INTRODUCTION
Much of the ornamental repertoire of Anglo-Scandinavian sculpture in this region is derived from Anglian, or at least Insular, motifs. This is not surprising, since stone carving was not common in Scandinavia, except in Gotland, until the end of the tenth century. The taste for interlace, human figures, and animal ornament, prevails from the Anglian period but changes in mode and detail. Viking-age styles tend to be expressed in decorative detail and embellishment, rather than in the introduction of intrusively Scandinavian forms or motifs. Even those 'Viking' elements are sufficiently distinctive from mainstream Scandinavian art of the period to be designated 'Viking colonial' or 'Anglo-Scandinavian' (Lang 1986). About eighty per cent of the pre-Conquest carvings in the area were produced in the Anglo-Scandinavian period.
INTERLACE
The majority of pieces bearing interlace are from the Anglo-Scandinavian period. These tend to have dense plain plaits, or closed circuit patterns, usually of four cords. At York, especially on the Minster shafts (for example, no. 4; Ill. 22), the strand can be deeply modelled, humped in section, and median-incised. In the Plait and Pellet Group at York (see Chap. 10), the interlace works in very short runs, confined in small panels. Median incision is a simple means of conveying apparent complexity, even in this primitive series.
The patterns can be as complex as Anglian ones on occasion, for example, the encircled interlace of the coped stones from St Mary Castlegate (no. 5; Ill. 316) and Sinnington (no. 15; Ill. 819), or the tenth-century shaft, Minster 4, where it is found in association with quasi-Jellinge style animal ornament (Ills. 21–3). Some workshops in Ryedale produced incompetent patterns, especially in the bungled interlace of the Middleton crosses, where they often resorted to simplified, closed circuit forms, such as the ring-twist (also known as free rings and long diagonals) which, at a distance, gives the illusion of true interlace: see, for example, Sinnington 2D and Middleton 4B (Ills. 802, 687).
There was some local eclecticism: Middleton 3A has an attempted ring-knot (Ill. 682) which apes an accomplished model at Gilling West, North Riding (Cramp and Lang 1977, no. 10), a similar type of cross, but the workmanship falls short. Debased forms, such as the Ryedale version of stopped plait, are found on the Helmsley hogback (no. 1; Ill. 478) and Kirkdale 3 (Ill. 552), the elements mere right angles locked together. Pickering 3 (Ill. 759) and Kirkdale 4 (Ill. 542) indulge in clumsy buckle knots which are equally debased, though probably based on more competent carvings like the Sherburn grave-cover (no. 9; Ill. 791).
One site in the tenth century is distinctive in its repertoire from others in Ryedale: in Stonegrave the taste is for interlace rather than animals (Firby and Lang 1981, 17–20, figs. 1–2). There is much plain plait covering large surfaces, for example, Stonegrave 1C (Ill. 835). It also occurs on Middleton 3 in the same position (Ill. 684). A pattern so quick and easy to accomplish was naturally suited to unprominent faces. Some of the Stonegrave interlace remains very open in its organization, an Anglian trait, and there are links with late Anglian ecclesiastical sites to the far west (ibid., 25). This survival may explain the eleventh-century revival of interlace on such monuments as Minster 41 (Ill. 171), though north of the Tees the tradition was still thriving at the end of the tenth: see for example, Durham 1 and 12 (Cramp 1984,
Two ring-chains (G.I., fig. 26 Cvi–vii) survive: on Kirkbymoorside 3, and Pickering 3 (Ills. 523, 755). The latter has its Y-shaped vertebral element reversed, in the Cumbrian manner (Bailey 1980, 217; Bailey and Cramp 1988, 94), but both are crudely cut, derivative pieces. A Scandinavian trait is seen on the interlace of Hovingham 1 (Ills. 486–9), where the strands are flat serpents with heads reminiscent of those on the Thames toggle and the Jelling stone (Wilson and Klindt-Jensen 1966, pl. XLV, f, pl. XLIX). On the cross-head the interlace has Jellinge-style terminal heads with ear lappets fused into the points of the loops.
STRAIGHT LINE PATTERNS (G.I., p. xlvi, fig. 27)
Straight line patterns are uncommon in the region, and none of the examples is pre Viking-age. As elsewhere in Yorkshire, narrow fields were often considered appropriate for this type of ornament: for example, the rim of the dial at Old Byland 1 (Ills. 729–30), which carries a meander or key pattern. Pickering 3 and, more clumsily, Lastingham 2 (Ills. 758, 580), have meander patterns running up the face. The rustic Lastingham shaft also carries step and zig-zag and experiments in right-angled meander (Ills. 578–81) similar to that at Old Byland.
Stonegrave 1, however, has diagonal frets with swastika-like incisions, a pattern which occurs nowhere else in this region but does have western parallels (Ills. 834, 836; Firby and Lang 1981, 25). It is one of the features which make the site distinctive. They also occur on the socket at Stonegrave (no. 6; Ill. 845). It is likely that the tenth-century manifestations of such frets, as at Sockburn (no. 8), co. Durham, were a result of Irish contacts during the period of the Viking supremacy in York and Dublin (Cramp 1984,
PLANT-SCROLLS
The uninhabited scroll saw no demise at the advent of the Scandinavian settlement, for it appears in a tight spiral version on two scroll type hogbacks (York St Mary Bishophill Junior 7, and Kirkdale 9; Ills. 241 and 244, 553), which are uncompromisingly Anglo-Scandinavian monuments (Lang 1984a, 101, fig. 9). The York hogback's scroll still has vestigial triple berries in the manner of the much earlier Kirkdale cross-slab (no. 7; Ill. 558), though the spirals are tight coils. The easy transition of such dense spiral scrolls is possibly indicated by Hovingham 3 (Ill. 491), a ninth-century piece, and the complete cross, Middleton 3, which has a brave attempt at a rooted scroll with closely wound spirals, and a tendril as the only approximation to a leaf (Ill. 683). This type of scroll is also found in an Anglo-Scandinavian context at Brompton, North Riding (Collingwood 1907, fig. i on 301), where it is so organic that it has been mistaken for an earlier piece.
The 'belated' scroll of York St Mary Bishophill Junior 2 (Ill. 224), an exploded form (G.I., fig. 10), is rough work, but still nods in the direction of the botanical origin, affirming the continuation of the plant-scroll tradition throughout the Viking period. One oddity is the scroll at Sutton upon Derwent (Ill. 871), whose parallels are in Wiltshire on the Britford door-frame, rather than in the locality (Kendrick 1938, pl. LXXVI; Lang 1989, 3).
ANIMAL ORNAMENT (G.I., p. xlvi; fig. 7)
In the Anglo-Scandinavian period zoomorphic motifs came to dominate the ornamental repertoire. Both standing and recumbent monuments display a predilection for long, single panels filled with animal ornament: in York often a beast-chain (see below), in Ryedale, a single dragon. The origin of many of these creatures, like their counterparts in tenth- and eleventh-century Cumbria, lies in Insular art of the eighth and ninth centuries, the Scandinavian elements being confined to embellishments rather than affecting the form or disposition of the animals. The long traditions of zoomorphic patterns in both Scandinavian and Insular art approach enough of a similarity in the late ninth and tenth centuries to make ethnic attributions a dubious act of criticism. Much of the animal ornament of the Anglo-Scandinavian sculpture of York and eastern Yorkshire has been attributed to, or assigned with, the Jellinge style (Wilson and Klindt-Jensen 1966, 103–5, for example), or even used to argue for the development of that style lying in Viking contact with the Insular world (Shetelig 1954, 135–8). **Brøndsted had distinguished two kinds of animals in the sculpture of northern England, one the substantial 'Anglian beast' and the other the ribbon-like Jellinge animal. His view of the Anglo-Scandinavian experiments in this field was restricted to the interplay of these two types: '... the conflict between these two systems, the classic-naturalistic, represented through "the Anglian beasts", and the Germanic conventionalizing system introduced by the Vikings in the shape of the Jellinge style' (Br{*ø*}ndsted 1924, 195).
Recent studies, especially of the Middleton group (Lang 1973, 16–23; Bailey 1978, 177–9) have led to a modification of Wilson's view that the Ryedale bound dragons of the Middleton crosses were produced 'within a generation of Halfdan's settlement' (Wilson and Klindt-Jensen 1966, 104; idem 1980, 23). Perhaps the earliest animal decoration of the Anglo-Scandinavian period comes from the York Minster cemetery and the city's Metropolitan School (Lang 1978; see Chap. 10).
Grave-covers of the York type carry two types of beast: a winged biped with its lower parts resolved in interlace (Pattison 1973, 220, fig. 1, l–m), here termed the York winged beast, and a profile ribbon beast with body extensions (idem, figs. 1–2; see, for example, Minster 39; Ill. 165). In the Minster group, and the related slabs from the city's parish churches, there is little variation in the animals; indeed the bipeds are identical enough to suggest the use of templates and imply, through their confined distribution within the city, an exclusive use of the motif by a particular atelier.
The bipeds have an immediate Anglian source some miles to the west in Wharfedale. One of the early shafts from Otley, West Riding, has a large biped with wings, sinuous tail, mane, and beaked head (Ill. 921). It is foreign to all other Anglian beasts of the period and may well be a copy from an imported textile (Volbach 1969, pls. 21 and 57, for example). On the narrow face of that Otley shaft (Ill. 923) the creature is adapted to a more native Anglian mode by being given a head and chest at each end of the symmetrical design, and joined by an interlace ribbon body. The device can be seen employed at an early stage on Hackness 3 (Ill. 471). At Ilkley, West Riding, it was perpetuated in the ninth century, sometimes in a confronted pair, and sometimes raising the fore leg in a 'Mercian salute' (Collingwood 1915, fig. a on 186); compare the stances and disposition of animals from sculpture in the Midlands and elsewhere in Yorkshire for the stylistic connections here (Cramp 1977, 228–9, figs. 62–3).
It is a short step to the bipeds of the Minster slabs, although similar developments were current in Yorkshire and throughout Mercia (Cramp 1978, 21, fig. 1.2). The origins and evolution of the winged biped or York winged beast are in no way Scandinavian (Lang 1978b, 150–1), and comparison with the paired animals of the Brunswick Casket confirms the native development (Cramp 1978, fig. 1.2, n, q). The pendant tongues of the York winged beasts also have parallels: at Breedon, Leicestershire, and Elstow, Bedfordshire (idem, figs. 1h, 2s), both in late Anglian contexts.
The profile ribbon beasts are usually interlocked and fettered by body extensions, exemplified on York Minster 2 (Fig. 7h; Ills. 12–13). They have triple nose-folds and often spiral leg joints. Indeed in many respects they resemble Jellinge-style animals and at this period it is natural to view them as a response to Viking taste introduced from Scandinavia (especially in York). It can be argued from their method of construction, however, that their roots could equally well lie in Insular linear animal ornament (Lang 1984b, 50–1, fig. 4; Lang 1986, 254–7). Decorative elements which might well typify the Jellinge style, such as scrolled joints to the limbs, double outline, and filiform extensions trailing from the head and rump to interlace with the torso, are not exclusively 'Viking'. Neither is the symmetrical, undulating disposition of interlocked ribbon animals. Such features have a long ancestry in Hiberno-Saxon and later Insular manuscript decoration, their geometrical basis of construction being preserved and dictating the form of the design (see Lang 1984; idem 1986).
Certain details of the Metropolitan School ribbon beasts are restricted to the region, notably the nose-folds of their heads. These are not comparable with the trailing lappets of Jellinge-style animals: contrast the Jelling cup (Fuglesang 1982, 158–9, fig. 29a–b). The York snout is crumpled into three folds rather than sporting a slender appendage, a characteristic common to both the ribbon beasts and the more substantial, Anglian-derived quadrupeds. The half-moon nicks of Jellinge-style leg joints hardly occur in this region's sculpture, except at Sutton upon Derwent (Fig. 7j), and the scrolled joints are often loose incised spirals rather than the tight shell spirals of Viking art, exemplified in York Minster 9B, which has the associated bar familiar in tenth-century Scandinavian styles (Ill. 36).
In York, the interlocking of these profile animals within a long panel produces beast-chains. This arrangement also evolved from English sources, probably in Mercia (Lang 1978b, 149, fig. 8.3) where, in the ninth century, large, long-necked quadrupeds, often linked by body extensions, take on the diagonal posture which is reflected in the eastern Yorkshire beast-chains. The late Anglian quadrupeds of Mercia, especially those of the St Alkmund's shaft at Derby (Lang 1978b, fig. 8.3A), are the starting point: long-legged animals with large chests and elongated necks stand in profile, set one above the other discretely. The double outline is already apparent, as well as fettering ribbons and body extensions, median-incised. At Breedon (Fig. 7a; Lang 1978b, fig. 8.3B) such a beast is disposed diagonally within the panel, whilst its neck is entangled in the tail of the creature above. The full bodied animals of Folkton 2 and Nunburnholme 1 (Fig. 7b, d–e) are close to these Midlands pieces (Lang 1978b, 149; idem 1978c, 15, fig. 2), though the fettering and layout convey a new density which is typical of Yorkshire beast-chains. This type of beast-chain is found as far north as the Tees valley, at Gainford (Fig. 7c; Cramp 1984,
The thrown-back head of the Folkton 2 and Nunburnholme 1 beasts, as well as the diagonal disposition, underlie the more ribbon-like design of York Minster 2 (Fig. 7h; Lang 1978c, 15, fig. 2D). Apparently truer to the Jellinge style, the system on which this chain operates is identical with Insular examples, such as those on the Macregol Gospels (Hemphill 1912, pl. I; Lang 1986, 254–7, fig. 4, pl. 10), based on a diagonal grid, with the fix-points concealed in the scrolls. The shape of the eye and the bar-scroll on the hip do suggest a Jellinge-style milieu, even if the layout is common to both Insular and Viking art.
The ribbon version of the beast-chain takes on a looser appearance in the sinuous animals of Folkton 1 and its probable copy, Sutton upon Derwent 1 (Fig. 7f, j; Lang 1978c, 16, fig. 3F–G). The legs are shorter and the bodies are almost S-shaped. A further distinction is the manner in which the internal areas of the torso are decorated with hatching or pellets, comparable to the beast-chain of Odd's cross at Braddan on the Isle of Man (Fig. 7i) (Lang 1978c, 15, fig. 3E; Cubbon 1971, 39). Yet that feature is not necessarily Scandinavian, since it is found at St Alkmund's Derby, infilling the body of a bird (Lang 1978b, fig. 8.1B), in a ninth-century Anglian context.
The ribbon beasts of York Minster 2D, and the slab, Sherburn 9 (Ill. 791), tend to act as interlace strands rather than interlocking, sub-naturalistic animals, with the fetter or body extension serving as a filiform element, in the manner of the tubular beast-chains of manuscript decoration. The emergence of these chains in the Anglo-Scandinavian period is difficult to explain, for only Kirby Misperton 1 (Ill. 508) has an Anglian expression of this mode of animal ornament and there is no secure argument for local continuity. Contact with Mercia is the most likely source.
The interlocking sometimes serves to unite two or three beasts which are not in a chain, for example, on St Denys, York, (no. 2; Ill. 213), but it usually works in a linear way to fill the long, simple panels typical of York and eastern Yorkshire. On North Frodingham 1 the beast-chain actually intrudes into the cross-head.
The chain can also be composed of birds (Fig. 7g; Lang 1978b, 147, figs. 8.1–8.2, pls. 8.5–8.7) with drooped tails deriving from Mercian monuments, like those from Gloucester and St Alkmund's Derby (ibid., fig. 8.1; Brown 1937, pl. CXI; Kendrick 1938, pl. XCVII). With a single exception, Sutton upon Derwent 1, these are found only in York. The English roots of the sculptural beast-chains are to some extent corroborated by the shaft from Newgate, York (Ill. 343). The bipeds which form its chain are derived from southern English manuscript art of the early ninth century, as their domed heads, stumpy jowls, and broad chests testify: see Chap. 10, The York Master.
In Ryedale the long single panel is usually occupied by a solitary animal, for example, Middleton 1–2 (Ills. 674, 680). The head is thrown back and the fore leg crammed against the end of the frame. The most assured versions are at Sinnington (no. 4; Ill. 807), and they are probable models for those at other Ryedale sites (Lang 1973, 21–3). They have an S-formation, with the ribbon torso prominent over the legs, which serve as fillers. The jaws are distinctive open loops threaded by a band, a feature which develops at several other sites such as Ellerburn (no. 1), Levisham (no. 1), and Kirkbymoorside (no. 1) (Ills. 428, 633, 516). The distribution of these animals is very localized. Whilst clearly a local motif, the type of animal may derive from the larger York beast-chain (Lang 1978b, 150). There are two sizes, the smaller ones tending to be more competent and more ingenious in the use of a variety of filler shapes.
The use of the single long animal and the beast-chain must be associated with the habit of having a single panel on the face of the monument. Further north, in Allertonshire, North Riding, the fashion was for a series of small, squarish panels, with a consequent effect on the dimensions and posture of the beast.
Free style, or naturalistic, animals usually occur in narrative contexts, like the Sigur{*ð*} slab from York Minster (no. 34; Ills. 145, 147), or the 'hart-and-hound' motif at Ellerburn (no. 5; Ill. 432). The best example is Middleton 1 (Ill. 671), whose stag hunt is naturalistically represented, like its counterpart on the Stonegrave slab (no. 7; Ill. 861). This conforms with practice elsewhere in the north, for example, on the cross from Gosforth in Cumberland (Ill. 915), and the hogbacks of the Illustrative type (Lang 1984a, 138–9, 164–7). Realistic animals are rarely used for formally ornamental purposes. There is, however, a narrow distinction between 'free style' and 'rustic' when crude carvings such as those on the Stonegrave socket (no. 6) or Skipwith 1 are considered (Ills. 845, 823).
Three-dimensional carving of animals occurs chiefly as the dragonesque end beasts to hogbacks. That type is confined chiefly to eastern Yorkshire, extending as far north as Lythe, North Riding (Lang 1984a, 99). On the large ring-head from St Mary Castlegate, York (no. 3), a small dog crouches in each arm, with its head to the centre, resembling metalwork mounts (Ill. 302). It is an unusual form of animal decoration in the region, its nearest analogue being the ring-head at Bilton in Ainsty, West Riding (Collingwood 1915, 139–40, fig. b on 140), where human figures occur in similar positions.
FIGURAL ORNAMENT
Deep relief never really died during the Viking period, nor did portraits of Christ and saints. York Minster 4 carries, alongside its flat Jellinge-style profile beast and bird, a head with dished halo cut on inclined planes and bevelled with an abrasive stone (Ill. 20). This kind of modelled, full face portrait was employed by the First Sculptor of Nunburnholme, though in a more stylized mode, with schematized drapery based on Carolingian models and a scrolled halo. A complete bust of a saint and a Mass scene survive (Ills. 724, 728); both have sophisticated cutting features and stand out with curved profiles from a flat background. Their posture is stiff and unrelaxed.
The flat planes and profile view of Nunburnholme's Second Sculptor are a mark of ineptitude rather than change of style. The attempts to ape his predecessor in the Virgin and Child (Ill. 723) are unhappy but marginally less clumsy than the same scene on Sutton upon Derwent 1 (Ill. 868). The closely related shaft at Shelford in Nottinghamshire shows that at this period stylistic links were with the North Midlands rather than with the regions west of the Pennines. The warrior portrait at Nunburnholme (Ill. 721) is a typical example of this eclecticism for it is apparently derived from Holme upon Spalding Moor 1 (Ill. 483), just as Sutton upon Derwent is derived from four distinct monuments (see p. 000).
The full-face warrior portraits of Middleton (nos. 2, 4–5; Ills. 677, 686, 688), Levisham (no. 1; Ill. 631), and Kirkbymoorside (no. 2; Ill. 518), are local variants of a tenth-century Yorkshire tradition. Bailey has shown that they were produced using templates (Bailey 1978, fig. 9.3). An atelier based in Allertonshire, North Riding, produced profile warriors at Sockburn (co. Durham), Brompton, and Kirklevington (Cramp 1984,
Most Crucifixions in Eastern Yorkshire have an Irish aspect. At Ellerburn (no. 8; Ill. 437), Christ wears a long gown; at Kirkdale 1, he wears a kirtle. The Kirkdale Christ also has a forked beard and is bound to the Cross like the later Scandinavian Crucifixions (Ill. 546). Sinnington (no. 11) and Kirby Grindalythe (no. 4) both have the figure on the cross-head and have the large hands associated with Irish versions (Ills. 814, 503).
The great cross at Stonegrave (no. 1) depicts an orans holding a book and a cleric wearing a book-satchel (Ill. 833; Firby and Lang 1981, 25, 27, fig. 7). These are extremely rare in eastern Yorkshire and exemplify the distinctiveness of the Stonegrave cross, which looks to the west rather than to the locality for its borrowings (see Chap. 10). The figures do suggest a form of monasticism current in the Irish Sea province which, in the tenth century, could only be explained by a short-lived resurgence at the hands of Norse-Irish settlers.
Only one sculpture in the region bears a pagan scene: Skipwith 1 (Ill. 823). It shows Ragnar{*ö*}k, with a god's foot being assailed by a large canine beast. It may be compared with the Kirk Andreas slab on the Isle of Man and a hogback at Sockburn, co. Durham (Kermode 1907, 192–3, no. 102; Cramp 1984, 143–4, no. 21), but it is in no way a monumental carving. There are two stones, however, which depict heroic scenes from Germanic sources. Sherburn (no. 3) shows Weland the Smith in his flying contrivance and holding a woman above his head (Ill. 768; Lang 1976, 91–2, fig. 7). It is very close to two carvings at Leeds, West Riding (Ill. 920), and another on a hogback at Bedale, North Riding, a small but widely distributed group. Other Sherburn pieces (nos. 1–2) pick up the theme of birds, as at Leeds, perhaps with a similar iconographic intention. York Minster 9 has a naturalistic representation of the winged Weland (Ill. 35).
The large slab, York Minster 34, has scenes from the **Sigurð story on two of its faces (Ill. 145, 147; Lang 1976, 83–4, figs. 1, 3) which link it with the Manx series and the cross at Halton in Lancashire. One of the intrusive carvings at Nunburnholme (Ill. 728) may also illustrate the heart-eating scene (Lang 1977, 78). In both the eastern Yorkshire Sigur{*ð*} monuments the carving is in free style (that is, not formalized or stylized either in disposition or form), as it is in the other examples in Northern England and Man. Such pieces are rare, despite their interest, and they do not necessarily imply paganism (ibid., 89).
