GRAVE-MARKERS AND GRAVE-COVERS

GRAVE-MARKERS (Cramp 1991, p. xiv, fig. 4b)

There are few recumbent monuments or shrines of the pre-Viking period in this region, and there are no more than four upright end-stones to graves. Any attempts at generic grouping should therefore be cautious. Nevertheless, the two inscribed grave-markers from Wensley (nos. 8 and 9) are clearly from the same school; their size and the form of their relief crosses are close (Ills. 883, 884). They are carved on one face only and it is likely that their original context was within a building, though whether fixed to a wall or tomb, vertically or horizontally, will never be determined. The two markers are peculiar to Wensley and are very distantly related to the Hartlepool and Lindisfarne series (Cramp 1984, 7, 98–101, 202–7).

A solitary round-headed grave-marker, much damaged, survives at Stanwick (no. 8, Ill. 773). An early example of this type, Jarrow 10, has been dated to the late seventh or early eighth century (Cramp 1984, 110), though of a form that was still current towards the end of the pre-Conquest period (e.g. Aycliffe 12 and Elwick Hall 2, also in co. Durham (Cramp 1984, 46–7, 76)). There are no parallels for this form south of Stanwick, so it may belong to a Bernician tradition. A much larger upright grave-marker, too short for a grave-cover, was found at Gilling West (no. 9, Ills. 291–3); its cross type suggests a ninth-century date, and it is the only one of its kind in the region.

Only one stele of the type known from the York Minster excavations (Lang 1991, 18) has been identified. This was found at Whitby (no. 37), and carries an inscription but no cross (Ills. 1031–6). Its surface is plain apart from a double edge moulding.

GRAVE-COVERS (Cramp 1991, pp. xiv, xxi, fig. 4a, g)

Several plain grave-covers or -markers survive amongst the material from Whitby, including a fragment found during recent excavations on the site (no. 41). One of these carries a cross in relief (no. 38; Ill. 1044), but the others have incised mouldings of the type found on the Whitby Plain Cross group (see Chap. 6; nos. 39–41, Ills. 1037–42, 1046–7). Other slab fragments from the site may have had an architectural function (Ills. 1048–60).

The lost grave-cover Barningham 1 (Ill. 9) was initially described as a coffin lid (Longstaffe 1847), a convincing interpretation in the light of its shape and taper. Its sides carried ornament, like Kirkdale 8 in Ryedale (Lang 1991, 162–3, ills. 563–7), which might suggest that it was the lid of some kind of free-standing sarcophagus. Its ornament is of ninth-century date, but there are no analogues for its form elsewhere in Northumbria.

In the Anglo-Scandinavian period composite arrangements of recumbent monuments and standing crosses or end-stones were common in the area of Viking settlement. The so-called 'Giant's Grave' at Penrith, Cumberland (Bailey and Cramp 1988, ills. 494–5) and the cemetery beneath York Minster (Lang 1991, 28, ill. 417) demonstrate both the custom and the variety of the recumbent elements. At York the slab was used because it was convenient to use Roman ashlar from the headquarters building as a stone source. Geological bedding was no doubt a factor which determined the form of the later recumbents; certainly, flat slabs are much less common in the north part of Yorkshire than elsewhere in the county. Ormesby 2 (Ill. 714) is a simple slab with panels typical of grave-cover design but not of shafts, and from the latest pre-Conquest period two almost matching grave-covers, Gilling West 8 (Ill. 261) and Spennithorne 2 (Ill. 745), share a design which echoes that of slabs in the East Midlands with a double-headed cross, their nearest parallel being the rather degenerate grave-cover Sockburn 25, co. Durham (Cramp 1984, 154, pl. 153, 798), itself an oddity. The Gilling and Spennithorne monuments have no stylistic links with other slabs in northern Yorkshire. Crathorne 2 and 3 (Ills. 127, 128), though fragmentary, could well date from after the English takeover in the mid tenth century, when simple cross forms were applied to slabs in the York area.

HOGBACKS (Cramp 1991, p. xxi, figs. 5–7)

The most striking innovation in the region during the tenth century was the hogback, a house-shaped stone with a bowed roof ridge and often with bombé long sides. They are usually not longer than 1.5 m (5 ft) and in many cases were accompanied by standing crosses (Lang 1984a, 97). Their origins and models have been thoroughly debated (Collingwood 1927a, 164–73; Schmidt 1973, 52–77; Bailey 1980, 85–100; Lang 1984a, 90–7), with propositions ranging from Scandinavian wooden house-types, to Anglian stone shrine tombs, or Irish house-shaped reliquaries. Arguments for a single exclusive source are weakened by the variety of the hogbacks' typology: there are eleven types (Lang 1984a, 97–103, figs. 7–9; Cramp 1991, xxi, figs. 5–6), all of which are represented in the region covered by this volume (see Table 2). Indeed the distribution of hogbacks in England is at its most dense in northern Yorkshire (Fig. 8), with major centres like Brompton with eleven and Lythe with seventeen, suggesting that the form was initiated in this region.

Seven of the types have some form of inward-facing animal that often obscures the gable-end, and consequently the plan and profile of the house-shaped skeuomorph. In Allertonshire the end-beasts tend to be large and occupy much of the monument, and the niche and panel types (Lang 1984a, 97–9, fig. 7; Cramp 1991, xxi, figs. 5a, 5c–d) often have three-dimensional end-beasts of bear-like appearance. These constitute the only plastic stone sculpture of the period: they are true statuary, cut in the round and depicting an animal. The form was short-lived, probably only a matter of some thirty years between the Hiberno-Norse incursion of c. 920 and the English takeover in 954, yet it was a remarkable departure from the low-relief decorative carvings of the mainstream Anglo-Saxon tradition. Five other hogback types have rudimentary or heavily stylised end-beasts (Lang 1984a, 106–9) which resemble the gable-ends of many Irish portable reliquaries (ibid., 95), underlining the Hiberno-Norse milieu of the hogback in an area where so much is looking westwards to Cumbria and beyond the Irish Sea. Though it is possible to find Scandinavian parallels from the next generation, they are too late to be models for the hogbacks (ibid., 97). The model for the end-beast need not be the same as that for the house-form.

The typology of the region's hogbacks is not always restricted to particular sites or localities. Certainly the wheel rim type (type j) is restricted to two monuments at Lythe (Ills. 569–71, 577–9), with related monuments in the East Midlands, such as Shelton, Nottinghamshire (Lang 1984a, 162; Everson and Stocker 1999, ills. 472–7), and whilst end-beast forms, vestigial and dragonesque (types e and f), do occur on the east coast, the preferred form is the enriched shrine type (type k) or the simple house type (type i), at the sites of Easington and Lythe. This may have been determined by the Lythe workshop's need for vertical plain ends to receive the stumpy crosses (Chap. 6). Elsewhere, especially in Allertonshire, there are sites with more than one type: Brompton boasts three (types a, c and d). There is not necessarily an internal chronology implicit in the typology, and the quality of the cutting is variable. It seems that, as with the mixture of cross-head forms (see below), the sculptors in Allertonshire experimented with a variety of hogbacks, and the types were concurrent within the decades of their popularity.

Cumbria boasts five types (b, f, h, i and k) also found in northern Yorkshire. The pilaster type (type b) occurs at Wycliffe (no. 5, Ills. 1125–6), and at Aspatria 6 and Lowther 6 in Cumbria (Bailey and Cramp 1988, 53–4, 131), re-affirming the cross-Pennine connection across Stainmore. Similarly the enriched shrine type (type k) occurs at Gosforth 5 (ibid., 106–8), and, like the Lythe workshop's products, has a form and distribution of panels analogous to fine metalwork reliquaries. The type is almost certainly a skeuomorph of a reliquary shrine rather than a house. It is, however, the wide distribution of the scroll type (type h) which defies attempts to localise the various forms. Not only is the disposition of the ornament common to all examples, but their form unites them in being tall, slim monuments, very narrow with their height over three times their width. Type h is best represented in this region at Crathorne (nos. 4, 5 and 6: Ills. 133–7) and the lost Bedale 5 (Ill. 8), yet there are six in Cumbria, two in eastern Yorkshire, and one each in co. Durham and Derbyshire (Lang 1984a, 101). There is even an influential one at Govan in Scotland (Lang 1994b, 123). Given this spread it is impossible to determine the likely centre for the type's origin. It is a consolation that it implies far-ranging contacts between sculptors who may have been itinerant. The form of a type h hogback, laid on its side, is very reminiscent of some grave-slabs' proportions.

The only hogback forms confined to Allertonshire are the niche and extended niche types (types c and d), which may have derived from a large ecclesiastical shrine with side openings to reveal its relics (Bede 1969, 346, iv.3; Lang 1984a, 95); and the panel type (type a) which is restricted to Brompton, except for related fragments at Stainton and Wycliffe. Since the pilaster type so closely resembles the three Brompton panel type examples, it is possible that type b was simply an elaboration of type a, whose careful and naturalistic carving (probably by the creator of the hogback form) was so difficult to emulate.

FREE-STANDING CROSSES

CROSS-SHAFTS (Cramp 1991, p. xiv, fig. 1)

Perhaps the earliest examples of free-standing stone crosses in England are those from the monastery at Whitby (see Chap. 6). Their finely dressed surfaces and minimal decoration are closely related to the stelae from York Minster (Lang 1991, 18, nos. 11–24, ills. 44–102), their distinguishing feature being the integral cross-head with semicircular arm-pits. The Whitby crosses are monoliths. Their shafts are rectangular, even squarish in section, with little or no taper. The evidence of the dressing indicates that they very probably stood indoors against a wall (see Chap. 6).

The majority of Anglian shafts in this region are rectangular in section, ranging from late seventh- to early eighth-century pieces like Northallerton 1 (Ills. 662–4), through to ambitious monuments of the late eighth to early ninth century like Croft 1 (Ills. 147–52), Easby 1 (Ills. 193–212) and some products of the Uredale master (see Chap. 6). The distinction between squarish and slab-like sections (Cramp 1991, xiv–xv, fig. 1a and b) was no doubt determined by the geological bedding and is certainly not a chronological indicator, since Cundall/Aldborough 1 (type a) (Ills. 160–84) and West Tanfield 1 (type b) (Ills. 889–92) are by the same hand. Croft 1 is noticeably slab-like, as is Hauxwell 1 (Ills. 311–14), but all the other Anglian shafts in the area are more square: Northallerton 1, Wensley 1 and 2 (Ills. 858–60, 861–3), Easby 1, Wycliffe 3 (Ills. 1108–11), West Tanfield 2 (Ills. 893–6) and Cundall/Aldborough 1. The taper is usually gentle, except for Wensley 1 and 2 which diminish sharply to the neck of the shaft.

There is but one columnar shaft in the region: Masham 1 (Ills. 597–603). It is the lower portion of a very large monument whose upper section slotted into a large mortise in the top of the pillar-like shaft: a carpentry technique. If, as in Collingwood's reconstruction (1927a, 5–9, fig. 13.5), it is assumed that Masham 1 was of the 'staff rood' type, with an upper square section mounted on the lower cylindrical pillar, then the parallels for the form are many and far-flung: from Beckermet St Bridget 1 and Gosforth 1 in Cumbria (Bailey and Cramp 1988, ills. 42–5, 288–308), to Midlands crosses like Leek, Staffordshire (Kendrick 1949, 70, pl. XLVI.1), Reculver in Kent (Tweddle et al. 1995, 151–62, ills. 108–22), and many a Yorkshire Anglo-Scandinavian cross, spanning a considerable date range. The way in which the shaft's decoration is laid out in four arcaded levels suggests that earlier Continental models may have influenced the columnar form for Masham 1, sources which were architectural in function (cf. Ill. 1198). Its nearest parallel is the group of fragments at Dewsbury in the West Riding (Collingwood 1915, 162–3, figs. a–c), the ornament of which is related to that of the Masham pillar (Lang 1999, 273).

There are two examples of idiosyncratic cross-shaft designs in the region. West Tanfield 2 (Ills. 887–8, 893–6) is a complete yet small element of a composite shaft. At its top is a mortise slot, and at the base is a tenon. Its size precludes its completeness as a standing cross, and it no doubt accompanied similar elements which made up a tall tapering shaft. The device is clearly derived from carpentry joints and may offer a clue to the structure of wooden monuments now lost to us (cf. R.C.A.H.M.S. 1982, 197–201). Unlike many Irish skeuomorphs of wooden crosses (Kelly 1991, 105–45), West Tanfield 2 is not a stylistic reflex, indeed not a skeuomorph, since its mortise and tenon are structural. One cannot now judge the monument's stability in its original form.

The elegant tapering shaft at Melsonby (Ills. 654–61) is unique in its polygonal section. Originally perceived as two separate grave-slabs, its re-interpretation as an upright shaft by Rosemary Cramp is entirely convincing (Cramp and Lang 1977, nos. 7 and 8). The disposition of its ornament and proportions support that view (Fig. 18). It is essentially a rectangularly sectioned shaft with elaborate chamfers on its corners. The quality of the carving is in keeping with this innovative feature, yet the form does not recur elsewhere in the region. It is therefore possible that this shaft may have been a piece of church furnishing, like Jarrow 22 in co. Durham (Cramp 1984, 115–17, figs. 15–16), rather than a cross-shaft.

The majority of Anglo-Scandinavian shafts in this region are of rectangular section, and in the Brompton school and Allertonshire workshop (see Chap. 6) almost square. The taper is often slight. The side faces usually consist of long single panels, whereas the principal faces are divided into squarish panels which are in harmony with the section of the shaft. In terms of proportion, the form dictates the relative size of the ornamental panels. There is some indication that the shafts were not very high: no more than about 1.5 m (5 ft). The connection between the Allertonshire shafts and some in Co. Kildare, especially at Castledermot (Lang and Henderson forthcoming), is confirmed by the shared proportions and panelling fashion, but the Irish monuments are taller than the northern Yorkshire crosses.

There is a solitary example of a collared shaft in the region: Crathorne 1 (Ills. 129–32). Its ornament is Anglo-Scandinavian and Collingwood considered it to be the neck of a cross, with an unusual thicker block below the cross-head (Collingwood 1907, 304–5). There are no local parallels for this form, though there are two striking analogues in Cumbria: Bromfield 2 and Rockcliffe 1 (Bailey and Cramp 1988, ills. 173–6, 539–42). There are many stylistic associations between this region's sculpture and contemporary monuments in Cumbria; in the tenth century Allertonshire was certainly looking westwards rather than to York or Ryedale.

This relationship is also exemplified in the round-shaft derivative crosses (Cramp 1991, xiv, fig. 1, g–h), though these forms do indeed spread into eastern Yorkshire (Lang 1991, 29). A small fragment from the centre of a round-shaft derivative, Gilling West 2, preserves the upper rectangular section and the lower cylindrical section, separated by a swag in triple mouldings (Ills. 266–71). It is close in form to the swag junction of the Gosforth cross (Bailey and Cramp 1988, 100–4, ills. 292–5), and one of its faces repeats an otherwise unique motif of a chain of wings which appears elsewhere only on the Gosforth monument (ibid., ill. 305). Indeed, there are grounds for considering Gilling West 2 as by the hand of the Gosforth master (Bailey and Lang 1975; Bailey and Cramp 1988, 33). Another example, though in a poor state, may have been Kirklevington 14 (Ill. 433). Ellerburn 5 in Ryedale is an easterly outlier (Lang 1991, 128, ills. 41–6), and the type remained popular in Cumbria into the tenth century. The Peak District examples form a parallel school (ibid., 30; Collingwood 1923), and there are pieces elsewhere in Yorkshire which show the tradition with a long life span (Kendrick 1949, 68–76).

The other variety of round-shaft derivative uses the skeuomorph of a rope-like binding or collar mid-way up the shaft, often with a pendant decorative triangle suspended below the encircling band on each face (Bailey 1980, fig. 51). These 'vandykes' are also skeuomorphs: of metal appliqués used to strengthen wooden poles such as Irish crozier-shrines (Lang 1986a, 246–9). Both binding and vandyke are preserved on Gilling West 1 (Ills. 262–5), where the plain lower part of the shaft is sub-rectangular in section but the usage is clearly drawn from the round-shaft derivative. The vandyke also occurs on Stanwick 2 (Ills. 760–3), though without the binding or collar, and its lower edges are rounded off as an echo of the cylinder from which the shaft derived. Three further monuments in the region, Brompton 1 and 2 (Ills. 30–2, 33–5) and Hawsker 1 (Ills. 319–22), retain their rectangular tapering section but they carry vandykes and have unadorned lower parts to the shaft. The five examples of vandykes in northern Yorkshire, whilst decorating shafts in form like Penrith 4 and 5 and Beckermet St Bridget 2 (Bailey and Cramp 1988, ills. 47–51, 489–506), have no Cumbrian parallels. The Northumbrian vandykes occur in eastern Yorkshire on a pair of shafts from the same workshop: Lastingham 1 and Sherburn 4 (Lang 1991, ills. 574–7, 772–5), and there is an even more rare outlier on Bywell 1, Northumberland (Cramp 1984, pl. 162, 853–5). The group at Brompton, Stanwick and Gilling West are geographically close to four in the Tees valley: Sockburn 4, 7 and 8, and Dinsdale 1, co. Durham (Cramp 1984, pls. 33, 170; 129, 707; 134, 729; 136, 733–6), all associated with the Brompton school (Chap. 6). Since vandykes also appear on the great cross at Leek, Staffordshire (Kendrick 1949, 70, pl. XLVI.1), it may be assumed that the model for the Yorkshire round-shaft derivatives with collar and vandyke must have been Anglian; the absence of Cumbrian pendants must corroborate this view.

There is one almost columnar shaft, Bedale 1 (Ills. 10–13), which could easily represent the form as an Anglian survival in an area strongly influenced by Scandinavian colonial taste. Each face of the shaft is convex and the corners are so bevelled as to appear sub-cylindrical. The cabled edge mouldings for each face treat the shaft as though it had a rectangular section, when in reality it is almost oval.

CROSS-HEADS (Cramp 1991, p. xiv, figs. 2–3)

The cross-head forms of the Anglian period have little variety, and apart from the series at Whitby are relatively few in number. The heads of the early Whitby crosses are noteworthy in treating the lower limb as the extended shaft of the cross (Kelly 1993, 221; Bailey 1996a, 51), forming a Latin cross. It has been suggested that Whitby 2 (Ills. 900–2) is an isolate, but there are many unpublished fragments from the site that confirm that it is part of a Whitby series (see the catalogue, and Ills. 903–57). The arm-pits are semicircular, not exclusively a Whitby feature, and the lateral and upper arms are straight-tipped, with no squaring of the limb; the tip springs immediately from the curve of the arm-pit (Fig. 10). Recent finds reveal that the type did extend beyond Whitby: a cross-arm from Wharram Percy in the East Riding (Ill. 1194; Lang 1992, 43, no. 17, fig. 22) is close in form to the parent group, its arm-pit inclined awry, like Whitby; and a cross-head at North Otterington (no. 4, Ills. 698–702) differs only in its convex arm tips. The widely curving arm-pit is found on Northallerton 5 (Ills. 672–6), the parallels for which range from Ripon in the West Riding (Collingwood 1915, 233, figs. a–c) to a metalwork analogue in the Ormside bowl (Ill. 1196; Webster and Backhouse 1991, 173, ill. 134). The cusped cross-arm, type D9, is uncommon in this region, though the form does occur in the Whitby Plain Cross series (Chap. 6), for example Whitby 3 (Ills. 907–10). Almost a century later the cross-arms Masham 4 and 5 (Ills. 632, 636), which probably belong to the great shaft (Masham 1), are of the cusped type but richly adorned with subtle plant-scroll stemming from the shaft and through the neck of the cross into the head. The form, however, of the Masham pieces is almost identical with the Whitby examples.

During the Anglo-Scandinavian period, though free-armed cross-heads continued to appear, by far the most popular forms were the ring-heads, both types, and the plate-head (Cramp 1991, xiv, fig. 3). Finely cut ring-heads occur in the Brompton school (see Chap. 6), for example Brompton 9 and 10 (Ills. 51–4, 55–7), alongside numerous plate-heads which are ubiquitous. It may be chance that true ring-heads in the region of northern Yorkshire are often more accomplished or more adventurous carvings than the plate-heads, even though they share the same ornamental repertoire, but the plate is a more cautious option than drilling the arm-pit of the cross. The two forms may be expressions of quality and expertise. It is unfortunate that none of the Allertonshire workshop's shafts has an identifiable matching cross-head (see Chap. 6); again quality of carving may be the only clue. None of the ring- and plate-heads was a separate composite element; the crosses were monoliths and consequently the lateral limbs are often stubby and the arm-pits tight so that the cross-head has no great width.

The form of the superimposed cross on the plate varies greatly. Brompton 13 (Ills. 66–70), for example, exploits curves in the shapes of its plate, limbs, arm-pits and arm-tips in a controlled Anglian-derived manner, whilst Kirklevington 18 (Ills. 454–6) attempts a similar cross-form against a plate with almost vertical sides (they are only just slightly convex). One might assume that Brompton 13 was designed and constructed on a series of curves within a circle, i.e. perceived as a circular form, whilst Kirklevington 18 applies the curving cross-form to a basically rectangular plate. The North Otterington 5 plate-head is similarly constructed (Ills. 703–5). The ring-head Thornton Steward 4 (Ills. 801–5), and the plate-head Stanwick 7 (Ills. 768–72), both have type 1 right-angular crosses with plates, the one most competent sculpture and the other fairly rustic; the combination of cross-forms in a mix-and-match manner seems to have been typical in northern Yorkshire in the first half of the tenth century. Both the ring- and plate-heads have a Celtic derivation, in this region of north Yorkshire almost certainly Irish as the Allertonshire carvings prove (Lang and Henderson forthcoming). Kirklevington 15 (Ills. 438–41) has several Irish features, especially the upper limb which is longer than the lateral arms: compare the South Cross at Castledermot, Co. Kildare (Lang 1991, ill. 912; Harbison 1992, ii, fig. 101). Such influence is compatible with the probable Hiberno-Norse settlement of northern Yorkshire (see Chap. 2), and sits easily with the accepted dating of ring- and plate-heads as post-920 to c. 960 (Bailey 1978b, 178–9).

A fairly rare form, the billet-head (Cramp 1991, xiv, fig. 3), may also derive from a western source. Kirklevington 16 (Ills. 442–5) is the best survival. Together with its Irish Crucifixion position, it retains billets very similar to two crosses at Kells, Co. Meath, and two others at Monasterboice, Co. Louth (Harbison 1992, ii, figs. 337, 355, 481, 490). The billet's function is identical with that of the plate: to give stability at the most likely points of fracture on the cross.

Free-armed crosses continued into the Viking period: Easington 4 and 5 (Ills. 217–20, 221–4), Finghall 4 (Ill. 244), Gilling West 7 (Ills. 288–90), and Kirby Hill 7 and 8 (Ills. 349, 363–7), for example. The Finghall piece has iconography and decoration that would place it without dispute in the Anglo-Scandinavian period. The cross-arms are either curved or wedge-shaped, but there is no survival in the later era of the cusped form.

CROSS-BASES

Cross-bases, where they can be confidently attributed to the pre-Conquest period, occur at Whitby where they were discovered in excavation north of the standing abbey church (nos. 55 and 56; Ills. 1088–94, 1095, 1096–8). The finer of these shows architectural modelling. The shafts at Hauxwell and Hawsker may also stand in their original bases (Ills. 311–14, 319–22). A large socket stone is built into the fabric of the church at Kirby Hill (no. 11, Ill. 368), and could also be from before the Norman Conquest. Other cross-bases of uncertain date are noted in Appendix A.

ARCHITECTURAL SCULPTURE AND CHURCH FURNISHINGS

The two earliest fragments of decoration from stone buildings come from what might have been the monastic estate at Streanæshalch. Lythe 37 (Ill. 586) is a gable finial of the Irish type (Leask 1955, 46–7, fig. 20; Harbison 1970, 54–7, figs. 18–20), a form which occurs twice in Northumbria – at Heysham, Lancashire (Andrews 1978, 2; Cramp 1994, 106–8, figs. 41–2) and Lastingham in Ryedale (Lang 1991, 171–2, no. 9, ills. 610–13) – though the Heysham piece has alternatively been interpreted as the side of a chair (Cramp 1994). The minimal decoration and the smooth dressing of the Lythe finial is closely related to the series of early crosses at Whitby, from a period when the Irish tradition from Iona and Lindisfarne was still influential in the monastery.

Lythe 36 (Ills. 593–6) is a door jamb, decorated with late eighth- or early ninth-century interlace (Collingwood 1911, 288–9, fig. k), and a plainer stone found at Whitby (no. 63) was regarded as such (Peers and Radford 1943, 33): both indicative of embellished stone buildings at the neighbouring sites. A more ornate version is exemplified in Wycliffe 8 (Ills. 1116–18), which bears cluttered linear ornament.

Two pieces from the Whitby excavations (no. 52a–b, Ills. 1079–83), whilst recorded by Peers and Radford as crosses (1943, 39–40, nos. 22 and 23, pl. XXIV, a and b), are more likely to have been parts of a lintel (Lang 1990a, 5). Reconstructed as crosses they would have been unstable and could not have stood upright; their slab-like section also differs from all other crosses on the site. Each fragment has a step on the corner and a restrained continuous decoration along its perimeter. When viewed horizontally, they make perfect sense as a flat lintel that would have been incorporated into the wall above the doorway. They have recently been conserved together, very convincingly, forming one end of the lintel.⁴

Perhaps still in situ is the impost at Kirby Hill (no. 12, Fig. 16, Ills. 369-71; Collingwood 1907, 338, 343) which may be compared with the nearby stones in the north transept of Ripon Minster (Collingwood 1915, 233–5, fig. e). Its ornament suggests a rather earlier date than Collingwood's: in the first half of the ninth century (Adcock 1974, 147). This would indicate a stone church of the rare 'B' period at Kirby Hill, closely related to nearby Ripon (cf. Ill. 1197).

The only balusters in the region come from the Whitby excavations (Peers and Radford 1943, 33), one full-length and a second fragment (nos. 53 and 54, Ills. 1084–5, 1086–7). The larger baluster has entasis and also both raised and incised decoration like the Monkwearmouth series (Cramp 1984, no. 14, pls. 120–1) but more restrained. Its original fine finish is reminiscent of the surfaces of the plain crosses at Whitby (see Chap. 6). It has lost half its longitudinal section and may originally have been in half-round relief projecting from the masonry.

At Whitby the graves were marked by substantial stone crosses, often inscribed (Chap. 7), but other inscriptions could well have been plaques set either on walls or tombs; for example nos. 47 and 48 (Ills. 1061–4, 1065–6). They are in the tradition of dedication plaques such as Jarrow 17 (Cramp 1984, 113–14), which was perpetuated in much later examples like St Mary Castlegate 7 in York (Lang 1991, 99–101). The epigraphy of the Whitby plaques confirms their early date (see Chap. 7).

The most decorative piece is the plaque at West Witton (no. 1, Ill. 882). Whilst it might have been the end panel of a box-shrine, it seems more likely to have been a wall-plaque similar to Middleton 9 in Ryedale (Lang 1991, 187, ill. 694). It is hardly weathered so must have been displayed indoors. Its size and cruciform motif would be appropriate for its function as a frontal to an altar base, for which there are Manx and Scottish parallels (see the catalogue entry).

From the very end of the pre-Conquest period comes the plaque Newburgh Priory 1 (Ill. 653). Two figures sit within an arch, each an Evangelist. It has no local parallels, though the proto-Romanesque drapery is reminiscent of that of the figure beneath an arch at Sompting, Sussex (Tweddle et al. 1995, 178–9, no. 13, ill. 181). There is nothing to prove that it is part of a shaft and one should expect another plaque depicting the two other Evangelists. There is no early church in the immediate vicinity yet it is tempting to see this novel carving as a decorative architectural feature; it is too small for a shrine.

The sundial fragment at Skelton is similar to those in eastern Yorkshire (Lang 1991, 43), and carries an inscription, partly in runes, which provides evidence for patronage by speakers of Old Norse in the region in the eleventh century (see Chap. 7).

NOTES

4. See also the catalogue. The author's comments have been allowed to stand here, but in the catalogue entry, written subsequently by Rosemary Cramp, a different interpretation has been put forward. (Eds.)