INTERLACE (Cramp 1991, pp. xxviii ff., figs. 14-26)⁵

The detailed study of this motif in Northumbria was first undertaken by Gwenda Adcock (1974) and this account is much indebted to her work. A significant body of the earliest material from the region, for instance from Whitby, is plain, and this relates also to the York tradition (see Lang 1991, 18, 23). The fine-stranded delicate interlace patterns which distinguish the early sculpture in Bernicia, at centres such as Wearmouth/Jarrow and to a lesser extent Hexham (Cramp 1984, 16–17) are only found in this area in the centres of cross-heads which are reminiscent of metal-work ornament, namely on Northallerton 5 (Ills. 672–3) or Masham 5 (Ill. 636). These monuments have close links with work at Ripon in the West Riding, which is perhaps the dominant ecclesiastical centre in north Yorkshire.

The most distinctive interlace from this region is the type which Adcock has called 'mature sculptured interlace' (1974, 92), which is not so dependent on metalwork models and which may be assigned to a period of creativity and experiment in the late eighth to early ninth centuries. These interlaces have a common unit measure of 3.5 cm and favour spiralled, surrounded and changing patterns (see Cramp 1991, xxviii–xlv). The West Witton slab (no. 1, Ill. 882) exemplifies the type as performed by a sculptor who was not fully competent, whilst the Easby sculptor produces interlaces on the narrow faces of the cross of the same high quality as the figural and plant-scroll panels (Ills. 204–6, 210–12). The strands at Easby are crisp and vertically incised with a median groove, and, as Adcock notes, the patterns chosen – half pattern F units with outside strands (see Cramp 1991, fig. 22) – are unique in sculpture although found on three pages in the Lindisfarne Gospels (Adcock 1974, 99). The control and elegance of the carving and the experimental variety of the patterns of interlace are of a piece with the whole design on the Easby cross, and this is markedly different from the heavy clumsy interlace on one narrow face of the Croft 1 shaft (Ill. 151) which is in stark contrast to the delicacy of the inhabited plant-scrolls on the other faces (Ills. 148–50). Nevertheless this interlace face at Croft is typical of the Deiran type in its changing patterns linked by long outside strands.

Interlace patterns which change and flow into each other, rather than being divided into neat panels, are boldly conveyed by a larger unit measure of turned pattern C and Carrick bends on a shaft from Wycliffe (no. 3, Ill. 1109), and on the closely similar carving at Melsonby (Ill. 659) with turned pattern F and pattern A. On both the broad and narrow faces of the Melsonby octagon the individual patterns are linked by long crossing or simple linking strands with smooth blank surfaces in between (Ills. 654, 660), a composition which is simplified and exaggerated in later work (see below).

There is however one notable monument, the Cundall/Aldborough shaft, where interlace is treated as a discrete and framed motif amongst other panels of animal, plant and figural ornament (Ills. 160–1, 163, 181). The unit measure of 4 cm is larger than the rest of this group, save for Melsonby, but the patterns favoured are patterns C and F with spiralled and twisted patterns A and D, which conform to the Deiran type, and the patterns are also linked by gliding strands against a plain background. The same technique of cutting as Cundall/Aldborough is to be found on one face of the impost from Kirby Hill (no. 12, Fig. 16, Ills. 369–70), but the design of three registers of encircled pattern D with loose pellets as fillers is one which is more closely paralleled in Bernician centres such as Monkwearmouth (Adcock 1974, 146–7). A further parallel with Bernician fashions is to be seen on Kirby Knowle 2 (Ill. 373). Panels of complete pattern E are popular in the more formal geometric interlace patterns north of the Tees (Cramp 1984, pls. 37, 189; 49, 234; 220, 1254) and this piece may imply some specific contact with that area. An individual linked pattern with fine strands, which is also not part of the northern Yorkshire regional tradition, is found at Ingleby Arncliffe (no. 1, Ill. 325), where the 'knitting stitch' is uniquely paralleled at York (Lang 1991, 110, ill. 370).

In the next generation, towards the end of the ninth century, the skill of linking changing patterns is lost. The shaft at Hauxwell (no. 1, Ills. 311–14), on which interlace is almost the only ornament, seems to represent an early stage in this process. This has a large unit measure, varying from 6 to 7 cm, and there are large open spaces between the units – on the broad face complete pattern A and on the narrow faces alternating half patterns A and D. This cross is very weathered, which may not do justice to the cutting, but the layout is skewed and the strands wavering when compared with earlier work. Hauxwell 1 has close affiliations with other monuments in the west, notably Wakefield in the West Riding (Collingwood 1909a, 185–7).

The Deiran motif of joining units with long glides is hardened and simplified in the Anglo-Scandinavian period in this region, for example at Stanwick (no. 1a–b, Ills. 754, 758) where on one broad face lines of twists are linked by double vertical strands with very wide plain surfaces in between, and at North Otterington (no. 3, Ill. 697) similar twists are linked by single strands, but the knots on the narrow faces at Stanwick are ring-knots – a feature typical of Viking-age art. Closed circuit patterns such as these, sometimes with a 'Borre style' link and bifurcating strands as on the Spennithorne slab (no. 2, Ill. 745), are typical of the tenth-century carvings. Simple four- or six-strand plain plaits become the norm, often accompanied by Como-braid twists, and, where competently carved, they are striking, as at Kirklevington 1 (Ills. 400–3). On the other hand at most of the sites where there had been earlier individual monuments there survive remnants of several smaller monuments carved with thick-stranded, grooved and close-packed interlace units, which represent the end of the tradition. In fact some form of interlace is the commonest form of ornament on monuments up to the Conquest.

PLANT-SCROLLS (Cramp 1991, pp. xxiv, xxviii, figs. 10–13)

Although there are few plant-scrolls surviving from this region, there is considerable variety in their typology. Moreover, many of the patterns appear to be contemporary, which makes for difficulties in establishing an internal chronology. One of the earliest monuments bearing uninhabited plant-scrolls is the shaft fragment Northallerton 1: its three carved faces carry respectively a medallion-scroll, a tree-scroll and a spiral-scroll, demonstrating that all three forms were current in the eighth century and that the sculptor was familiar with a range of models outside the region. The medallion-scroll (Ill. 662) has a naturalistic, organic appearance, its heavy blossoms and seed-heads pulled by gravity and the leaves being pendant too. The lay-out of the scroll is very open with some splitting of the stem. Conversely, the spiral-scroll of face C (Ill. 664) is very densely wound with an additional moulding cramming the space between the scroll and the border. In the case of this shaft both openness and density of layout are employed concurrently, a warning that horror vacui is not necessarily a chronological indicator. The tree-scroll of face B (Ill. 663) again contrasts with the other faces in its mirror-image symmetry and very rigid axial stem. The scroll forms a geometric progression of alternating pairs of simple scrolls tipped with small round berry-bunches and pairs of split plain leaves. It is in effect a rhythmic movement of a pair of motifs. Northallerton 1 belongs to a widespread group associated with monastic foundations across the whole of Northumbria, both in Bernicia and Cumbria (see Chap. 6). The sources for these plant-scrolls are not necessarily confined to the locality or even to the region as a whole.

The tangled scroll of Wensley 1 and 2 (probably the same monument) has something of an echo in Wycliffe 3 (see below), but the design is much wilder with no apparent underlying geometry controlling the stems (Ills. 858–66). There is more of a sense of upward movement at Wensley than there is at Northallerton, with seedpods and leaves reaching in every direction. The plant is organic with ridged nodes and thin tendrils which have attendant single leaflets. If there is a botanical source for this scroll it must be like convolvulus. The Wensley fragments have at least four leaf types and two seedpod types. The upper part of the pattern has the main strands flanking a turmoil of clinging shootlets. However, the narrow sides, like Northallerton 1, work in registers even though the leaf forms vary. Wycliffe 3 on its principal face has a form of scroll which is a halfway house between the medallion and the tangled plant-scrolls (Ill. 1108). Its stem is fleshy and well-modelled and it is based on a vertical axis. Face C of Wycliffe 3 has a simple but naturalistic scroll with nodes from which roundels erupt, and small six-petalled rosettes with gouged centres (Ill. 1110). There is an attempt at plasticity despite the formality of the controlled registers.

On Melsonby 1 and 2 the sculptor has confined the plant-scrolls to three of the chamfered faces. One face carries a simple scroll with ridged nodes and an abundance of leaves (Ill. 661). Another narrow face has fleshy stems expanding into nodes which almost resemble cornucopias, and its scrolls become more light and airy towards the top of the panel (Ill. 656). The third plant-scroll at Melsonby is the simplest of scrolls with slender tendrils and drop leaves (Ill. 658). The master carver at Melsonby conceived the monument as a whole rather than as four discreet panels, and all faces including the plant-scroll have denser design and more substantial motifs at the base of the shaft. There is an increasing lightness of design as the eye proceeds upwards. A similar open scroll containing trilobate leaves occurs on the impost, Kirby Hill 12 (Fig. 16, Ill. 371), a rare example of the motif (in this region) on decorative architectural sculpture. The Melsonby sculptor employed a wide range of plant-scrolls on a single monument in tune with the ever-changing designs of the total shaft. The scrolls are however all conventional enough, drawn from the Anglian repertoire of Yorkshire and Mercian monuments.

Whilst there are kindred motifs of the plant-scrolls so far discussed, at the turn of the eight century the Uredale master (see Chap. 6) and the two other adventurous sculptors who produced Croft 1 and Easby 1 developed a local expression of the motif. Croft 1, face B carries remains of naturalistic scroll with a ridge node from which sprouts a curved pointed leaf terminal (Ill. 149). This leaf form is fairly common in Mercia and on monuments of the late eight and early ninth centuries in Yorkshire; for example, Lastingham 10 and York Minster 10 (Lang 1991, ills. 38, 625), with further analogues in the East Midlands (Cramp and Lang 1977, no. 3). The inhabited plant-scroll at Croft has been turned almost into an architectural feature; its axial stem branches into only three registers so that the plant becomes a kind of arcade for the animals (Ill. 152). The nodes even resemble capitals of columns. This mixture of two fashions of the period underlies the Mercian / Yorkshire links in motif and organisation. The relative absence of leaves naturally facilitates the incorporation of the birds and the beasts, though the examples discussed above already show a disinclination for heavily foliated forms, and leaves tend to serve as terminals with few shootlets from the mid stem.

Masham 1, 4 and 5 were all produced by the Uredale master (see Chap. 6) and are very probably parts of the same monument, of which only the cylindrical base of the shaft and the arm fragments survive (Chap. 4). Nothing is known of the upper section of the shaft, but the face of cross-head no. 4 is covered with finely modelled plant-scroll which may once have stemmed from the shaft (Ill. 632). If this is the case then any iconographic significance of the scroll would be placed above images from the Bible and a line of apostles with Christ at the centre. This placing might convey an impression of the foundation of the Eucharist being associated with the true vine. The strands of the scroll are laid out geometrically about an axis. It is possible that Masham 5 (Ill. 636) represents the neck of the cross and the topmost part of the shaft. The typology of the scroll would move from a putative medallion on the upper shaft through to loops based on a Stafford knot at the neck. The interlacing shoots at the top of the medallion are strikingly more organised than other examples in the series. The leaf forms of Masham 4 are based on ivy and their scale on both nos. 4 and 5 is miniature.

The most elaborate plant-scroll in the region is undoubtedly that found on Easby 1 (Ills. 197–211). Kitzinger has argued cogently for the scroll's origins in the Carolingian world (Kitzinger 1936, 70–1, pl. VI). It is not a slavish copy, however: the leaves at Easby are fewer and smaller than the Carolingian analogues. On the narrow sides of the cross the alternating panels containing plant-scrolls and interlace may well provide evidence for continuity in layout and repertoire from the early eighth-century monuments like Bewcastle 1 (Bailey and Cramp 1988, 61–72, ills. 92, 93). The Easby plant-scrolls retain a glimpse of Late Antique forms which were being resurrected in Charlemagne's Europe as well as in Yorkshire and Mercia. It is the layout of the plant-scrolls and their adjacent ornament that maintains the tradition of Bewcastle and the prestigious monuments of the beginning of the century; it is the variety of the embellishment of Easby 1 and Cundall/Aldborough 1 (Ills. 160–84) in leaf forms, flowers and berry bunches which distinguishes the plant-scrolls of this region and demonstrates their southerly, ultimately Carolingian and Late Antique milieux. The eclectic habits of the region's principal carvers included filtering scroll patterns through native, contemporary European and Early Christian traditions of which they seem very aware.

The plant forms of the inhabited scrolls produced by the Uredale master reduce the tendrils even more to form an impressionistic roundel around the animal (Chap. 6). In some cases this is not organic and is really a trail, usually with a volute tip, with no junction to a main stem or root-stock. West Tanfield 1 separates out the animals within the panels, the scroll sometimes acting as a loose fetter only half framing the creature (Ill. 889). This encircling of the beast can sometimes form a transition to scrolled body extensions of the animal itself. This is a distinctive feature of the Uredale master: the plant-scroll has become rudimentary. This is most noticeable on Cundall/Aldborough 1, where on face C the unique layout of cruciform panels creates a succession of discrete animals, with the smallest vestiges of tendrils indicating their origin in the inhabited scroll (Ills. 162, 182).

The Uredale master also had a fondness for the bush vine, usually only of one register on the slenderest of stems with the loops disposed symmetrically. On Cundall/Aldborough 1 separate bush vines alternate with the cruciform panels containing animals, so that instead of a long uninterrupted progression of scrolls the effect is of checkers in small square entities. This effect is reinforced by the unique stepped borders (Ills. 162, 182). On Masham 1 the smallest of bush vines, almost sprays, are used to fill the spandrels of the arcades containing figure sculpture (Ills. 597–603). There is considerable variety amongst these fillers, which should be compared with the small symmetrical sprays from the Hovingham slab (Lang 1991, 146–8, no. 5, ills. 494–9) and the fragment from Castor in Northamptonshire, as well as the Hedda stone in Peterborough cathedral (Cramp 1977, 213–14, fig. 57b and c). Yet again the plant types used by the Uredale master and his Yorkshire contemporaries have more in common with contemporary Mercian sculpture than with counterparts north of the Tees, where bush vines are much less common.

In the Anglo-Scandinavian period plant-scrolls are found less frequently; but there is Anglian survival on Brompton 3 (Ill. 38), where a rooted scroll with gouged pointed leaves and rudimentary nodes convinced many, such as Kendrick (1938, 196) and Oakshott (1959, 117), that this was genuine Anglian work, though the corner locking-ring and template evidence show that the piece is from the Allertonshire workshop (see Chap. 6). Unlike many Viking-age scrolls, this does resemble the real plant with gravity affecting the drop leaves. The scroll is designed upon a grid with a one-inch unit of measure. The registers are ten inches high and the undulating stem is dictated by the diagonals of each register – the same technique that this workshop employed for figure carving. Whilst the design layout and geometry of the pattern have close links with Anglo-Scandinavian fashions in this region, such grids can also be found on earlier Anglian monuments such as Otley 1 in the West Riding (Collingwood 1915, 224–6, figs. a–i; Cramp 1970).

The thick fleshy scrolls of the Crathorne hogbacks, nos. 4, 5 and 6 (Ills. 133–4, 135–7), leave no room for leaves or other botanical features, and domed pellets are crammed into any space left by the scroll. The scrolls are very tightly wound and are the last vestige of plant-scrolls in the Viking period in Yorkshire, though all the Crathorne examples are on hogbacks of type h, which is the most widely spread type (see Chap. 4).

Stanwick 1a–b has an idiosyncratic plant-scroll which consists of a clumsy undulating stem with gourd-like fruit pods and rudimentary shootlets with heart-shaped leaves (Ills. 752, 756). This is apparently a rustic copy of an Anglian plant-scroll with its terminal blossom at the top. The associated ornament on other faces is equally peculiar, with only a ring-twist (Ill. 755) to suggest a date at some time in the tenth century. The work could therefore be late in the series, even extending into the eleventh century, as a revival of a perceived English motif.

ANIMAL ORNAMENT (Cramp 1991, p. xlvi)

The Anglian animal ornament of this region is of very high quality. It is chiefly found in inhabited plant-scrolls, and, whilst obeying the conventions in terms of layout, sculptors tended to play with the design of the creatures. On Croft 1, face C, six animals are disposed in three registers (Ills. 150, 152). A pair of confronting parrot-like birds constitute the lowest register, a pair of adorsed quadrupeds are in the middle and a pair of winged quadrupeds occupy the top. Indeed, the topmost pair are in effect amalgams of the other creatures in the panel. The birds and the avine foreparts of the transitional beasts have sharply pointed upright wings, though in each pair the sculptor has been careful to alter the lie of the wing in order to avoid strict mirror-image symmetry. Different types of feathers are carved in light relief, though the creatures stand proud and create an additional three-dimensional effect by the tail, wing and legs being caught in an 'Anglian lock' (Kendrick 1938, 141): for example, in the bottom left the tail passes over the scroll, the wing under it, the first leg over it and the second leg under it. Its confronting companion reverses this order. The huge talons occupy a great arc of the roundel, and whilst the birds are feeding, they are not perching but resemble star-shaped elements within the roundel. This is accentuated by the straight lines and tautness of the birds' design. The quadrupeds are leaping through the scroll and belong to that class of canine beasts which was shared by both Yorkshire and Mercia c. 800 (Cramp 1977, figs. 62–3). Nevertheless, the legs do not salute in the front of the creature's face, but whimsically the tail is made to do so. The sculptor here is playing with the convention. Again, the feet of the quadrupeds are exaggerated into elongated claws and hocks matching the talons of the birds below. The right-hand beast of the middle register prepares the way for the great leaping beast of St Leonard's Place 2 in York, where the last vestiges of the foliate scrolls are collapsing in the face of such dominant animal ornament (Lang 1991, 109–10, ill. 369).

Face A of Croft 1 (Ills. 147, 148) has in its upper panel a well-executed bush vine springing from a parabolic root and occupied by a pair of feeding birds, who do indeed perch with one foot whilst clasping the berry branch with the other. The birds are of the same type as on face C, and whilst they are certainly not peacocks there is enough detail in this panel to support the view that the sculptor was aware of the iconographic significance of the scene. It is a local development of an early Christian motif where the plant-scroll springs from a chalice, and birds representing eternal life feed from the Eucharistic vine. The layout should be compared with Jarrow 19, co. Durham (Cramp 1984, 114, pl. 98, 526). The lower panel has a pair of winged bipeds whose tails carry arrow-terminals reminiscent of the pointed leaves elsewhere on the shaft. From the canine head an extended tongue droops to a volute tip, much in the manner of animals at Elstow in Bedfordshire (Tweddle et al. 1995, 208–9, ill. 269). It is juxtaposed with an identical creature whose position is reversed (i.e. it is upside down) and whose tail wraps the first beast so that it can adopt the 'Anglian lock'. Immediately below, but separate from the first pair, is a pair of adorsed quadrupeds with similar pendant tongues and arranged in the same way as the winged creatures, one upside down. The tails are extended into a Stafford knot, both of which are made to occupy the base of the panel, thereby avoiding perfect symmetry but resolving the total pattern neatly at the bottom of the stone. Each pair are fettered by their own body extensions, but their disposition gives the illusion that they are joined by a common tail which is not the case; compare Otley 2 (Collingwood 1915, 227–8, fig. r; Cramp 1970, pl. 46, 3). This is lively animal ornament and whilst it is balanced it is not slavishly symmetrical. The sculptor must have been aware of current fashions in eastern England but adapts them to his own less regular mannerism. His patron must have been aware of the Eucharistic iconography, and perhaps not only of its immediate Carolingian models but also of the early Christian motifs which were being resurrected and filtered through Charlemagne's court at a period when Alcuin was deeply involved with the Emperor.

The polygonal shaft Melsonby 1 has on one of its broader faces a panel filled with animal ornament (Ill. 655) which yet again has similarities with Mercian and Yorkshire examples. Its distinguishing feature is the lack of either a plant-scroll to contain all the animals or even transverse mouldings which would separate each animal motif into its own panel. The upper parts of a pair of affronted quadrupeds are deeply carved so that the muscles on the neck and shoulders are realistically modelled, and where a particular tendon is taut it is raised into a modelled strand echoing the curve of the stretching neck. The solitary leonine animal shares this feature, with its attenuated neck liveried with a series of softly modelled ridges. Even its front legs have this tightness and it is a rare example of a sculptor at this period attempting to convey the anatomy beneath the skin. One of the forelegs is raised, for which there are Mercian analogues (Cramp 1977, fig. 51) as well as parallels in Wharfedale, West Riding (Collingwood 1927a, figs. 61, 63). The creature stands very erect but, unusually for an animal seen in profile, its face is turned to confront the viewer. It has a mane of tight curls and carries a serpent in its mouth: this must be one of the earliest manifestations of the 'combat motif' of later Anglo-Scandinavian art. Between the quadrupeds of this panel is a pair of biped dragons each with one leg raised, whose other leg and tail are stretched to link in a pair of loops, much in the manner of Otley 2 (Collingwood 1927a, fig. 60), though here the beasts are side by side and viewed from above. Their necks, too, have prominent strands of muscle. There is an interesting copy of this motif on the Anglo-Scandinavian shaft Brompton 3, where a cruder pair are disposed in the same manner (Ill. 39). The cutting is very deep and the edges bevelled to catch light and create shadow. The long undivided panel uses the animals to give the impression of upward movement in harmony with the tapering shaft, and the tension of the muscles is a device to heighten the effect of strenuous reaching.

It comes as no surprise that the most experimental animal ornament came from the hand of the Uredale master (see Chap. 6), especially on the large shafts Cundall/Aldborough 1 and Masham 1. The animals at Masham are isolated as individuals within an arcade in the lowest register of the shaft (Ills. 597–603, 625–31). Each consists of a standing quadruped perched on the slenderest of legs, with attenuated necks tapering to such a fine point that the head itself is often difficult to identify. Body extensions, especially around the head and neck, loosely fetter the creature in a light filiform strand. These slender extensions float upon open ground, so that whilst enmeshing the animal the beast has a sense of freedom and space. There are manuscript analogues in books such as the Canterbury Codex Aureus, now in Stockholm (Nordenfalk 1977, 106, pl. 38), but the exotic diminishing heads and twiglike growths about the legs are unique to this carving. Some of the animals are mirror images and it is more than likely that they are based on common templates. The template was reversed to establish the pairing, in the same way that the panel depicting Samson was reversed at Masham from its parallel at Cundall/Aldborough 1 (see below). This arcade of animals is the most baroque expression of the Mercian/Yorkshire quadruped of the late eighth/early ninth century; the nipped waists and long tapering neck and legs belong to the standard repertoire of animal ornament of that period (Cramp 1978a, 13–14).

On Cundall/Aldborough 1 (Fig. 14; Ills. 160–84) the Uredale master set himself the task of never repeating an animal form. He also experimented in filling a variety of panel shapes in such a way that symmetry would be impossible: face C of this shaft is unique in having cruciform spaces in which to accommodate individual beasts (Ills. 162, 182). Consequently, although there are four zoomorphic panels on face C, the Uredale master adopted four distinct strategies for filling them. Nonetheless, the basic form and proportion of the animal conforms to the Yorkshire and Mercian type with fleshy flanks, proud chests and a raised foreleg. One of these cruciform panels depicts a quadruped elegantly stepping down the chequered border (Ill. 174), whereas the one above is contorted as though it has slipped on the stairs (Ill. 173). A third panel on this face treats the legs and back in very strict diagonals, making the beast with raised foot as geometrical as the stepped border (Ill. 182). Other sides of the shaft such as face B have confronting beasts in mirror-image united by a single roundel which interlaces them (Ill. 181). The plant-scroll is subservient to the paired animals and the modelling of the flanks and chests is almost naturalistic. Together with the birds and creatures of face D they are the most conservative on the shaft, with large feeding quadrupeds leaping up into the plant-scroll in the manner found in Wharfedale and at Heversham in Cumbria (Collingwood 1927a, figs. 47, 49; Bailey and Cramp 1988, ill. 351). The animals of faces B and D have naturalistic features such as a foxy head and the realistic posture of a feeding bird with raised wing. The only fabulous beast is found on face D (Ill. 177), where a biped with a long tapering foliate tail is given the same modelling as its more realistic neighbours. The sculptor strove to convey diagonal thrusts, and owing to the deep cutting the relief carving of the beasts often approaches the half-round. A feature of Cundall/Aldborough 1 is that the zoomorphic panels are made to alternate with interlace, bush vines and figural scenes (Fig. 14). This accentuates the animals, since being discrete they do not have to be disentangled from the clutter of their plant-scroll context.

As in other parts of Yorkshire the growing importance of animal ornament at the expense of foliate designs was begun in the late Anglian phase in the early ninth century. As the Viking settlements burgeoned in the late ninth and early tenth centuries the secular sculpture of this region responded, with some of its embellishing details modified to Viking taste. Nonetheless, there appears to be a distinction between the animals of this region and those found in York and the east of the county. Beast chains are rare indeed, the only true example being Easington 1 (Ill. 190), and even that lacks the clutter of body extensions and the contortions of interlocking bodies which are features in the eastern group. Only two pieces from this region have a close similarity with those to the south of the North York Moors. One of these is Gilling West 1 (Ill. 262), which has other Ryedale features in its vandykes and ring-knot: it should be compared with the Middleton series, near Pickering (Lang 1991, 181–7). The flat ribbon with scroll appendages that constitutes the Gilling West animal has also the flat cutting of the Ryedale series and the same floppy S-disposition. The short scroll tendrils may well be a response to the smarter Ryedale animals such as Levisham 5 (ibid., 177–8, ill. 648). The animal could be regarded as an outlier of the Ryedale school, though its models were more likely the reflexes at Middleton than the Sinnington beasts, which were the real origin of this type of animal (ibid., ills. 674, 680, 804, 807). The hogback Pickhill 4 has a pouncing profile beast with double outline and body extensions, which even possesses a spiral joint on the hip (Ill. 737). This is by far the most convincing example of Scandinavian taste in the area, and the balance between clear Jellinge-style features against Insular parallels remains a matter for debate (Kendrick 1949, 90; Shetelig 1954, 136; Lang 1973, 16; Bailey 1980, 91). In terms of this region, however, the piece is a solitary with no local analogues. Double outlines are not exclusively Scandinavian, though it is more likely in this region that it is an expression of Scandinavian taste. The shaft Forcett 1 has a lorgnette cross accompanied by three quadrupeds, probably boars with spinal ridges (Ill. 250); yet despite this rustic naturalism each is given a double outline.

In the Lythe workshop (see Chap. 6) the taste is for plain naturalistic profile animals cut in flat planes, for example Lythe 1 and 19 which are possibly from the same template (Ills. 463, 533). Other parts of these two monuments have modelled cutting so it is likely that the flat animals are the result of stylistic considerations rather than geology. These creatures stand clear in their own panels with no embellishments and adopt realistic poses, especially the birds (Ills. 535, 553). This is in tune with taste further to the west in Allertonshire, where fettered creatures are rare, Stanwick 2 being an isolated example of a fettered beast (Ill. 760).

The isolated animal within a square panel is a notable feature of the Allertonshire workshop (see Chap. 6). The pot-bellied birds of Brompton 3C (Ill. 39) and Kirklevington 3A (Ill. 408) have been shown by Bailey (1978b, 180–1) to have been constructed from the same template that was used to construct the helmets of Brompton 3B (Ill. 38), Kirklevington 3A and also Sockburn 5 in co. Durham (Ill. 1189; Cramp 1984, 137, pl. 131, 715). The result is not a naturalistic bird, but its stylisation is rigidly determined by the half-moon template (Bailey 1978b, 184–5). Brompton 3C also has a pair of joined dragons seen from above (Ill. 39), and it is tempting to see this as an Anglo-Scandinavian copy of the monsters of Melsonby 1, only a few miles distant (Ill. 655). The Brompton sculptor had also borrowed a convincing Anglian plant-scroll which retains its traditional characteristics (Ill. 38). Local eclecticism, as can be observed in the hogback series in Allertonshire (Chap. 6), was very common in that area. The Allertonshire workshop has strong stylistic and iconographic links with monuments in Co. Kildare in Ireland (Lang and Henderson forthcoming), where solitary beasts occupy square panels in the manner of the Allertonshire workshop. At the time of writing it would be unwise to attempt an internal chronology for the Kildare and Allertonshire pieces, though crosses like Moone in Co. Kildare (Ill. 1200; Harbison 1992, i, 155, ii, fig. 158) pre-date the northern Yorkshire monuments. The inference is that these animal panels were inspired by the sculpture of Kildare.

The 'hart and hound' motif occurs on Forcett 1, Kirklevington 11, Melsonby 3 and Wath 4 (Ills. 251, 431, 651, 845–7). Whatever its iconographic significance these carvings are always accomplished in freestyle, which is usually reserved for narrative or symbolic scenes. The motif extends westwards, for example at Dacre 2 in Cumberland (Ill. 1193; Bailey and Cramp 1988, 91–2, ill. 245) and beyond to Lancaster and the Isle of Man (Collingwood 1927a, 150–2, fig. 171; Kermode 1907, 55, 64). It also occurs sporadically in eastern Yorkshire at Ellerburn 5 (Lang 1991, 128, ill. 432). It is tempting in the light of the Co. Kildare analogues to see the origin of the motif in the many hunt scenes carved on the sockets of Irish high crosses. The northern Yorkshire examples seem to have enjoyed a secular milieu without any monastic association.

The most innovative animal carvings in the region have to be the end-beasts of the hogbacks (see Chap. 4). There is a lot of variety in the size and appearance of these end-beasts, ranging from naturalistic muzzled bears at Brompton 17 to puny rabbits at Pickhill 3 (Ills. 82–3, 740). It is near impossible to establish a stylistic sequence for these animals in the light of their very brief floruit between 920 and 960 when the axis between York and Dublin was at its strongest. Indeed it may be possible that those beasts consisting only of inward-facing heads derive from parallels on fine metalwork shrines from Ireland, where the roof-ridge stems from the mouths of such zoomorphic terminals (Lang 1984a, 95). The innovation of a much larger beast could well mean the amplification of the head into a complete animal. This would explain to some extent the stylised bodies of end-beasts such as Wycliffe 5 (Ills. 1125–6), where the head is naturalistic but the flanks consist of panels of formal decoration as one would find in Irish reliquaries. The huge end-beasts, which can each occupy up to a third of the stone, resulted in what can truly be called free-standing sculpture. It was a device which did not develop, probably because of the English takeover in the middle of the tenth century. Had it not been for that, many of Yorkshire's late pre-Conquest churches might have been embellished with prominent three-dimensional animal carvings.

The work of the Lower Wensleydale workshop (see Chap. 6) is crude beyond measure, often festooned with dangling serpents, for example Coverham 1 and Masham 3 (Ills. 125, 646, 648). The work is so clumsy that the snakes may simply serve as fillers in place of interlace patterns which would be beyond the expertise of the carver. However, in the Stanwick and Forcett area there are examples of tightly coiled snakes, for instance Forcett 1, Melsonby 4 and Stanwick 4 (Ills. 251, 652, 765). The Forcett example is juxtaposed with a 'hart and hound' motif (see above) so there may have been symbolic significance intended.

FIGURAL ORNAMENT

During the pre-Viking period carvings of the human figure are restricted to a handful of monuments, and it is not until the end of the eighth century that the fashion develops. They are also restricted to the truly monumental crosses and do not occur on funerary sculpture. The subject matter is limited principally to historiated biblical scenes, as on Cundall/Aldborough 1 (Ills. 167, 180), and arrays of apostles on Masham 1 (Ills. 597–624) and Easby 1 (Ills. 193–6, 201–3). The monuments at Masham and Cundall are from the hand of the Uredale master (see Chap. 6) whose figures are in a miniaturist style: they have narrow straight sides and light drapery, though in both cases the monuments are very worn so that drapery details are hard to decipher. The apostles at Masham have a parallel in Hovingham 5 (Lang 1991, 146–8, ills. 494–9) but lack the movement and fluid postures of the shrine fragment. At Masham they are rigidly posed (Ills. 604–10) and this may have been dictated by their iconographic significance as 'pillars of the church' (Lang 2000, 116). These delicate figures are reminiscent of the small angels and saints in the Carolingian gospel book of Saint-Médard at Soissons (Hubert, Porcher and Volbach 1970, ills. 73, 76). A particular characteristic is the fall of the garments and the lifted wisps of drapery folds, suspended to one side. There is clearly an awareness of current European fashions, and the links through Alcuin with Charlemagne's Palace School of manuscripts suggest the most likely source for the Yorkshire examples of this style.

Cundall/Aldborough 1 (Fig. 14; Ills. 160–84) eschews the arcade containing individual figures which is a tradition that is more widespread in eastern Mercia, for example on the Hedda stone at Peterborough and a piece from Castor, Northamptonshire (Cramp 1977, 213–14, fig. 57b, c). At Cundall the busts of well-modelled angels are carved near the top of the shaft with a hint of half-profile (Ills. 164, 172). The well-rounded features and the holding of wands are comparable with fragments at Fletton near Peterborough (Clapham 1927, pl. XL). It is impossible to define an internal chronology between the Yorkshire and East Midlands examples, indeed it is more likely that both areas were indulging in a classical revival often filtered through Carolingian models. On face A of Cundall/Aldborough 1 are two rectangular panels, each with an arch containing a biblical scene. At the base is the Raising of Lazarus (Ill. 180) and above is Samson carrying off the gates of Gaza (Ill. 167). The carving is in a miniaturist style, and Richard Bailey has shown that the Samson scene which also occurs on Masham 1 (Ill. 615) is cut from the same template (Bailey 1996a, 114). The template was reversed for each monument in the manner of the animals in the lowest register at Masham.

The recognisable scenes on Masham 1 in the historiated arcade (Ills. 611–17) are not easily identified owing to erosion, though it is generally agreed that David the Psalmist (Ill. 614), David or Samson killing the lion (Ill. 612) and Samson at Gaza are sure identifications (Cramp 1965b, 9; Bailey 1972, 146). Contained within an arch, the scene of David and his musicians may be compared with the Canterbury psalter, BL MS Cotton Vespasian A.I., fol. 30v (Nordenfalk 1977, 95, pl. 32; Alexander 1978, no. 29, ill. 146), where the enthroned Psalmist sits frontally with his subordinate musicians grouped below him. The Masham version only differs in David sitting on a profile chair, though the lyre-like instrument is much the same and resembles that in the Cassiodorus commentary from Jarrow, Durham MS B.II.30, fol. 81v (Nordenfalk 1977, 85, pl. 27; Alexander 1978, no. 17, ill. 75). The relaxed postures of the figures and what remains of the delicate drapery folds have more in common with the Canterbury psalter scene and with the smaller figures of the Palace School of Carolingian manuscript art (Hubert, Porcher and Volbach 1970, 85, 89). The wafting drapery of David in the lion-slaying scene draws upon the same source (Ill. 612), as does the long pendant 'sleeve' of Samson at Gaza on both Masham and Cundall (Ills. 167, 615). Indeed, as noted above, Bailey has convincingly demonstrated that the two Gaza scenes came from a single template. The other historiated panel on Cundall/Aldborough 1 is at the base of the shaft, now in Aldborough, on the same face as the Gaza scene. Though it is much damaged (Ill. 180) it represents the Raising of Lazarus, as Collingwood first proposed (1915, 133). Christ stands tall at the left, whilst Lazarus emerges in his grave wrappings from a classical porch with pediment. Another figure kneels at Christ's feet. The scene occurs on a fifth-century ivory diptych from Milan and was reproduced in the ninth century on a Carolingian ivory, now in the Victoria and Albert Museum (Volbach 1952, 61–2, no. 119, pl. 37; ibid., 100, no. 23, pl. 63). An Early Christian sarcophagus in the narthex of San Lorenzo in Rome display the image in reverse. This is a striking example of the north Yorkshire sculptors drawing upon the antique, either directly from early Italian sources or from Carolingian models which were created with similar intentions in mind. The lay-out of the scenes betrays a high degree of accuracy in the copying of the models and an understanding of their iconography. The Cundall/Aldborough panels, one above the other (see Fig. 14), have a shared theme, for the Gates of Gaza image was often used as an ante-type for the Resurrection (Hawkes 1997, 156). Northern Yorkshire, c. 800 and the neighbouring decades, well beyond the hinterland of York, had ready access to current European theological trends; and the revival of classical cutting techniques, especially in the figure carving, shows a thorough knowledge of the Continental models.

The Apostles and two portraits of Christ on Easby 1 (Ills. 185–6, 193–7, 201–3) are half-figures or busts, so whilst the scale of the cross is neat and small, the figures have more space to be substantial, and their grouping, which is unique in the region, allows for the illusion of perspective. Their style owes much less to direct Late Antique models, as in the manner of Otley 1 in the West Riding (Cramp 1970), or the miniature style adopted by the Uredale master (see Chap. 6), than to Carolingian analogues: for instance the figure groups from ivories such as the cover of the Dagulf Psalter from the end of the eighth century (Hubert, Porcher and Volbach 1970, 227, ill. 208). In the overlapping of the figures and their individual heads there is a variety not seen in the miniature-style figures, and, underlying the apparent naturalism, there is a carefully planned logic to the overlapping elements which is as rigidly defined as an interlace sequence. Together with the vertical drapery folds, this gives a upward thrust to the design which matches the gravity of the Apostles themselves. It is an 'Apostle pillar' (Lang 1999; id. 2000), and the authority of its purpose, to assert the Apostolic origins of the true Church and to link these with baptism, is conveyed in truly monumental figure carving. The topmost panel of face A depicts the Risen Christ (Ill. 194), and on the cross-head he gives a blessing (Ill. 193). The iconography of the figures, unlike Masham 1 and Cundall/Aldborough 1, does not refer to incidents from the Bible nor to any symbols of redemption. An imposing pillar of stony figures, the 'pillars of the church', Easby 1's figure styles are entirely appropriate to its function, in much the same way as many Carolingian ivories and manuscripts from the Palace School of Charlemagne.

With the coming of the Viking settlements into northern Yorkshire, especially from c. 920 onwards, the patronage of stone monuments shifted from the ecclesiastical to the secular, and as in other parts of the country a taste developed for warrior portraits and lay persons. It is most apparent in the Allertonshire workshop (see Chap. 6): for example Brompton 3 and Kirklevington 2 (Ills. 38, 404). The workshop employed two distinctive styles. The profile warriors (Ills. 38, 408, 1189) were constructed on grids and with templates (Bailey 1978b, 184–5; Lang 1986a, 248); they are cut in low modelled relief in a flat plane. The clerics of Brompton 3 (Ills. 37, 40) and particularly the full-length portrait of Kirklevington 2 (Ill. 399) employ a much greater degree of relief which allows for more subtle features on the face. The modelling of the Kirklevington 2 figure is almost in the half-round, which, as Cramp pointed out, resembles Irish figure carving of the period (Cramp 1972, 147–8; see the catalogue entry). Facial features of the modelled type are a mixture of incised work and light facial relief. Brompton 4 and 5 demonstrate less expertise but draw upon the Allertonshire workshop tricks of style (Ills. 43, 44).

Warrior portraits were fashionable in eastern Yorkshire in the Ryedale school (Lang 1991), and also in Wharfedale in the West Riding, for example at Weston and Otley (Cramp 1982, 17, pl. 20). Civilian portraits are less common, but Kirklevington 2 is by far the most accomplished of those depicted; for the relief is subtle in catching light, and the feet stand on an inclined lower edge moulding which pushes the figure forward in an illusory manner from the containing frame (Ill. 399). Females are rare, though Kirklevington 6 depicts a couple, the woman with raised arms, gripped by the man who wears his sword sheathed (Ill. 420). This may be a narrative episode, and its flat planes comes close to free style. It has a very close parallel at Weston in the West Riding (Bailey 1981, 92, no. F16, ill. on 61). But it would be wrong to assume that ecclesiastical portraiture was totally absent, despite the large number of warriors on Anglo-Scandinavian shafts in northern Yorkshire. Brompton 3 has two clerics holding maniples in authentic mass vestments (Ills. 37, 40), one of them on an adjacent face to a typical profile warrior.

CRUCIFIXIONS⁶

There are no Crucifixion scenes from this area dating from the pre-ninth century, and most are combined with Viking-age ornament. This area is however particularly rich in examples of crosses where the head is used as a crucifix, and amongst the eighteen examples (see Fig. 9 and the Form and Motif Table) many are closely related in details of their iconography. This material has been fully considered by Dr Elizabeth Coatsworth (1979 and 2001), and both the individual entries and this section are closely based on her work.

Three of the examples – Finghall 4, Great Ayton 3, and Kirby Hill 7 (Ills. 244, 304, 363) – depict a naked Christ in a loin cloth, a type which was popular elsewhere in Northumbria and in the rest of England at this date (ninth to tenth century). The others display a 'robed Crucifixion' in which Christ is dressed in a long garment with sleeves reaching to the wrist. As Coatsworth pointed out, this robed type up to the ninth century is, in Northumbria, mainly Bernician in distribution, but the type does not continue in that area (Coatsworth 1979, i, 139–40). Instead its distribution in the tenth century coincides with areas of Scandinavian settlement, particularly in northern Yorkshire where it cannot be seen as a survival. Although, as she says, the popularity of the type could have reached Scandinavia from Ottonian art and then been transmitted to the Viking colonies, it seems much more likely that it came to the area with influences which had been transmitted to the Scandinavians in Ireland, along with a new form of monument, the ring- or plate-headed cross (see Chap. 4).

The region in Ireland which has produced examples of Crucifixions with very similar iconography and design is a small area west of the Wicklow Hills in Co. Leinster. The crosses which are most similar to those in northern Yorkshire are the North and South Crosses at Castledermot in Co. Kildare, but similar features are also found at Ullard, St Mullins, Ballyogan, Aghailten, and Moone (Coatsworth 1979, i, 142–5, ii, pls. 53–9). In all of these, characteristic features of the Yorkshire crosses are prominently displayed. These features are the flattened heads, with halo or hair flattened against the top of the cross-arm, and the long robe; the arms of the Crucified are straight and extended to reach the edge of the horizontal cross-arms, with splayed and prominently displayed fingers (see Ill. 1200, and Lang 1991, ill. 912). On two crosses in northern Yorkshire, Brompton 14 and Kirklevington 15 (Ills. 71, 438), the Christ figures have the characteristically flattened head and halo, but are set centrally in the cross-head. There is also a very crude version of this type of figure on Thornton Steward 5 (Ill. 807), which seems to have been affected by a type where the figure of Christ is placed higher in the arm of the cross (Coatsworth 1979, i, 133–7, type b).

This latter type is most competently displayed on Thornton Steward 4 (Ill. 801). It is a cross of some pretension since on the other face there is a seated figure of Christ (Ill. 803) and the figure there and in the crucifix is carved in deep relief. This may represent a new type in the area which is copied in cruder form elsewhere. The head of Christ does not reach the top of the arm and his elaborate halo may possibly be, as Coatsworth suggests, a crown (1979, i, 134). He wears a long-sleeved robe and is bound on to the cross, a feature which is known elsewhere in Viking-age art, for example a silver crucifix from Gotland (although here Christ is dressed in a loincloth and trousers), and it is noteworthy that another crucifix in gold filigree from Birka depicts a robed Christ with similar head type and the extended fingers so stylised that they appear as mere fronds at the end of the arms (Graham-Campbell 1980b, ills. on 188). This type of stylisation of the prominent fingers is found to a greater or lesser degree on all of the sculptures in the Yorkshire group, including two others at Thornton Steward (Ills. 788–9, 807). One cross, Thornton Watlass 2 (Ill. 813), depicts in a very abstract fashion a Christ figure with splayed legs: this has been taken by Coatsworth to be stylised folds in the garment (1979, i, 136), but could, like the Gotlandic crucifix, represent a trousered figure.

However crude some of these crucifix figures appear to us today, they may have delivered quite sophisticated messages to their contemporary viewers, many of whom must have been recent converts or even pagans. They could have served as aids to understanding the centrality of the Crucifixion in Christianity and also how the mass is its re-enactment. In several of the crosses such as Stanwick 7 (Ill. 768), Thornton Watlass 1 (Ill. 812) and 2, there is a prominent boss or circle in the centre of the body which plausibly represents the communion host or bread. This then could be a reminder that Crucifixion and Communion are inextricably combined in the Christian mass. It could also recall the well-known biblical words of John 6.51, 'I am the living bread which came down from heaven, if any man eat of this bread he shall live for ever'.

Beyond the region in the west there are parallels for this image, as these are for several other distinctive features in the sculpture of northern Yorkshire. In a cross at Lancaster (Collingwood 1927a, fig. 128) the centre of a robed figure, presumably of Christ, is shown on one face with a large central roundel, and on the other with five bosses which have traditionally been taken as representing the five wounds of Christ's passion (Bailey 1980, 148–9). Five such bosses are also found in the central roundel of the Northallerton 1 cross-head (Ills. 672, 673) and may have the same significance.

NOTES

5. The section on Interlace was written by Rosemary Cramp.

6. The section on Crucifixions was written by Rosemary Cramp.