In the church

First mentioned in 1903 as having been found in the rockery of Hornby Castle (SD 587686) and then, after a period in the castle, being placed in the church in 1904 (Taylor, H. 1903, 99; id. 1906, 299).

St Margaret

All the upper faces are damaged and much of the border between faces A and D has been lost.

A (broad): The fragmentary remains of the head, which was slightly wider than the shaft, are divided from the shaft by two horizontal mouldings. In the head is the lower part of a complex knot pattern which seems to resolve into two interlocked pattern F loops with paired strands.
The shaft is filled with a delicately-carved scene set within lateral and bottom moulding borders. At the base are two fish with, above, five roundels, placed three above two; the two roundels are flanked by mouldings which presumably represent the sides of the basket in which they are contained. Springing from the central upper roundel is the stem of a tree whose upper termination is formed by a rosette set between two branching stems; these stems end in plain triple round leaves to the left and a pointed leaf to the right. Two short branches spring upwards from the middle of the trunk which terminate in similar triple leaf forms. To the left of the tree stands a haloed figure, clothed in a long garment and half turned to his left, with classically modelled hair reaching down the back of the neck and with delicate facial features; the brow and eyes are well marked, the head slightly bowed. One arm, with distinct fingers, is bent across the body. The standing figure to the right is also haloed but here the hair is not shown as long. The face is half turned to the left, with well-delineated brow and eyes. This figure appears to wear a long garment and has the right hand raised pointing at a book held in his left hand.

B (narrow): The end of the cross-arm protrudes some 2 cm beyond the plane of the shaft, the two parts of the cross being separated by a double horizontal moulding. The right-hand border of the cross-head survives as do most of the lateral and lower borders of the shaft. The head is occupied by complex knotwork, possibly the remains of four simple pattern E knots forming linked triquetras, and the shaft is filled by two registers of interlocked pattern F loops with paired strands, terminating below in a double-stranded simple pattern E knot.

C (broad): On this face the remains of the lower cross-arm are distinctly wider than the shaft. In the head is a haloed half figure with deeply-carved wings, holding a book across the chest; the book is held from below in the left hand whilst the right hand points towards it. The figure's right shoulder carries a double moulding, possibly indicative of an elaborate collar. This figure is flanked by panels of zigzag moulding, each resultant triangle filled with a small pellet.
The shaft has its own inner border but, to the left, the remains of an additional outer moulding survive to suggest that this face had a double surround. Within the single panel is complex knotwork formed by two parallel and inter-connected rows of complete pattern C interlace in four registers, the terminals at each end formed by alternately joined strands. At the bottom of the panel is a worn inscription.

Inscription The text is now practically illegible (Ill. 551). Okasha (1971, 82) thought to see perhaps +E in the centre of the panel. This is not consistent with Collingwood's reading [D]N.DIRI[GE], which he interpreted as 'possibly, but not by any means certainly, the opening of the Dirge or Antiphon in the Office for the Dead' (Collingwood 1929a, 57).[2]

D (narrow): The end of the surviving fragment of cross-arm protrudes some 2 cm from the plane of the shaft and its decoration is separated from the shaft by a double horizontal moulding. In the head, its borders cut away, are the remains of two linked simple pattern E knots, whilst the shaft is occupied by three registers of spiralled half pattern A interlace with bar terminals at each end. Only the lower and left hand border remains on this face of the shaft.

This cross is sculpturally of very high quality, its delicate small-scale figural sculpture ambitiously representing the half-turned forms of natural engagement, and its interlace giving a satisfying impression of complexity. Collingwood (1927a, fig. 71) restored the head as type D9 or D11. This is inevitably speculative but it is perhaps important to note that the large churchyard shaft at Bakewell and a fragmentary shaft from Ripon (Hawkes 2007b, fig. 24; Coatsworth 2008, ill. 632) have similar large blocks which project beyond the top of the shaft, acting as the lower element of the head.

Its inscription identifies it with a group of literate monuments centred on the Lune valley (see Chapter IV, p. 20). The zigzag element around the head, with its pellet infills, is exactly paralleled on a lost pre-Viking carving from Lastingham in Yorkshire (Lang 1991, ill. 622), but is an elaborate version of a metalwork-derived form popular in early Northumbrian monastic sites and found locally on Hornby 2, Heysham 3 and Lancaster Vicarage Field 4 (Ills. 513, 553–4, 619; see discussion under Heysham 3, p. 200).

The interlace patterns appear complex but are in fact relatively simple; they depend for their effect on the fine density of the lay-out and the clever use of double strands. The type on face C (Ill. 549), a paired version of complete pattern C, is unique within Northumbria and northern Mercia, though, as a single row, it is found on both pre- and Viking-age monuments at sites like Sandbach Market Square 1, Leeds, Ripon and Bakewell (Ill. 257; Coatsworth 2008, ills. 493, 672; Routh 1937, pl. V). The type on face D (Ill. 550) is spiralled half pattern A, which occurs early at Ripon and Cundall/Aldborough but also in Viking-age contexts at Stanwick and Aspatria (Coatsworth 2008, ill. 673; Lang 2001, ills. 161, 763; Bailey and Cramp 1988, ill. 34). The seemingly complex pattern F interlace on the head of face A and on face B (Ills. 548, 552) owes its complexity to the use of double strands. In its single form it is frequently employed at all periods.

Though many of these patterns have a wide distribution over time and space it is probably significant that complete pattern C and spiralled pattern A are both found on the fragments of an eighth- or ninth-century architectural frieze at Ripon (Coatsworth 2008, 240, ills. 672–3), whilst complete pattern F — which is closely related to the double-stranded version — also occurs there (Coatsworth 2008, ill. 674). For there are other links to the Ripon area which can be traced on this stone. Ripon has a cross-head using both zigzags and angular frames enclosing small pellets; as we have seen it also has a shaft with a block at the base of the head (Coatsworth 2008, ills. 632, 637, 639). What is more, Lang has identified a sculptor working in the Ripon area, the so-called 'Uredale master', who produced the Cundall/Aldborough shaft. This not only carries half pattern A but has delicately carved panels with figures engaging with each other in the manner of Hornby 1. Moreover it has central-stemmed bush vines, one of which is topped by a pelleted flower identical to that at Hornby 1, and has an angelic bust set at the top of one, if not two, of its sides (Lang 2001, 41–4, 93–7, 102, fig. 14, ills. 159–84).

The role of the boldly carved and winged bust, holding a book, is not entirely clear (Ill. 549). It could be the symbol of St Matthew in a set of four decorating the cross-head; this is the position occupied by the symbol on the early eleventh-century slab, Durham 12, and it is also where we find Matthew on the Ruthwell cross-head (Cramp 1984, 73–4, pl. 51.241; id. 1978b, 126). Alternatively it may function like the angelic figures, holding a Book (of Life?), at Halton St Wilfrid 3 and 5 (Ills. 481, 483, 489).

The scene on face A is clearly a depiction of the miracle of the Loaves and Fishes (Ills. 547, 552), a theme whose popularity reaches back to the earliest representations in Christian art in the catacombs and on early sarcophagi (in general see Schiller 1971b, 164–7; Jensen 2000, 84, figs. 12, 17; Harbison 1992, i, 255–7). In Insular sculpture representations of the scene appear nine times in Ireland, once in Scotland but only once elsewhere in England at Dewsbury (Harbison 1992, i, 255–7, iii, fig. 843; Coatsworth 2008, ill. 207). There is a good deal of variety in representing the miracle across both continental and Insular realisations. Some have accompanying figures of apostles or crowd, but types with only a single accompanying person, as here at Hornby, seem to be both rare and early: Schiller (1971b, figs. 464 and 477) illustrates sculptural examples from Rome and Gaul to which can be added an engraved and gilded glass-base of similar fourth-century date in the Vatican (Du Bourguet 1965, pl. 144). Admittedly a fresco panel at ninth-century Müstair comes close to this composition in depicting a richly dressed man alongside Christ but this still has two additional smaller figures below (Birchler 1954, 201, no. 47). Hornby's composition could thus depend on an early model, or the constraints of a narrow panel may have forced abbreviation of the accompanying crowd. The second Hornby figure to the right could be Andrew, Simon Peter's brother who, in the narrative of John VI, 1–13, identifies the lad with the basket but feels doubtful if his food-supply will suffice. If so, he is not given the unkempt hair so frequently associated with that apostle (Bailey 1996a, 58–9; Henderson, G. 2007). It is therefore more likely that the youthful-looking figure is that of the Evangelist John, pointing to the Gospel text which, as we will see, explains the implications of the event.

Even among the variety of forms adopted for showing this miracle in early Christian continental and Insular art, the Hornby panel is unique in its incorporation of a fruiting tree between the two figures. This visual riddle calls for explanation.

In his sermon for Mid-Lent Sunday, Ælfric, drawing on St Augustine, suggests an approach to a solution: 'with regard to the miracle which God performed with the five loaves, it is not enough that we wonder at the miracle or praise God because of it, without also understanding its spiritual sense (gastlice andgit)' (Clemoes 1997, 277). St Hilary provides something of the 'spiritual sense' which can be teased from the presence of this unexpected tree. He saw the miracle as a sign of God's fecund power to sustain his people, fecundity expressed through bread and, vitally here also, through wine: 'wonder not that the fountains run, that there are grapes in the vines, and that wine comes forth from the grapes ... for this so great yield of loaves proclaims the maker of all this' (Migne 1844, col. 1001). Hornby's central fruiting vine thus picks up the God-given generosity of the miracle's spiritual significance.

But rumination yields more. John VI, 35 ('I am the bread of life'), like 1 Corinthians X, 6 ('Is not the bread which we break a participation in the body of Christ'), links the miraculous loaves and fishes to the Last Supper and to its later Eucharistic re-enactment. Patristic commentaries reinforced these links: 'Who can the bread of heaven be but Christ?' asks St Augustine, 'But for man to eat the bread of angels, the Lord of angels became man. Because if he had not become this, we would not have his flesh; if we did not have his flesh, we would not eat the bread of the altar' (Migne 1865b, col. 726; see also Schiller 1971b, 164–5). That Eucharistic dimension is present in this Hornby panel in the way in which the central tree-vine takes on a cruciform shape, a visible emblem of the thoughts lying behind the famous poem by Venantius Fortunatus: 'you are powerful in your fruitfulness, O sweet and noble tree ... hanging between your arms is a vine, from which sweet wines flow as red as blood' (Walpole 1922, 179, 181).

What at first appears to be a simple depiction becomes, through visual manipulation and riddling puzzles, a powerful statement of the symbolism of the Eucharist, and of God's fecundity. Past event and present experience are linked together. The image is multi-valent, layer upon layer of meaning emerging from informed contemplation of the elements which disrupt its expected form.

Inscription Higgitt (1986b, 130) adduces parallels for inscriptions consisting 'of single horizontal lines ... placed on a strip above or below a panel with ornament or figures'. Little more can usefully be said about the text.

Late eighth or early ninth century

Taylor, H. 1903, 99–103, pl. facing 99; Collingwood 1904a, 332; Collingwood 1904b, 36–9, figs. 1–3; Garstang 1906, 267, pl. facing 266; Taylor, H. 1906, 391–5, pl. facing 391; Collingwood 1907a, 281, 292, 337, 351; Ditchfield 1909, 119; Collingwood 1913, 169; Farrer and Brownbill 1914, 200; Collingwood 1915, 164, 218, 226, 233, 270, 292; Collingwood 1916–18, 40, 46, fig. 14; Collingwood 1927a, 57–8, 73, 119, fig. 71; Collingwood 1929, 27; Collingwood 1932, 48; Brown, G. 1937, 187–8, 197, pl. LV; Cramp 1965, 9; Pevsner 1969b, 16, 146–7; Bu'lock 1970, 292; Okasha 1971, 81–2, pl. 56; Adcock 1974, I, 148–50, II, pls. 49a, 50a–d; Edwards, B. 1978a, 64; Fellows-Jensen 1985, 402, 403, 404; Higgitt 1986b, 130, 147; Edwards, B. 1988a, 205; Lang 1990a, 14, 15; Kenyon 1991, 102; Harbison 1992, I, 256–7, 323, 325, 343, III, fig. 838; Okasha 1992a, 81; Higgitt 1995, 230; Bailey 1996a, 3, 59–61, pl. 2; Bailey 1996b, 26–8, 30, 38, pl. 2; Newman, R. M. 1996, 98; Rollason et al. 1996, 12; Crosby 1998, 30; Noble 1999, 35, fig. 48; Lang 2001, 52, 252, 268; Bailey 2003, 219–21, 232, fig. 9; Hawkes 2003d, 359–61, 367, pl. 22.5; Okasha 2004, 274; Blair 2005, 216; Salter 2005, 44; Newman, R. M. 2006, 103; Coatsworth 2008, 61, 137, 138, 187; Cramp 2008, 19; Bailey 2009, 24–5, fig. 6

[1] The following are unpublished manuscript references to the Hornby stones: BL Add. MS 37550, items 650–62, 688, 692, 696; Manchester Public Library, Hibbert Ware S. MSS: Msf 091 H21, vol. 4, 158.

[2] Originally he read, and drew, –N.AIRI–, 'the meaning of which I cannot suggest' (Collingwood 1904a, 39).