Kendrick 1949, 90), but only Workington 3 seems to merit that description and, on the metalwork evidence, should consequently belong to a date between c. 950 and c. 1010.
On these stylistic arguments some of Cumbria's sculpture can thus be dated to the Viking period. It is, however, a measure of the relative lack of importance of these Scandinavian styles in sculpture that this approach has given us dating brackets for, at most, seventeen stones. Earlier scholarship has understandably lavished much time on these zoomorphic themes, but this work is largely irrelevant to a study of most of Cumbria's Viking-period sculpture.
A further, rather smaller, group of carvings can, however, be gathered into the period because they carry depictions of Scandinavian mythology. Admittedly there are problems with these identifications (Bailey 1980, 101–42) and clearly we must bear in mind that Christian Anglo-Saxon England shared knowledge of many of these myths (Wormald 1978, 42–58). Moreover, on the Franks Casket, we have evidence that these themes achieved a form of sculptural representation in England long before the Viking settlement. There is, however, no trace of them in stone sculpture during the Anglian period and it is therefore reasonable to assign Gosforth 1, 4 and 6, together with the two Lowther stones 4 and 5, to the tenth century.
This dating does not rely solely upon the negative arguments that such myths never appear on Anglian sculpture. We have already seen that Gosforth 1 can be assigned to the tenth century because of its use of ring-chain; its presentation of a female with trailing dress and hanging pigtail must also be seen as a Scandinavian-derived mannerism supporting this dating (Ills. 692–3). It follows that Gosforth 2, 3, 5, 6 and 7, which, we will see, are all by the same hand, can be placed in the same period. Gosforth 4 carries a Jellinge beast and its figural scenes are carved in the same lightly incised manner as the Lowther stones 4 and 5; this implies that there is no great difference in date between them. What is more, since the boat scene on Lowther 4 is closely paralleled in Gotland, and since both Lowther 4 and 5 set their figural scene over the coils of an all-encircling serpent (surely the world serpent of Scandinavian mythology), these two carvings cannot be other than of tenth- or eleventh-century date. Presumably these mythological scenes were inspired by depictions in perishable media like the tapestries, woodwork and shield-paintings attested in the literature.
Beyond this point, where we have dated stones by the presence of Scandinavian-based styles or of Scandinavian mythology, we move to less certain ground. Certain motifs, whilst not specifically Scandinavian-derived are, nevertheless, regularly found in association with Scandinavian-derived ornament (Collingwood 1915a, 262–78; Bailey 1980, 58–75). Since, significantly, they do not seem to occur in English sculpture alongside motifs or styles which are pre-Viking in date, it seems reasonable to use them as Viking-period indicators. These motifs include free rings with long diagonals (or ring-twist), ring-knots and certain straight line patterns (G.I., figs. 24, 27), along with the so-called 'hart and hound' theme which, on English sculpture, is a Viking-period abstraction from the crowded hunt scene so popular in Celtic art (Bailey 1977, 70–1).
Sculpture can, then, be dated by its ornament. It can also be dated, in some cases, by its form. Two particular kinds of monument from Cumbria are important in this respect: the ring-headed cross and the hogback. The presence of a ringed form of head (G.I., fig. 3, 1–3, 5) has long been recognized as diagnostic of a Viking-period cross when it appears in an English context (Collingwood 1928c; Bailey 1978, 178; idem 1980, 70–1). The distribution of the form suggests that it was first introduced into England in the north-west and that its origins lie in some Celtic area, though which Celtic area is still a matter of dispute (Bailey 1978, 178). Recognition of the chronological significance of this type allows crosses like Addingham 1, Brigham 7, Burton in Kendal 4, Penrith 2, and the various circle-heads (see below, pp. 31–2) to be added to the tally of tenth- and eleventh-century carvings.
Collingwood's early work on the hogback argued that it was a Viking-period introduction, though his system of style-dating later obscured this conclusion and misled later writers (Collingwood 1907a, 276; idem 1915a, 284; idem 1927a, 164; Radford 1967b, 176–7). His case was based upon a survey of the ornament on these stones and remains valid. Further studies, notably by Lang, have only served to confirm this dating (Schmidt 1973; Lang 1972–4; Bailey 1980, 85–100; Lang 1984). Essential to the argument is the fact that the **bombé ground plan and curved roof of hogbacks echo the form of a type of building which, whilst it may have made a brief appearance in southern England during the early Anglo-Saxon period, seems to have been introduced to the north by Viking settlers. The distribution of this type of monument, with a notable absence from the Cuthbert community lands north of the Tees and from Mercian Danelaw, further suggests not only that this is a Scandinavian-linked form but that it may also be associated with areas settled by Gaelic-Norse groups. Lang has plausibly dated its development to the first half of the tenth century (Lang 1984). The hogback concept probably derives from a re-shaping of the traditional shrine-tomb of Anglian England to give it the appearance of a contemporary building, whilst one of the most prominent hogback features, the end-beast, can also be explained in terms of a development of a motif already present on Insular Christian shrines (Bailey 1980, 97–8). Cumbria shows certain regional variations on both the form and decoration of the hogback and these are examined below (pp. 29–30).
We have now isolated features whose presence can be used to identify a Viking-age carving in Cumbria: Scandinavian-derived styles; depictions of Scandinavian mythology; motifs used persistently in association with those styles and depictions; hogbacks; and ring-headed crosses. What must not be ignored, however, is that the majority of the motifs and monument forms of Viking-age sculpture are nothing more than developments from earlier Anglian art. Some motifs, indeed, have hardly developed at all; the affronted birds of Waberthwaite 2 differ little from the winged forms of the Gandersheim Casket (Wilson 1964, pl. 1a), yet the presence of split bands and Jellinge details on the cross show that it was carved some two centuries after the casket. Similarly the Jellinge-influenced beasts of Aspatria 1, Brigham 9, and Cross Canonby 1 are closer to Trewhiddle animals than they are to Scandinavian art. All the motifs of the spiral-scroll school (see below, pp. 33–6) were developed from well rooted Anglian treatments and even their eccentric Cumbrian renderings were already occasionally anticipated in earlier art. The continued popularity of the free-armed head in the area is but another facet of a strong Anglian tradition which is everywhere apparent in this Viking-age sculpture.
It will be clear from the foregoing that close dating of Viking-period sculpture is impossible. On the historical grounds outlined above (pp. 3–6) it seems unlikely that sculpture showing the impact of Scandinavian-derived art can pre-date c. 920. We cannot, however, know how long Anglian-derived motifs continued to be used in this 'village vernacular' art. Nor can we be certain that Borre and Jellinge styles did not persist in sculpture later than is attested by metalwork hoards. We may feel that the depiction of pagan mythology at Gosforth, allied to its presentation in forms so closely paralleled in Scandinavia, ought to imply an early tenth-century date. But such datings rest on unproven assumptions rather than demonstrable proof.
