a (shaft):

The shaft is edged by a well cut roll moulding. The top of the shaft is irregularly broken. A socket was recorded in it in the early nineteenth century (Lysons and Lysons 1816, cxcix), implying that the head was a separate stone dowelled into the shaft rather than having been of a piece with it.

An outline drawing (Fig. 11) shows the numbering of the panels and the location of the inscriptions.

Inscriptions The inscriptions are cut on the flat-band mouldings and other spaces between the sculptured panels of the cross, and on the north and south faces (D and B) at the head of the shaft. The east and west faces (C and A) at the head of the shaft now have groups of weathering grooves which may have originated in similar runic inscriptions, though no letters can now be identified there. Imaginative nineteenth-century investigators also found texts on other flat-band mouldings of faces B and D, but it is safe to ignore these. The surviving inscriptions are all runic. They may be read:

==R.I.P.

A (west, broad): The face is worn but the figures are deeply cut, in high relief from their background. (i) Worn panel, no decipherable text or ornament. (ii) A square-headed panel, edged on three sides by a fine roll moulding, encloses a standing frontal figure. The figure is slightly inclined so that his right shoulder and upper arm are pressed against the frame while the left side of his body is inclined away from the frame. He supports a lamb in his left arm and his right hand emerges from a fold of drapery and appears to be pointing across his body towards the lamb. The lamb is haloed; its body straddles his draped hand with head to the left and its face is turned towards the viewer.

The human figure is male, since there are surviving traces of a beard and moustache. There are no traces of a halo. The dress that he wears is a long full garment that crosses his chest in close-packed folds and covers his feet with soft voluminous folds. A scarf-like feature is draped between his arms with a heavy loop in front and sharp V-shaped folds at each end. (iii) A runic inscription (a). (iv) A round-headed panel, edged on the sides by fine roll mouldings, encloses a standing frontal figure. His right arm is raised and the hand is partly cut or broken away. His left hand holds a scroll which is angled across his body. He stands on the heads of two beasts.

The face of Christ, (identified by inscription (a) above), is slightly inclined. It is surrounded by a large dished halo, which may have faint cruciform incisions. His hair is long, smoothly combed from either side of a central parting, and falls in long locks behind his ears and on to his shoulders. The details of the features are worn, but he appears to have had delicately lidded eyes with lightly incised pupils. He wears a long garment which covers his shoulders and crosses at the waist, and possibly an under-tunic. The garments hang in heavy folds to his ankles, emphasizing the line of his right leg, and over his arm, ending in sharp V-shaped folds.

The animal heads on which he stands emerge from opposite corners at the base of the panel and their muzzles touch in the centre. Two of their feet are crossed and fill the triangular space between their heads and another two feet are raised towards the hem of Christ's garment. They have small pointed ears but no other details of their heads are visible. (v) A runic inscription (b). (vi) A round-headed panel edged at the sides by a fine roll moulding encloses a standing figure. He is half-turned, presenting his right shoulder to the viewer. His feet are pointing to the right but his head is turned back to the viewer, as though in arrested movement. In his right hand he holds a long rod or stick pointing to the ground. His left arm is extended in front of a front-facing bird. Underneath his outstretched arm, at about thigh level, is a T-shaped object. The man has long curling hair but the details of his face are difficult to distinguish and may have been recut. The eyes are very deeply hollowed but may not be in their original form. The man's dress is significantly different from that of the figures above. He wears an overgarment of a cloak-like type with a double 'frilled' feature around the neck. His undergarment is pleated at the front and is of lower calf length, and either has very full sleeves tapering in to the wrist or he wears heavy gloves.

The bird's head is difficult to distinguish. It is not clear now whether its head was originally turned towards the man or whether it was hooded.

B (south, narrow): The face is edged by broad outer and fine inner roll mouldings which, at the top, also enclose the five ornamental panels. The panels are sub-divided horizontally by wide, flat-band mouldings. (i) Plain, but there are some vertical incisions that could be meaningful. These may be traces of a fragmentary runic inscription (c). (ii) An interlace pattern formed from three registers of simple pattern F, turned horizontally, with outside strands and alternately-joined terminals. (iii) A complex plant trail springing from a single root. Two composite plant forms, in which two tiers of flowers are framed with a bell-shaped sheath, hang on either side of a ridged node above the root. Above the smaller composite flower on the left is a tiny trilobed berry bunch, and from the stem of the right flower sprouts a single half-moon leaf. The volute above contains a sundial attached like a large leaf to the stem and sprouting from above is a sheathed berry bunch. The hole for the gnomon survives and the dial is divided into four major divisions, each subdivided into three. The main lines (at 45, 90 and 135 degrees) seem to run to the edge of the semicircular dial, while the lesser dividing lines end in small circular depressions short of the edge. The space on the right above this volute is filled by crossed leaves, a variety of the serrated triangular type. The uppermost volute is formed from the looped stems of two composite plants – the pendent one a veined triangular leaf with florets at the base, the upper a composite of berries and sheathed stamens. The top corner is filled with a bold berry bunch sheathed with horizontal pointed leaves (see Fig. 5). (iv) A single register of complete outward-facing pattern C with terminals formed by simple pattern E knots. (v) A section of a complex interlaced medallion scroll which springs from two globular roots framing a type of fleur-de-lys. From each root spring three stems: from one sprouts two berry bunches; from another, a composite of sheathed stamens which shoot up into the volute above; while the third and wider stems form the medallion, at the base of which the subsidiary stems interlace. Two sheathed berry bunches hang from the top of the medallion and the main stems cross to form another smaller medallion above, with interlaced stems of pendent fruit and leaf-flowers at the top. The whole volute is surmounted by a pair of berry bunches framed in horizontal leaves. (vi) Possible traces of a fragmentary runic inscription (d). (vii) An eight- by twelve-cord pattern composed of a single register of double-stranded complete pattern C. The upper terminal is a pair of simple pattern E knots; the lower varies the crossing to form a register of pattern D. The box pointed loops in the centre form a cross shape.

C (east, broad): A double roll moulding frames a simple inhabited scroll which springs from a single root. The scale of the volutes and their associated creatures diminishes sharply towards the top of the shaft. The scroll terminates in a plant knot, (i), in which two composite sheathed stamens or fruit sprout upwards, while a large, elaborate leaf falls from the centre of the knot. The pendent leaf and the crossed leaves in the spacing below are mirror images of iii on face B. The next pairs of volutes, (ii) and (iii), contain tiny squirrel-like creatures, hardly bigger than the leaves; they are shown in profile right and left; each has its tail curled over its back and is nuzzling into a sheathed berry bunch. Dividing the volutes are veined and lobed leaves and a single berry bunch framed in curled-back leaves. Below, in (iv) and (v), are two birds of a thrush-like type. The uppermost faces right; its claws are braced against the main stem and its beak is half-open to peck at the stamens or seeds of a plant composite. Its wing line is delicately marked. Its tail fills the space between the volutes on one side; there is a pair of serrated and veined leaves on the other. The bird below is frontal, perched naturalistically on the volute, head turned to the left to peck at a sheathed berry bunch. The spaces below the volute are filled by plain triangular leaves with a raised central rib and a composite leaf with two rows of stamens. Below, in (vi) and (vii), are two bipeds facing right and left. Each grasps between its paws the stem of a sheathed berry bunch which it is eating; the tails loop over and under the main stem of the plant. The upper beast has small rounded ears, a punched eye and its tail has a double termination, first a fish-tail, and then a leaf. The lower beast's tail merges even more intimately into the enclosing plant and also terminates in a leaf. The nodes from which spring the volutes which enclose the three lowest creatures in the scroll appear to be ridged, and small split leaves are attached to the interior of the volutes. (viii) At the base of the scroll is a quadruped leaping up to grasp a fruit bunch in its front paws. It is very worn, but it appears to have a canine head which faces right. The details of the base of the scroll have been obliterated.

D (north, narrow): This face is the least weathered and so surface details of the ornament, as well as the inscriptions, survive well, although the edge mouldings are indistinct. (i) Traces of a runic inscription (e). (ii) A complex plant trail springs from a ridged root and terminates in a plant knot, the three stalks sprouting one trilobed and two triangular berry bunches. The strands of the knot below terminate in one lobed and veined leaf, and one veined with a flowery tip. Opposing it is a veined triangular leaf with florets sprouting to left and right below. From the lowest volute hang berry bunches, and at the base of the stem is a fan-shaped composition of stamens (see Fig. 5). (iii) A small squarish panel of twelve-cord interlace: a single register of pattern C with long loops and two added diagonals, and concentric edge breaks. The box-pointed loops in the centre form a cross shape. (iv) Twenty-five rows of alternate four sunken and four raised chequers. (v) A single register of double-stranded half pattern F with an outside strand and breaks. (vi) A runic inscription (f). (vii) A medallion plant scroll with two rounded roots from which the main strands spring and are 'clipped' together at the base and the top. The strands terminate in two composite leaf-flowers. The leaves are pointed and delicately veined, and a small rosette flower sprouts from between them. Below the top of the medallion the space is filled by two tiny opposing berry bunches, and below that, two composites of fanned stamens sheathed in wide spreading curled back leaves. At the base, the stems of two pairs of berry bunches or leaves form complex spiral knots and the space between the roots is filled by two small plant composites (see Fig. 5).

b (base): The base is now mostly below the modern ground surface, but was fully surveyed and repaired in 1891 by a competent master mason, Mr Baty, working on the instructions of chancellor Ferguson. He found that part of the base of the shaft within the socket had been broken and replaced by a loose stone. The shaft was held into the socket by lead. The base was found to be an irregular octagon in shape at the top (the sides parallel to the faces of the shaft being much longer than the others), and to splay outwards so that the chamfer peters out and the base is square at the bottom (Ferguson 1893b, 54–5).

Inscriptions [2]. The Bewcastle cross has spent its life in the open, subject to the attack of men and elements, and its inscriptions are severely weathered. Though they look promising at a distance, their condition is disappointing when they are examined close to – this applies particularly to the main inscription, (b), on the west face (Ills. 104, 116). Consequently, any reading given here can only be tentative; the linguistic forms given in the transcripts should be received with caution. Indeed, the various modern scholars who have examined the cross have produced rather differing texts; some have read fuller texts than can now be seen (and fuller than can be justified). The earliest drawings of the inscriptions – from the late seventeenth century – suggest that the runes were in much the same weathered state as now. Elsewhere I have suggested that in more recent times the inscriptions suffered further damage, perhaps from mischievous attack, also from the attempts of nineteenth-century scholars to clean and evaluate the texts (Page 1960, 50–1).

Inscription (a) (Ill. 103) is a form of the Latin Jesus Christus. Inscription (b) (Ills. 104, 116) is in a very bad state, and much of the transcription, notably the last four lines, is suspect. It is in part comprehensible as a memorial inscription, beginning: þis sigbecn [ . . ] setton Hwætred, [ . . ]þgær, . . . . . . . . æft [ . ]lcfri[ . ] (Translation: 'This token of victory (alternatively victory cross, victory memorial) Hwætred, . . gaer and . . . . . . . . set up in memory of . lcfri . .'). Thereafter, it is impossible to make any sense of the fragments, though the penultimate line seems to contain a form of the verb gebiddan (to pray). The two fragmentary inscriptions on face B, (c) (Ill. 108), and (d) (Ill. 111), are so badly weathered that only occasional letters are to be identified, and no reconstructions of the texts are possible. The beginning of the text of (e) (Ill. 109) is too badly damaged to be readable, but it is very likely that this is to be reconstructed: [ge]ssus, Jesus. The cryptic penultimate rune of (f) (Ill. 110) makes the exact form uncertain, but it is the recorded Old English feminine personal name Cyneburh.

Bewcastle compares with other inscriptions of north-west England in using the local rune calc (transliterated 'k'), and is particularly close to the Ruthwell cross in that it also has the differentiated g-rune gar (transliterated 'g͞'). It also has a letter-form of uncertain significance (given here by the asterisk).

Linguistically there is little dating evidence, partly because two of the legible texts are in Latin, partly because of the difficulties of reading the main inscription (b) – for example, some scholars have seen 'settæ where I have seen 'setto/n', and this makes it unsafe to use either verbal ending æ / -on, as a point in dating. Perhaps the only two inscriptions which can be used with confidence are (f) and the one on the lost cross-head (no. 7), if that was contemporary with no. 1. Their evidence is too slight to be of much use. They postdate i -mutation (cf. forms dryhtnæs – if that is the correct reconstruction of no. 7 – and kyni-), and also the Old English syncope of medial vowels, as shown by dryhtnæs. The condition of the unstressed vowels (Primitive Old English æ retained in ricæs, dryhtnæs, and Primitive Old English i in kyni-) suggests an early date, though three examples are little to go on. Runologically, Bewcastle is close to Ruthwell, Dumfriesshire, in their common use of 'k' and 'g͞' and, as far as it survives, its language is at a similar stage.

==R.I.P.

a (shaft):

A general discussion of the possible historical context of the cross-shaft and of its relationship to other carvings in the region and beyond may be found in the Introduction (pp. 19–22). Here the discussion concerns itself more closely with an analysis of the ornamental details of the monolith. The carver (and there does seem to be only one carver or master designer at Bewcastle), demonstrates a striking ability to relate his ornament to the form of the monument. His scheme of alternating panels of interlace and plant scrolls on the narrow faces could have become a mechanical formula, but he varies the order: on the south face, interlace, plant, interlace, plant, interlace; on the north face, plant, interlace, chequers, interlace, plant. In the centre of each face is a surprise break – the sundial on the south face, the chequers panel on the north. The differences in the panel heights are something that the Bewcastle master uses to provide a deeper variation, for example, Dii and Bv are the same height but their positions on the shaft and relative widths create a totally different impression (Fig. 11). On the whole, the carver manages very skilfully to vary the scale of the detail within the panels so that each makes an individual impact, whether high or low on the shaft.

Now it could be maintained that this form of theme and variation is characteristic of the layout of the best Insular manuscripts, for example, the Lindisfarne Gospels (Bruce-Mitford 1960, 255), or indeed the Durham MS. B.II.30, which is discussed below. The most obviously Insular motifs of the cross are the interlace types. Baldwin Brown recognized the similarity to (although not identity with) some of the interlace panels in the Lindisfarne Gospels (Brown 1921, 172–3, pl. XXIII). Such panels are typical features of a Bernician group of crosses which Adcock has called the 'designed panel school'. In this group Adcock discusses the Bewcastle interlace at some length (Adcock 1974, i, 158–65). She notes some similarity with a Lindisfarne, Northumberland, cross-head, (Cramp 1984, pl. 195, 1099, 1101), and also notes that the six-cord double-stranded pattern occurs in the Lindisfarne Gospels, fols. 95r and 211r, but states: 'the essential concept does not come from the Lindisfarne Gospels although the well gridded double-stranded patterns have a superficial similarity. The concept does have a great deal in common with the work of the Durham 'Cassiodorus' artist, especially on folio 81v.' (Adcock 1974, i, 164). She feels that the likeness is so striking that manuscript and cross must have been produced in the same generation and probably at the same centre. She also sees similar interlaces at Rothbury, Northumberland (Cramp 1984, pl. 214, 1222) and Jedburgh, Roxburghshire (Adcock 1974, i, 168) as strongly related but more developed, so they should fit somewhere into an atmosphere of continuous and creative traditions not greatly separated in time. Bewcastle she sees as contemporary with Abercorn, W. Lothian, but rightly stresses the difference between the motifs on the Bewcastle cross and the Insular frets and interlaced animals or birds on the Lindisfarne/Abercorn/Aberlady group (Cramp 1984, 27). Bewcastle, then, in the use of interlace, is exploring an Insular form, but in the choice of pattern and details of development, is not closely linked to the centres which favour other Insular motifs. The panel of alternately sunk and raised chequers (Div) could be seen as an Insular motif, since coloured chequered patterns occur in millefiori inlays (Bimson 1983, 928–33, figs. 664–5) and in manuscript illuminations such as the Book of Durrow (Nordenfalk 1977, pl. 4). Nevertheless, such chequers occur in stone carvings from the Roman period to the Norman. The fashion, however, seems to have been directly transmitted from Roman stones at Corbridge to carvings at Hexham, where they decorate a capital (Cramp 1984, pl. 184, 1003). Baldwin Brown also noted the resemblance between Bewcastle and a fragment from Hoddom, Dumfriesshire (Brown 1921, 171–2, fig. 13, 3).

The sundial on the south face is the only one on a cross to survive from the pre-Conquest period. The others, when in situ, are set into the south-facing walls of churches, as at Escomb, co. Durham (Cramp 1984, pl. 56, 277) or Kirkdale, Yorkshire (Collingwood 1907a, 347). The solitary Welsh and the early Irish dials (Hamlin forthcoming) are found on free standing pillars or slabs, as were also some Roman dials, and it could be that the Bewcastle cross is to be seen as part of the 'free standing' tradition.

The octaval division of the dial is the most frequent pre-Conquest form and the division into twelve hourly segments is rare. It has indeed been suggested that the original fourfold division has been subsequently subdivided (Aked 1973, 501–4). The differential weathering of the stone today makes it difficult to verify this theory, although the subdivision lines do appear fainter. Aked also considers that the large gnomon-hole may have been caused by replacing a horizontal with an inclined bar in order to achieve greater accuracy, especially in winter (ibid., 503–4).

Most horologists who have commented on this dial appear to believe that it is scientifically constructed in the tradition of Greek prime vertical dials (S. J. H. Daniel, pers. comm.) and, if it does conform to that tradition, then twelve divisions would be the norm. In the light of the evidence for linking Bewcastle pattern motifs with the monastery of Wearmouth/Jarrow (Cramp 1965b, 7–9) it is not without interest that Bede commented on the division of the day into twelve unequal 'hours' and also provided tables for telling the time of day by noting and measuring the length of shadows and stepping them out with one's own feet (Waugh 1973, 1–2). It is possible, then, that the dial is of a piece with the Classical and Mediterranean taste of the rest of the ornament. As with the inscriptions, it is impossible to tell whether its twelve divisions were meant to inform a monastic or a lay community. It may, however, throw light on the function of the cross here in that it could have been used in Easter calculation in the manner described by Bede :'. . . Aequinoctium autem . . . duodecimo kalendarum Aprilium die provenire consuevit, ut etiam ipsi horologia inspectione probamus' (Bede 1969, v, 21). [3]

The plant forms which fill some of the larger panels on the cross have long been recognized as deriving from oriental-Hellenistic models and have been most fully discussed by Brøndsted (1924, 35–6, 88–9) and Kitzinger (1936). The individual details of the scrolls are so distinctive that it is important to consider them first before discussing their origins. On face B the uppermost scroll does not terminate in a plant knot or volute as on Dii or other Northumbrian plant trails, but the carver fills a large triangular space with a sheathed berry bunch of the type repeated in a smaller paired form on panel v (see Fig. 5Ai). This large composite balances the sundial below which has a plant-like appearance despite its precise outline and layout. The leaf forms, the bell-like sheathed flowers or berries, and the flowering leaves are all repeated in the two scroll panels on this face and so provide a visual relationship despite the different organizations of the scroll. All spaces are filled; even the space between the roots in Bv is filled with a small stiff plant. The scrolls on face D are also based on an elaborate single stemmed scroll, Dii, and a bush and medallion scroll, Dvii. On this face three new leaf types are introduced and a composite of leaves and fan shaped stamens (see Fig. 5Aiii, Biii, Ei). The plant-knot terminal at the top of Dii occurs elsewhere in Anglo-Saxon sculpture, for example, at Easby, Yorkshire (Collingwood 1927a, fig. 53), but the large leaves with delicate veins and sprouting florets are unique in form and composition. Medallion scrolls such as this, which do not cross but are clipped or pinched together, also occur on face A of the cross at Otley, Yorkshire (Cramp 1970, pls. 41, A; 42, 1; 43, 3) which, like Bewcastle with its plumed terminals and small rounded leaves, demonstrates elements which are to be found in late Classical art. The clipped medallions with plumed tips are characteristic of scrolls on late consular diptychs, as for example, the diptych of Sividius, AD 488, (Volbach 1961, pl. 257), or the diptych of Justin, AD 540, (Delbruek 1929, pl. 134). It is indeed possible that ivories as well as textiles served to provide the English artist with models. The medallion scroll is found on twenty-eight Northumbrian crosses, many of which can be associated with the Hexham school of carving (Collingwood 1927a, 27–38); but these heavy scrolls with their complex detail are most like the scrolls at Otley I (Cramp 1970, pls. 42–3). The latter, in its inhabited scroll, contains elements such as the flowery leaf and the composite leaf-flower, which occur also at Ruthwell, Dumfriesshire (Fig. 5Ai–v; Ills. 682, 685–7). Comparisons with other exotic plant forms are discussed in the Introduction (p. 20), but one must note that Bewcastle is pre-eminent in the variety of such forms.

The inhabited scroll at Bewcastle is of the type with a strongly marked undulating stem unconcealed by subsidiary plant trails. This also occurs at Ruthwell (Ills. 682, 685–7), Edlingham, Northumberland (Cramp 1984, pl. 164, 868), Otley, Yorkshire (idem 1970, pl. 42, 2), and Auckland St Andrew, co. Durham (idem 1984, pl. 5, 14). Each volute, as in the paired scrolls of Jarrow, co. Durham (ibid., pl. 98, 526), and Jedburgh, Roxburghshire (Cramp 1983a, fig. 120a–b), encircles the animal or bird in a single coil, but assigns a distinct space to the plant elements and to the animal within the volute. In nearby Closeburn, Dumfriesshire, and in some later Northumbrian scrolls, such as Croft, Yorkshire, or York (Collingwood 1927a, figs. 59, 68, 146), the animals fill the volutes in an almost plant-like way so that their bodies are strained and contorted into writhing shapes (Henderson 1983, 246–9, fig. 105). At Bewcastle the beasts and birds peck or nibble composedly at the fruit or seeds. Their tails droop over or coil naturalistically round the volutes of the scroll, but they are disposed very formally nevertheless. The volutes pass either in front of or behind each body in regular progression. Above the leaping canine at the base each type is paired left and right and they diminish in scale up to the tiny squirrel-like creatures at the top. Only the central bird is en face and even it turns its head in the expected direction. The plump, well modelled birds and the tiny, squirrel-like creatures occur in almost identical form on friezes and a cross from Jarrow (Cramp 1984, pls. 98, 526; 90, 478) and on the narrow sides of the cross at Ruthwell, (Ills. 682, 685–7). The biped also occurs at Ruthwell, Edlingham, (Cramp 1984, pl. 164, 868), and Jedburgh (ibid., pl. 265, 1429). Well modelled birds of the same thrush-like form also occur at Otley (Cramp 1970, 57, pls. 42, 2; 43, 4), but the present east face of Ruthwell (Ills. 685–6) is closest in scheme and detail to Bewcastle. At the bases of the narrow sides of Ruthwell is a bird or beast and a leaping quadruped, above two birds, and two bipeds with the same head types as at Bewcastle. The leaf types and the composite plant elements on this face are also closely similar to Bewcastle, although all the volutes of the plant pass unvaryingly in front of each creature. On the west face of Ruthwell the animals are shown in livelier attitudes and the lowest creature is something like a winged biped, while at the top of the shaft are two bipeds which do not appear at Bewcastle. One has a canine head and is savagely gnawing the plant; the other is a sprawling gripping beast. The canine type appears also in Jarrow and Rothbury, and both the winged biped and the canine quadruped can be found at Jedburgh. In Yorkshire at Otley and Easby (Cramp 1970, fig. 52; Wilson 1984, ill. 80) a different, more eagle-like bird is contained in the scroll, and a crouching quadruped, such as is found in a later form in the west at Heversham 1 and Lowther 2 (Ills. 351, 432).

Although one would not wish to follow Brøndsted's single linear progression for the inhabited scroll, the Bewcastle cross must belong to the group already defined and whether one thinks that the heavy emphasis on the central stem is a late characteristic, as Brøndsted does (Brøndsted 1924, 38–42), is a matter of opinion. I have stated the sequence that I see for these scrolls elsewhere (Cramp 1978, 2–8), but would modify that somewhat in seeing that the fashion for inhabited scrolls could have been differently explored at the same time in various centres and the York school could have been unrelated stylistically, but nevertheless contemporary with, the Bernician school of Jarrow. All the same, there are certain broad trends which seem to affect the whole of Northumbria, not only in that in later scrolls the foliage elements diminish in importance and are less varied, but also in that the creatures spread themselves through the volutes and are not contained by them. Moreover, the creatures can then take on the space-filling characteristics of berry bunches or flowers, and their own bodies become patterned and contoured like the plants that surround them. It has been remarked that prancing creatures or canine quadrupeds with plant-scroll backgrounds are found on Anglo-Saxon sceattas by the mid eighth century and this could be of some chronological value in considering the dating of the crosses. The sceatta motifs are not very like anything on Bewcastle unfortunately (Brown 1915, pls. VII–VIII), but like the man and falcon, discussed below, the coins could reflect in a general way the popularity of inhabited scroll motifs (see Morehart 1984).

The west face of Bewcastle which contains the three human figures and the lengthy runic inscription has provoked more discussion than those faces which are devoid of figure carving. Before discussing the stylistic affinities of the figural sculptures of the west face, however, some decision is necessary on the varied interpretations of their identity. The figure in panel (ii) which, even when first discussed, was very worn, was at first identified and visually represented as the '. . . Blessed Virgin with the Babe in her arms . . .' in Nicolson's letter of 1685 (Nicolson 1685, 4). This identification was followed by Smith (1742, pl., 318; idem 1914, 13), and by the anonymous artist of Ill. 117. Armstrong (1914, 17–18) saw it as a bishop with a mitre. Daniel and Samuel Lysons said that only a general outline of the figure could be distinguished, but did identify the 'holy lamb' (Lysons and Lysons 1816, cxcviii–cxiix). Haigh, in 1857, pointed out that the figure holding the lamb was bearded and identified it, and the uppermost figure on the north face of Ruthwell (Ill. 684), with St John the Baptist (Haigh 1857, 151–2). Since then, until the recent article by Paul Meyvaert, this identification has never been challenged. Meyvaert specifically discusses the identification of the figure on the Ruthwell panel which he would interpret, following Dr Duncan (as reported in Maughan 1857, 14), as God the Father with the Apocalyptic Lamb (Meyvaert 1982, 10–22). To a great extent this identification is dependent on the interpretation of the [.]DORAMUS inscription on Ruthwell, as well as the lack of closely similar depictions of John the Baptist with the Lamb of God elsewhere at this period. [4]

The Latin inscriptions at Ruthwell have been convincingly explained by Meyvaert as conforming to a common layout formula, which would mean that the [A]DORAMUS would occur at the end of the inscription, providing some sort of sequence such as 'adoramus in aeternum, amen' (Meyvaert 1982, 10–14). This common sequence in the epigraphic layout is the most convincing rebuttal of Howlett's otherwise attractive reconstruction of the inscription in a manner that could include the words read by Cardonnel and Gough (Howlett 1974a, 333; Gough 1789, pl. 57): DORAMUS | TNONEUM. Howlett reads:

(left): [AGNUM DEI A]DORAMUS
bottom): [E]T NON EUM
(right): [SINGILLATIM TOTAM VERO]

Now there is no doubt that the Apocalyptic Lamb surrounded by symbols of the Apostles is well known in Anglo-Saxon art (Cramp 1978), and the complex linking of Old and New Testament texts which Howlett very convincingly demonstrates for the panels on the Ruthwell cross would be perfectly acceptable to a world which was accustomed to the type and prototype icons which surrounded the church at Monkwearmouth; moreover the Apocalyptic themes were also illustrated in the churches in Bede's monasteries (Bede 1896, 6). Should we, however, see the two crosses as depicting the same scene? There are differences: on the Bewcastle panel, the major figure is not haloed, whereas Christ below on the same face is; it is not standing on globes; and it is bearded, and points very clearly to the haloed Lamb.

It is also noteworthy that, in the parallels for a seated God the Father with the Lamb produced by Meyvaert, the Lamb is either in the bosom of, or on the right of, the Father, whereas in both the cross and in the depiction of John the Baptist on Maximian's throne at Ravenna, the Lamb is held in the left arm. It is fitting that Christ should sit on the right hand of God and that the Baptist should point with his right hand at the Lamb of God, and the position of the Lamb in relation to the attendant figure could be very important. The two 'globes' under the seated figure at Ruthwell (Ill. 684) have been discussed in detail by Meyvaert, who provides two convincing parallels which depict God the Father, the earliest being a tenth-century ivory Gospel-cover at Darmstadt (Meyvaert 1982, pl. VIII). These 'globes' on Ruthwell need not be considered as orbs however. The tradition in the Early Christian world for depicting 'the outside' as rocky ground shown in circular formations could have provided an indication of a desert milieu. (This does not, of course, explain why such a significant image was left out on Bewcastle.)

The text from St John (John i, 29), 'Ecce Agnus Dei, ecce qui tollit peccatum mundi' seems to provide the most obvious explanation of the scene. It could have been developed further in Ruthwell by the addition of the halo and 'globes' as well as a rather unsuccessful attempt to seat the standing model in order to provide an image from which more could be drawn in explanation. Thus the Agnus Dei first identified by John the Baptist, and the Lamb on the right hand of the Father on the day of Judgement could be explicitly linked. The figure of John the Baptist on Bewcastle, if we can accept this identification, is shown as draped completely in a voluminous garment which is rather different from those worn by either of the figures below. This may be another feature used to distinguish him. He does not wear a skin or fur as on Maximian's throne, but perhaps he was not seen as a wild desert ascetic. Saxl, in his percipient discussion of the relationship of the Ruthwell scenes, notes that St John is characterized not only as the Baptist but as '. . . the prophet of Him whose grace and truth were to take the place that the law had occupied in the Old Testament. The Jews had sacrificed the paschal lamb, and obtained salvation from Egyptian servitude; Christ is the paschal lamb, which taketh away the sin of the world. St John, the 'voice of one crying in the wilderness', is therefore shown on the cross above Christ in Majesty whom the beasts and dragons adored in the desert' (Saxl 1943, 4). The link between the Apocalyptic Lamb and Christ as Judge is not lost on the Bewcastle cross either, and the central figure of the latter is specifically named as Christ. The Christ figure likewise appears on Ruthwell. [5]

Despite minor differences of style between the two figures of Christ (Ills. 94, 683) – for example, the drapery arrangements differ; and Christ's halo on Bewcastle is apparently plain, while on Ruthwell it is cruciferous – a common model for this panel on the two crosses can hardly be denied. Christ's stance is the same on both, and the strange mole-like animals on which he treads are the same, although rather less of these appear on Bewcastle than on Ruthwell. Their submissive posture and calmly clasped paws separate them remarkably from the writhing creatures that occur elsewhere in Early Christian and early medieval art under the feet of the victorious Christ (Saxl 1943, 8–11). Their heads, instead of being pressed down into reluctant submission, are eagerly uplifted towards Christ. This happy relationship between God and the beasts is remarked upon by Bede and can be seen as a hall-mark of Celtic saints such as Kentigern, or of his Northumbrian successors such as St Cuthbert (Schapiro 1980, 174–5). Obviously this relationship was a hall-mark of many saints, not just those of Anglo-Saxon England, but Schapiro was probably right when he saw the emphasis on Christ as the reconciler and tamer of the beasts as especially appealing to the Germanic peoples – nurtured as they were on stories of animal and human conflict such as Beowulf (Schapiro 1980, 153). The animals under Christ's feet are, like some of the beasts in the inhabited scrolls on the crosses, representatives of the beast genre, not of a specific species. They are as anonymous as the strange beast head discovered at Monkwearmouth (Cramp 1984, pl. 124, 673–6). Recently Éamon Ó Carragáin has provided a striking literary explanation for the figure of Christ with the beasts in Bede's commentary on Habbakuk where it is said '. . . in the midst of two animals you will be recognized . . .'. He also makes the challenging suggestion that the cross might have served as the backing for a temporary altar (Ó Carragáin forthcoming).

It could be said that each scene on Bewcastle illustrates some relationship between a human or divine figure and the animal or bird world; and perhaps to an earlier viewer more accustomed to the transformation of man and beast, that remark would appear less superficial than it does today. Nevertheless, one might expect on Bewcastle, just as much as on Ruthwell, that the images could operate on different levels, or could be susceptible of multiple meanings. Schapiro (1980, 171) mentions another possible 'echo' between the figure of Christ as Judge represented in Paul's address to the Greeks ('He has appointed a day when he will judge the world in righteousness' (Acts 18, 31)) and the words that Bede puts into the mouth of Oswiu when he undertakes to convert Sigberht, ruler of the East Saxons (Bede 1969, iii, 22). At Bewcastle this could provide a link between the Christ figure and the figure below as well as the inscription (b), where royal names have been read.

The identity of the figure with the bird is still a matter of debate. When first described it was seen as a falconer with his eagle perched beside him. He appeared to hold a stick in his heavily gloved right hand, and there was a perch by his left arm below the bird. Unlike the other two scenes, no help in identification is provided by the Ruthwell cross. In support of the idea that here one has a secular figure, perhaps even a depiction of the deceased, one may note that his more 'active' and unorthodox posture could be an attempt to differentiate him from the solemn hieratic figures above. [6] Moreover, although the details of his drapery are difficult to determine, he seems to wear distinctively shorter garments and at the neck a large collar or ruff, rather like the secular figures on a shaft from St Mary Bishophill Senior, York (Collingwood 1927a, fig. 147). His long curling hair would be appropriate for a layman, but equally, some Evangelists are so shown (see the figure of St John in the Lindisfarne Gospels, fol. 209v.). Kingsley Porter felt that the bird was a dove and therefore that the figure could have represented Columcille (Porter 1931, 101). I do not find his analogy convincing, but certainly birds and human figures could have a variety of meanings. The bird could well be a distinguishing attribute of an aristocratic figure, as on Pictish slabs. In addition, there are strange figures of men and birds in Anglo-Saxon sceattas which may be symbols of rulers (Browne 1915, pl. VI; Metcalf 1977, fig. 13). Against this theory of a lay 'portrait' is the general point advanced by Saxl that it is very unlikely that '. . . on any sacred monument of the early Middle Ages the donor or the person to be commemorated would have been shown on the same scale as the holy figures.' (Saxl 1943, 7, n. 3).

More recently, Kitzinger has explored the fragmentary evidence that survives for icons of the seventh century and the devices of portraiture in Byzantine art. We do not know in relation to Anglo-Saxon images whether '. . . a high degree of geometric simplification and linearism is often the hall-mark of the contemporary portrait, while a modicum of Hellenistic pictorial techniques just like the ancient costume of tunica and pallium remain characteristic of saints' (Kitzinger 1976, 260). It is possible that some figures on Anglo-Saxon crosses were intended to convey to their viewers not just a generalized ecclesiastical figure, but a specific saint or bishop without assuming any attempt at the depiction of likeness (Cramp 1982, 18).

The figure carving at the base of the Bewcastle cross may have been a precocious essay in depicting a secular type, or a secular person, but it seems to have no development in surviving Anglo-Saxon art until the tenth century 'portraits' of such kings as Aethelstan or Edgar. On the whole, it seems more prudent to suppose that the figure is one of St John the Evangelist with his symbol the eagle, despite the lack of halo and the unorthodox posture. This interpretation of the figure also has the merit of unifying the iconography of the west face in that Christ was recognized as the Lamb of God by John the Baptist and the incident was recorded by John the Evangelist (Ó Carragáin forthcoming).

The question of the relationship between Bewcastle and other figural sculpture has been discussed in the Introduction (pp. 19–20). Specialists in Early Christian art

Camden 1607, 644; Worm 1643, 161, 168; Nicolson 1685, pl. 1c; Roscarrock 1691, 92; Camden 1695, 834–5; Cox 1720–31, i, 388–9; Camden 1722, 982, 1028; Smith 1742, 132, figs. on 318, 369, 529; Armstrong 1775; Nicolson and Burn 1777, ii, 478–9; Boswell 1786 ad loc.; Gough 1789, 176, 200, 322–3; Hutchinson 1794, 80–92, fig. following 80; Britton and Brayley 1802, 116; Howard 1803, 117–18, pl. XXXIV; Camden 1806, 455, pl. XXVI, figs. I–IV; Lysons and Lysons 1816, cxcix–cci, pl. following cxcviii; Grimm 1821, 165–9; Whitaker 1823, 229; Liljegren 1833, 186; Kemble 1840, 346–7, pl. XVI, fig. 14; Maughan 1854, 130–4, figs. on 130, 131–2; Haigh 1856–7, 517–19; Haigh 1857, pl. following 192; Maughan 1857, 10–38, figs. on 17 and 28; Charlton 1855–7; Haigh 1855–7; Whellan 1860, 8, 630, 633, 635–44; Haigh 1861, 37–41, pls. I–II; Stephens 1866–7, 398–404, fig. on 399; Stuart 1867, 16–18, pls. XXI–II; Way and du Noyer 1868, 210; Haddan and Stubbs 1873, 54; Simpson 1874, 9; Nanson 1876–7, 218–23; Nicolson 1877, 56; Ornsby 1878, 506–7, n.; Stephens 1878–83, 353–6, 364, 381; Haigh 1879, 133, 196, 198, 202, 209; Smith and Cheetham 1880, 1979; Browne 1880–4a, lxxiii; Browne 1880–4b, lxxviii; Browne 1880–4c, cxi; Black 1883–6, 124; Stephens 1884a, 189; Stephens 1884c, 128–30, figs. on 129; Allen 1885, 342–6; Allen and Browne 1885, 354; Sweet 1885, 124; Way and du Noyer 1868, 210; Allen 1887a, 85, 261; Sweet 1887, 86; Allen 1889b, 158–9, 207–9, 226–7; Browne 1893, 99–100; Calverley 1893b and pls. facing 243, 244; Ferguson 1893a, 495; Ferguson 1893b; Stephens 1894, 10; Viëtor 1895, 13–16, figs. 6–9; Lowson 1895–6, 60, 65, 67; Browne 1897, 37, 235; Calverley 1899a, 39–47, 54–5, pls. facing 39–43, 45, 54; Chadwick 1901, 84–5; Haverfield and Greenwell 1899, 45–7; Collingwood 1901a, 254–7, 277–8, pls. facing 255, 278; Curwen 1902, 242; Prior and Gardner 1912, 1–17; Allen 1903, xxix–xxxi, xlviii–xlix, 442–8, 515; Collingwood 1903c, 261; Collingwood 1903–4, 220–4, pl. facing 220; Stevens 1904, 46–54; Collingwood 1905b; Browne 1906, 188–213, 223, pls. 3–4; Taylor 1906, 349–52, pl. facing 348; Collingwood 1906–7a, 119–33, fig. 9; Paues 1907, 12; Browne 1908, 9, 57, 296–7, pl. XI; Dinwiddie 1909–10, 117–18; Wilson 1910; Bainbrigg 1911, 355; Dalton 1911, 103, 236; Brown 1912; Conway 1912; Cook 1912, 18–22, pl. 2; Prior and Gardner 1912, 111–12, figs. 92–3; Rivoira 1912, 24; Brown and Lethaby 1913, 43, 46; Hewison 1913, 352–9; Lethaby 1913, 156–8; Collingwood 1913c; Collingwood 1913–14, pls. facing; Conway 1913–14, 85, 89, pl. b on 87; Cook 1914a; Cook 1914b; Forbes and Dickins 1914; Hewison 1914, pls. X–XII, XV, XVII–XXIII, XXV–XXVII; Howarth 1914, 47–55; Browne 1915, 215–16; Collingwood 1915b, 126–30; Forbes and Dickins 1915; Browne 1915–16, 260; Conway 1915–16; Lethaby 1915–16, 204; Browne 1916, 171–94, pls. 4–6; Cook 1917; Howarth 1917, i, 336–48 pls. facing 340, 342, 344, 346, 348, III, pl. facing 310; Collingwood 1918, 41–3, figs. 16–17; Reeves 1920; Scott 1920, 55–6; Astley 1921, 238–43, pl. on 241; Brown 1921, 1–317, frontis, XI–XIII, XVII, XXIII, XXVI–IX, figs. 10, 18; Collingwood 1923c, 207; Hewison 1923; Strzygowski 1923, 231–2, 239, fig. 67; Brøndsted 1924, 33, 38, 40–3, 55, 68, 74–9, figs. 22–4; Collingwood 1925d, 70, 81; Hodges 1925, 161, 164; Hughes and Faulkner 1925, xxv–xxviii; Peers 1926, 48–50; Collingwood 1927a, 19, 67, 77, figs. 102–3, 135; Findlay 1927, 134, fig. 1; Green 1928, 495–6, fig. 3; Baum 1930, 61–4, 69, 167, fig. 65; Clapham 1930, 55–7, 61, 63, 66, 68, pl. 11; Langenfelt 1931–2, 182–3; Porter 1931, 69, 87, 99–101, 112, figs. 142, 160; Collingwood 1932a, 37, 46–8, fig. 7; Friesen 1933, 53–5, 59, 61; Rivoira 1933, 151, 209; Ross 1933, 144–5; Collingwood 1935, fig. on 25; Kitzinger 1936, 69–70, pls. IIIb, IVb; Mowbray 1936, 436, pl. facing 385; Pfeilstücker 1936, 116–18, pl. 12; Baum 1937, 36, 43, 106, pl. 56; Brown 1937, 100, 121–2, 132, 135–6, 147, 151, 169, 182, 191, 196, 198–200, 203, 205, 207–8, 210, 269, 271, 282; Gardner 1937, 22, 24, figs. 8–9; Arntz 1938, 147, 150, 208; Dahl 1938, 7–8, 43, 65, 179, 187, 192; Keller 1938, 28–31; Kendrick 1938, 128–34, 149, 152–4, pls. XLVII–XLVIII; Dauncey 1941, 115, pl. XVI, fig. 13; Morey 1942, 187; Sievers and Brunner 1942, 4; Åberg 1943, 50; Saxl 1943, 7–13, pl. 8; Arntz 1944, 123–4; Stenton 1947, 150–1; Saxl and Wittkower 1948, 15, 17–18, 20, pls. facing 17(1–5), facing 18(5); Bennett 1950–1, 270, 277, 281; Charlton 1951, 191–2; Gardner 1951, 21–2, 26, 28, pls. 23–4; (—) 1951–6, 305–6, pls. facing 305, 306; Hodgkin 1952, i, 300, 362–4, pl. 50; Derolez 1954, 50, 183; Dickins and Ross 1954, 12–13; Radford 1955, 45, 54; Stone 1955, 7, 10–11, 13–14, pl. 4a; Blair 1956, 159, 307–8, pl. VIII; Colgrave 1956, 123; Dickins 1956, 306–7; Kitzinger 1956, 294–8; Willett 1957a,, 95–100, 113–36, figs. 7–10; Willett 1957b, 19–22, pls. VII–X; Fyson 1957, 71, fig. 11; Radford 1958a; Elliott 1959, 80, 83, 87, 90, 95, 102, fig. XLI; Page 1959a, 287; Page 1959b, 399; Fisher 1959, 68, 82, pls. 26a–b; Bruce-Mitford 1960, 16, 255, pl. 52g; Page 1960, pl. 5; Cramp 1959–60, 11–13; Marquardt 1961, i, 17–26; Curle 1964; Mercer 1964, pl. XLI; Wilson 1964, 9, n. 7; Cramp 1965b, 7–9, pl. 5; Hill 1966, 134, 136; Pevsner 1967, 68, pls. 8–9; Verzone 1968, 244–5; Page 1969, 31, 36; Cramp 1970, 57–8, pl. 46, 5; Swanton 1970, 157; Aked 1973; Page 1973, 3, 5, 8, 28, 30, 34, 48, 134–5, 140, 147–8, 217, 219, 221, pl. 16, fig. 28; Adcock 1974, i, 158–65, figs. 28–9, 37ci, ii, pls. 2, 53–5, 56a, 57; Bailey 1974a, I, 20, 24, 26, 32, 34, 38, 41–2, ii, 47–48a, 49; Cramp 1977b, 265–6, 268, fig. 2(d); Farrell 1977; Alexander 1978, 12, 46, 50; Cramp 1978, 2–7; Ó Carragáin 1978 and figs. 7–10; Bailey 1980, 27, 46, 77, 83, 148, 242, 263–4; O'Sullivan 1980, 274, 279, 284–5, 290, 292, 306–9; Schapiro 1980, 153–60, 167, 172, 191 n. 59, 179, 190, 224 n. 80, fig. 3; Backhouse 1981, 75, pl. 57; Dodwell 1982, 41, 115, 309, n. 120, pl. 27; Meyvaert 1982, 9–10; Cramp 1983b, 277–9; Henderson 1983, 244–8; Tweddle 1983, 32; Calvert 1984, 543, n. 3; Cramp 1984, 5, 16–18, 27, 53, 107, 171, 174, 216, pl. 264, 1426; Wilson 1984, 70, 72, 75, 77, 84, 105, ills. 72–4; Ó Carragáin forthcoming

==R.J.C.

1. The sections on the inscriptions are by R.I. Page

2. This section is by R.I. Page

3. I am grateful to Éamon Ó Carragáin for pointing this out to me, and for suggesting the possibility that it could be related to calculating the feasts of the church. He also sees the three figural scenes as having a liturgical function.

4. For parallels, see the Munich Ascension ivory of the fifth century (Volbach 1961, pl 93). Meyvaert (1982, 14 and n. 47) notes that in 692 the council in Trullo took note of the fact that scenes in which the Baptist pointed to a Lamb as an image of Christ had become common, and prohibited the depiction of Christ in this manner. Not all the prohibitions of the eastern councils affected Western iconography, however, and the bearded patricarchal figure (in contrast to the beardless figure below) and the emphasized pointing gesture could have been sufficiant to convey the Baptist's identity to the Anglo-Saxon viewer. Ó Carragáin, in a detailed discussion of the Agnus Dei theme, notes that Pope Sergius rejected the canons of the Council of Trullo and suggests that the theological controversy about religious art may have inspired the iconographic programme of he Bewcastle cross (Ó Carragáin 1978, 134-6).

5. It is possible that on Bewcastle originally each of the figures on the cross was named by inscriptions above it. St John's name could have been obliterated, although there are some marks above his head. The name of the Christ survives above his head. The name of the lowest figure, if he were a layman, coul have occurred in the log runic inscription above his head.

6. His features have been almost obliterated and the strange appearance given by the deeply marked eyes is probably not original. They appear to have been recut. The eyes of the figures on both Ruthwell and Bewcastle are lightly outlines and lidded, with a very definite marked pupil. On the other hand, there is a remoter possibility that the eyes of this figure were always meant to be different and that we have here a blind figure with a stick. This could provide support for those who would like to see some link between the Life of St Kentigern and the Ruthwell and Bewcastle crosses, since one of Kentigern's miracles was to heal a blind king (Schapiro 1980, 174-5).