This well-known building was begun c.1100 as the main church of a Benedictine abbey. It is a very large church in the centre of the town, comprising a C12^th aisled nave of eight bays, crossing, transepts and tower, with a later medieval aisled chancel.

The E end of the church was apsed, as found by excavation in 1890-91 (Harrison and Thurlby 1995, 51), but the ends of the chancel aisles are of unknown form. The central tower collapsed onto the S transept in 1690, the N and W arches remaining in place, together with the NW angle of the tower containing the stair vice. It is said the Norman S transept is lost, but there are fragmentary remains to be seen in its interior walls. For example, the restorers put a window in the E wall: this is in place of the opening to a chapel for which the N jamb remains; there would have been a chapel on each transept, Pevsner 1967, 439 says they 'seem to have been square on the outside and apsed inside'.

The N and S transept walls retain evidence of the original triforium and clerestory, and the fenestration can be reconstructed (Harrison and Thurlby 1995, 51-54; Figs. 1-4). This plain 'Early Romanesque' work is the earliest remaining phase (Fernie phase I); 'Later Romanesque' introduces chevron mouldings as in three of the first two bays of the nave and the arches of the crossing (Fernie phase II). Both phases date from the time of Abbot Hugh, 1097-1123. Later work in the nave arcades and N clerestory is Transitional (Fernie phase III), with keel mouldings and shafts (Fernie 1995, Fig. 2; 40-44).

The cruciform church became the parish church without, apparently, any demolition or damage at the Dissolution. In modern times, there was a restoration in 1871-3 and 1889-90, but a fire in 1906 severely damaged the church. The central tower and S transept were restored after this fire (in 1908 and c.1912 respectively). The lead from the roof melted, and some of the interior stonework is still marked by it, for example, SW of the crossing. The upper stages of the west towers were built in 1935. The most detailed plan of the church, in Hodges 1892, pre-dates these major restorations but it shows well the phases of work relevant to the Corpus at ground level; see also Fernie 1995, fig. 2. The post-fire works included the reconstruction of the S transept and W front and the building of a vestry off the W end of the S nave aisle. For restorations and sources, see Harrison and Thurlby 1995, 50-51.

Twelfth-century remains range in date from 1100-1110 for the N transept to the 1170s for the N porch (Pevsner 1967). Surviving parts include the N transept, much of the crossing and two stages of the central tower, the aisled nave with most of the galleries or triforia, also the lower parts of the west façade, and the N porch, which is on the nave N aisle.

In 1069, a site for a Benedictine monastery in Selby was granted by William the Conqueror to the monk Benoit of Auxerre (Dugdale). The first site may have been too near the river. Building in stone is thought to have only started under the second abbot, Hugh (1097-1123), often called de Lacy. It is not certain that he was a Lacy (see Wightman 1966 and Dobson 1969). However, there are a number of grants from the family recorded in Dugdale (Fernie 1995, 48, n. 2, n.5). Dates are suggested for the nave fabric by Pevsner 1967, 441.

The site the monks used for their first attempts at a church has not been identified, but is likely to have been nearer the river, perhaps where Hodges marks 'site of parochial chapel'; abbot Hugh is thought to have removed inland a little to the present site. There are no other abbey buildings left apart from the church (Hodges 1893, plan of Selby as above). The remains of a tithe barn in a garage in James Street are now demolished. Ground floor masonry at Abbot’s Staith, The Quay, north of the bridge, is likely to have been later, and has not been looked for. For a drawing of the medieval gatehouse (which was roughly where the open space outside the W gate is now), see Dobson 1969, 23. For the W front and an adjacent timber building, see watercolour by John Buckler, 1816, BL Add. MS 37121 (on parish web-site).

The earlier phases of the building used a homogenous Magnesian limestone, but later work, especially around the lower parts of the W front and the N porch, seem to have deliberately made use a limestone with crystalline inclusions and veins: these tend to weather proud of the main block.

The restoration or reconstruction of the exterior of the tower and the E and S crossing arches are said to be 'reasonably close copies of the surviving masonry' (Harrison and Thurlby 1995, 55). The upper stages of the early tower are of unknown form, but it seems likely that two levels of round-headed windows were the limit in this period. An engraving of c.1655 (Cobb 1980, fig. 106, from Dugdale) shows the church from the NW, but with no good detail; another of unknown date showing the 'old tower of Selby church which fell down in 1690' - the church from the SW - is a glass slide (digitised) in the William Rawling photograph collection of the Yorkshire Archaeological Society, and published in Young 1995, p. 88.

The earliest picture of the abbey church is from 1655, by Daniel King (Cobb 1980, fig. 106; from the original in Dugdale); this shows the N and W faces of the tower with two further levels of openings, inset above the surviving Norman levels. The third level recalls the W towers of Durham cathedral, in that it appears to have narrow blank arcading, in which tiny windows have been cut; the uppermost level has three large windows, the central one of which might be an addition. This picture does not agree with the undated image from the William Rawling Collection which shows only one higher level on the S and W faces of the tower; they have three pointed windows, the central one larger than the other two; the top of the tower is finished with a parapet having a wavy design and pinnacles similar to those on the E end of the church. The title of the engraving is retained, in part, on the glass slide: 'Old Tower of Selby Church which fell down in 1690'. It actually shows the whole church from the SW; the glass slide (no. 634 as digitised) in the William Rawling photograph collection of the Yorkshire Archaeological Society was published in Young 1995, p. 88.

The E end of the church is thought to have projected only just over 15m beyond the transepts (Fernie 1995, 47; Harrison and Thurlby 1995, 51). It was apsed, flanked by two aisles with uncertain terminations, and these in turn flanked by the apsed transept chapels (see Hodge 1892, plan).

For similiarities with Durham and other sites, see Fernie 1995, 46-7.

The breaks in coursing which accompany the change between stylistic phases II and III (as defined by Fernie) can be best interpreted as intentional breaks, made once the part of the building required for the clergy was completed. Thus a temporary wall between piers 2 N and S of the nave arcades, and a roof over bays 1 and 2 resting on the floor of the unbuilt clerestory passage, would have sealed off the W end of their church (Fernie 1995, 42, 44-5), while the E end was presumably complete. It is likely that an internal screen would later have been in this position.

The abbey buildings must have been taken apart after the Reformation - from the 17^thcentury a lot of building took place in this region which would have needed quality stone.

Photographs, probably by George Zarnecki, in the Conway Library include capitals etc. in the gallery level. Negatives and prints by Kit Galbraith (Society of Antiquaries?) include the loose fragments (box XI)

Work of the time of Abbot Hugh

Hodges 1893, 349, describes two sculptural features as indicative of the work under Abbot Hugh, and which can be linked to the work under Bishop Flambard at Durham; they are said to mark the extent of Abbot Hugh's building. The first feature is the distinctive plinth construction with the prominent string course. This is used around the tower piers, inside the transepts, along the S aisle wall almost to the W wall, and in the N aisle for about two and a half bays. It is rather more refined than the work at Durham. The second feature is the decoration of the bases of the earliest parts of the nave arcades with simple patterns - cable, diamonds, etc.

The exterior billet string-course. Pevsner compares this to Durham, c.1100-1110 where it is used on the S wall of the nave and is visible from the cloister. The billet string course arching over a window can be seen at Brayton, a few miles SW of Selby.

W doorway to nave

The particular Magnesian limestone used in architecturally plainer parts of the Romanesque W front near the doorway seems to have been chosen for its visual interest, note, for example, the corner of the plinth/base of the NW tower, and the column on the left of the doorway, which have intrusive crystalline veins (outcrops of this kind seen at Bramham). Other late 12th-century buildings, for example, the W doorway at Cawood, use a similar type of stone.

W doorway, N doorway, also N Arcade, bay 4 first order

There is much similarity in the patterns used in these arches with the fragments which were found at Kirk Ella (YE), see Wareham 1960. All the patterns there occur at Selby: the most similar capitals are on the W respond of Selby's nave arcades (though many of these are reconstructions). Other comparisons are possible with the three arches of the entrance to the chapter-house of St Mary's Abbey, York, preserved and restored within the Yorkshire Museum. In that facade the first order has a net of chevrons round a roll moulding, the second order includes a lozenge pattern on a roll which has a twelve-rayed motif with central boss and fillings of foliage in the outer spandrels. The third order has syncopated chevrons. The effect at the chapter-house is more florid than at Selby, and the foliate capitals have no comparisons at Selby.

N porch

Pevsner 1967, 437, dates the porch to ‘a little after the ground floor of the W front’ since it has some Early English features. There are examples of stone in the older surviving columns on the exterior surface which seem to have been chosen for their curious qualities: they have crystalline inclusions, most larger ones of which are hollow. Similar stone is used on columns of the W doorway. There is a second stage to the porch. This is in a later style, but it may be useful to record for the Corpus what was told me by an assistant verger in 1998 - that, internally, there is access to this space by steps and a doorway off the N gallery. The space has no natural light or (apparently) ventilation. It contains a stone slab, and loose or broken columns.

Figure sculpture on W façade

‘Antiquarian sources’ suggest the possibility that the W front had figures inserted in arcading as at York Minster, see Oosterwijck & Norton 1990, 26, note 9. The figures found at Cawood church, 5 miles away, did not originate there.

E and W crossing arches

The corbelling-out of these arches recurs at Romsey Abbey.

The two outer corbels on the E arch, suggested to represent rolls of parchment bound together in a strap, is a form which occurs in a simpler form in some exterior corbels elsewhere (I think there is one at Birkin). A similar form occurs in the great hypostyle hall of the Mosque at Cordoba (9th or 10th century). The painted examples at Selby would be useful in defining what these carvings represent in a 12th-century Christian context.

The two men on the central corbel could be seen as a priest and a layman. Again with exterior corbel-tables in mind, they could be looking for the Second Coming, hence the firm, steadfast expression of the priest, and the open-mouthed amazement of the layman.

On the W arch there are comparisons for two of the corbels: the two embracing people on the S side might be compared to a corbel reset inside at Campsall, and to another in Bayeux cathedral in Normandy; at Drax there is at least one corbel with a small animal held in the mouth just as at Selby, this is reset in the porch.

S crossing arch

Hodges 1893, 351, says stone was reused in the reconstructed parts of the SE pier.

N crossing arch

The pattern of a row of lozenges on the ring of one capital recalls several examples on the tower arch at Etton (YE).

Transept Windows

Hodges (1893, 351) mentions that the interior E wall of the N transept contains remnants of the Norman clerestory arches; he proposed three arches in each bay, the centre one only pierced to form a window. However, recent analysis of remnants of string-courses and window arches in the walls of N and S transepts suggests a different fenestration (Harrison and Thurlby 1995, 51-54, figs. 1-4).

Windows in nave aisles

No Romanesque windows survive in the aisles; remains suggest there was originally a pair of windows in each bay; they 'may have had an oculus set between their heads like the transept chapels at Fountains Abbey' (Harrison and Thurlby 1995, 58).

Arcades

S arcade Bay 3, Pier 2, W face of main capital: the two lions and the two men. The good man embraces the lion: the bad man fights the lion. At the last Judgment, it was believed Christ would appear as good to the good and fearful to the bad, (Emile Male citing Honorius d'Autun; probably an earlier authority somewhere). A window on the W front at St Peter's, Besalu, Spain, is flanked by two lions, one with a naked man holding up a large fruit, and the other standing over a prostrate man who is perhaps entwined with a snake; this pair of carvings could have a similar lesson (de Champeaux and Sterckx 1989, pl. 101, 277-8).

S arcade Bay 5, capital of second order to nave, E side, on pier 4; also N arcade, Bay 4, capital of second order to nave, W side. Assuming that the plentiful lush foliage belongs to Paradise, and that the men are showing their enjoyment of being up there in it, the dark rectangular holes might represent empty graves. On vault supports in the N aisle pier 4, the capital of the transverse arch of Bay 4 and the capital of the SE rib of bay 5 have domes within circles, six-fold and many-petalled 'daisies' just above the ring; these small motifs are likely to represent stars rather than flowers; combined with the ring, which is sometimes cabled, the firmament is represented.

N arcade, waterleaf detail. The crockets on the W respond seem a world away from the Romanesque motifs on pier 4, but there is at least one connection. Near the ring and alternating with the crockets at the top, the recessed embayments between the leaves are a design feature that forms a foil to the projection of the crockets. This recess is a feature of some waterleaf capitals on pier 6, and even on pier 4, in the E capital of the second order to the nave.

Bay 5 of both arcades. The divergence in everything between the E side and the W side in this bayis stark. The piers are worlds apart in terms of architectural ambition, construction methods, models, symbolism, focus and human concern. This free-standing sculpture might be compared to work on arcade capitals at Conisbrough, but is not common in Yorkshire.

Vault, bays 1 and 2 of nave aisles:

Deformation of rib profiles as at other sites. These matters are discussed in Harrison and Thurlby 1995, 57 and 58.

Profile (Harrison and Thurlby 1995, 58). Regarding the profile of two rolls separated by an angle fillet, noted by the authors as occurring at Dunfermline and Durham, this is common in mid-century works of the 'Yorkshire School', being used in the first order jambs of doorways at Riccall and Healaugh, for example, as well as Brayton. It might, therefore, be diagnostic of a continuing strand in the fabric of the group - or alternatively so standard a feature as to merit no importance.

Additional material

There are at least two items of sculpture mentioned by Harrison and Thurlby that the fieldworker did not get to see - these are in the upper parts of the N transept accessed from the vice and in an upper room in the tower. 'A shaft and incised scallop capital' of the triforium on the W wall can be seen from the stair (1995, 52); in the lantern stage of the tower there is 'a simple leaf capital in the north arcade' (1995, 56).