In the churchyard of St Mary and St Andrew's church at South Stoke (now itself more commonly called Stoke Rochford: NGR. SK 920273), mounted upright and set in an ancient (possibly original) undecorated base on a modern concrete plinth, c. 30m west of the west tower

The three pieces were said to have been originally dug out of the ruins of St Andrew's church at North Stoke, in the park of Stoke Rochford Hall (Davies 1911, 6; id. 1912–13, 148). They were reassembled 'near the lake ... under a canopy of four stately trees' at an unknown date but perhaps in the 1840s, when much of the landscaping was undertaken. The date of removal to the churchyard at South Stoke appears unrecorded, but it was post-1926 and pre-1964 (Davies 1926, 19; Pevsner and Harris 1964, 643). The move may have been made around 1948 when Stoke Rochford Hall was converted into a college. The church at North Stoke was in ruins before 1776 and its site was identified and excavated by D. Kaye (Whitwell and Wilson 1969, 115; Wilson and Hurst 1969, 278–9; Wilson and Moorhouse 1971, 144).

St Andrew

Very severely weathered

A large shaft originally decorated with interlace in low relief. The present single shaft, composed of three fragments of the original, was probably reassembled when it was re-erected as a garden ornament (perhaps in the 1840s). At that date also, presumably, the top of the shaft was trimmed and provided with its pyramidal shape, the original head having been lost in antiquity. The reassembly was achieved with a very tough mortar which has been matched closely to the colour of the stones, but unfortunately it was not quite exact. Although care has been taken to make sure that the three sections are in their correct relative orientation, the central section is mounted too close to the lower and slightly too far spaced from the upper, so that the interlace patterns do not join exactly. The surviving length of shaft should be 5–10 cm taller than it now is. The shaft has undecorated angle mouldings of rectangular section.

A (broad): Divided into three panels by undecorated transverse borders of rectangular section which are integral with the angle mouldings. The upper part of the shaft has a single panel across the width of the face decorated with an incoherent jumble of loops. A single strand is wound back on itself as it proceeds around the field, but the procession is erratic and has little symmetry. Half way up on the left side the sculptor achieves a single knot. The lower part of the face is divided into two vertical panels. That to the right has a moderately accurate four-strand plait, but that to the left seems to have a more complex pattern. The interlace in this panel terminates in a regular-looking knot, but in its upper parts it seems to degenerate into a single strand with loops of the same type as in the main panel above.

B (narrow): Although undivided by borders, this face has two independent runs of interlace. The upper part is occupied with a four-strand plait which terminates in unorthodox fashion with a pair of loops rather than the more conventional points. The lower part has an elaborate mixture of free rings and looped strands, this time carefully laid out and symmetrically arranged.

C (broad): This face may also have been divided into two panels by a frame, but the area where the frame would have occurred is now missing. The decoration is divided into two quite distinct groups. The upper part of the face has a poorly cut version of simple pattern E in which the knots are very small and occupy only the borders of the panel. The centre of the panel is consequently dominated by the criss-crossing of interconnecting strands between knots. There are two drilled holes placed carefully at the intersections of the strands, towards the bottom of both the upper and middle sections. These holes are filled with lead and are probably the pouring holes for the lead seating of two iron dowels holding the central section to the bottom and the upper section to the central one. The lower part of the face is occupied by a distinctive square interlace knot, which consists essentially of a frame formed by a single strand with looped corners enclosing a simple interlaced cross.

D (narrow): Decoration only survives on the lower and upper stones of this face. The central stone preserves a hint of the angle moulding, and a secondary drilled hole set asymmetrically. The decoration originally consisted of a run of interlace. At the base this included a free ring enclosing two crossed strands. The pattern must have changed in the central section, however, as on the upper fragment the interlace is a run of three-strand plait.

The base is stepped, with a chamfered plinth, but its angles are heavily weathered, particularly below face A.

This shaft has most of the characteristics ascribed here to the South Kesteven group (Chapter V). It is thin relative to its width, it is made of the characteristic shelly ragstone and it has undecorated angles of rectangular section. It also has some decorative motifs recognisable from other examples (the precisely cut runs of three- and four-strand plait along the narrow faces for instance). The simple pattern E on face C is also found on shafts belonging to this group at Colsterworth 1 (Ill. 94) and Creeton 1 (Ill. 126), but the style and quality of the interlace is quite different on these two shafts. The sculptor(s) at Stoke Rochford regularly used loops rather than proper interlace. It might be thought that this was because they could not manage proper interlace, but this does not seem to be the case, as the interlace on faces B, C and D is perfectly competent. Furthermore, the disposition of the loops on face B is itself carefully laid out. The upper panel of face A, however, is neither well laid out nor well executed. Perhaps there was more than one hand at work here, but if so, both hands were probably responsible for both interlace and loop-work. Similar eccentric loop-work is rare in Lincolnshire. There is a run of it on a cover from Lincoln St Mark (no. 5, Ill. 242), and two mid-Kesteven covers also have a well organised loop pattern (cf. Cramp 1991, fig. 26, Div) – those at Colsterworth (no. 3, Ill. 89) and Coleby Hall (Ill. 91) – and it may be that the decoration on covers such as these provided the inspiration for the Stoke Rochford work. The rectangular cross-knot on face C, however, is only paralleled in the county on the tympana at Rowston (Ill. 493) and Haltham-on-Bain (Ill. 494), which probably date from the final third of the eleventh century. [Profesor Cramp has suggested to us that this panel may have been recut, but we can see no evidence for this.]

The Stoke Rochford shaft, then, is clearly a member of the South Kesteven group, but much of its decoration appears unconventional. These features may be best explained by placing this shaft right at the end of the South Kesteven tradition, not just because some of the workmanship is so poor but also because several motifs used seem to be debased versions of designs found on other shafts (the loops seem to be a debased form of a design found on the mid-Kesteven covers, whilst the simple pattern E is a debased version of a pattern found on earlier shafts in the South Kesteven group such as that at Colsterworth). A mid eleventh-century date for the Stoke Rochford shaft can consequently be proposed, and given that the cross-knot on face C occurs on the tympana at Haltham and Rowston, which are probably post-Conquest, it may be that the Stoke Rochford shaft is as well.

The church site at North Stoke had its origin in a parochial chapel in the parish of Skillington. This chapel served a new settlement in the assarts of Stoke which seems to have been planted around the time of Domesday, i.e. at precisely the date of the cross-shaft (Owen, D. 1975, 20; id. 1979, 40). It seems likely, then, that the shaft itself marked the newly founded site, being placed, perhaps, to mark the new churchyard.

Mid eleventh century

Davies 1901, 4; Davies 1911, 6; Davies 1912–13, 130, 148; Davies 1915, 53; Davies 1926, 19; Pevsner and Harris 1964, 643; Pevsner et al. 1989, 720; Stocker with Everson 1990, 89