The Romanesque church was a cruciform building with an aisled nave with triforium and clerestorey; N and S transepts and an aisled eastern arm with a gallery rather than a triforium. Of the nave, the four eastern bays and the beginning of a fifth survive. In the fifth bay was a 13thc. north doorway under a porch, and west of the sixth stood the façade. There is no evidence for the original form of this beyond the ruinous lower part of a NW tower. This tower collapsed partially in 1572 and more drastically in 1574, destroying the western bays of the nave, and was rebuilt on a magnificent scale. Until 1881 it was reportedly the glory of the exterior and a notable Chester landmark, but in that year, while long-overdue repairs were taking place, it collapsed again, destroying the Early English north porch, which was rebuilt by J. Douglas in 1881-82. The eastern arm of the church was originally aisled and of five straight bays, but now the entire north aisle has been removed (except for its eastern chapel; see below). Of the main vessel and south aisle only a single bay survives within the building, which terminates in a straight wall. The remainder of the eastern arm was abandoned in 1547, when the King's Commissioners decided that the nave alone was sufficient for the parish, and that the lead on the choir roof along with the metal of four of the church's five bells should be removed and sold. To the east, outside the building, parts of the S choir aisle wall still stand, along with what remains of the east chapels. Originally the main vessel terminated in a deep apsidal chapel, and the aisles in shallower ones. All three chapels were remodelled and enlarged in the later middle ages, but the 12thc. wall containing their entrance arches still stands. This is in a disastrously eroded condition, which should be borne in mind while reading the descriptions of its elements in this site report.

The central tower of the church collapsed in 1468, and again in 1572, and at some point, presumably after 1547, the transepts were removed. The only other medieval part of the church is the enigmatic two-storey structure of c.1300 built in the angle between the south transept and the choir and accessed through a doorway in the S choir aisle. Its undercroft is square and vaulted in four bays with a central pier: the upper storey has lost its roof. Locally it is known as the Chapter House, but neither its form nor its position make this very likely, and it is here suggested that it was a two-storey treasury. In the early 19thc. it was incorporated as a kitchen in a house (now demolished) which became the residence of Thomas de Quincey's mother. What remained of the treasury was renovated in 1937 and the undercroft taken over in 1939 as a public air-raid shelter. It now serves as a stone store.

There was apparently a thorough repair to the chancel in 1813, but the external appearance of the church today is of a 19thc. building in Early English style, and this is largely due to the restorations of J. C. Hussey who rebuilt the south side in 1859-60 and the north in 1886-87. Included in the latter restoration was the construction of a modest bell-tower in the angle of the north transept and the choir. No firm dates are available for the Romanesque fabric. The present church was traditionally begun by Bishop Peter de Lea, who moved the see to Chester from Lichfield in 1075, but judging from the sculpture, none of the fabric is this early.

Collegiate church (c.1057-75), cathedral (1075-1541), parish church from 1547.

St John's is an Anglo-Saxon foundation, traditionally first founded by King Aethelred of Mercia c.689, and rebuilt and enlarged by another Aethelred, the Earl of Mercia and husband of Aethelflaeda, daughter of Alfred the Great in the early 10c. In 1075 Peter de Leia, Bishop of Lichfield, moved the see to Chester and began to build a new church. At his death in 1082 he was succeeded by Robert de Limesey who promptly transferred the see to Coventry. It has usually been assumed that the earliest parts of the present building date from Bishop Peter's time, and that the time lapse between this campaign and the late-12thc. work in the upper levels of the nave reflect Bishop Robert's lack of interest in the building. Gem (2000), however, has observed that after Robert moved the see to Coventry he is recorded using part of its revenues to fund rebuilding work at Lichfield, and suggests that he saw the increased revenue available as an opportunity to enhance all three of his cathedrals. At some time before 1540 St John's had been demoted from cathedral to collegiate status, and in 1547 it became a parish church.

Parker (1855-62) took the view that the building was begun in the late 11thc., attributing the choir arcades, the transepts and the main arcades at the east and west end of the nave to a campaign of the late 11thc. and early 12thc. He considered that the central section of the nave arcade had been completed last, and that the choir aisles were either delayed in their completion or rebuilt later. He dated the upper storeys of the nave to c.1190. Pevsner and Hubbard's analysis is less detailed, but suggests a starting date before 1095, and dates the nave triforium to c.1190 or later and the clerestorey to the 13thc. A very different view was taken by Clapham (1934), where it is suggested that the church was not begun much before 1130-40. He was to modify this in 1937, dating the inception of work 'not before the first half of the 12thc.', and the nave arcade to the third quarter of that century. This historiographical analysis relies on Gem (2000), whose own suggestion involves three campaigns, the first including the entire original eastern arm (c.1100-17), the second the crossing arches and the nave arcade (c.1125-50), and the third the nave triforium and clerestorey (late 12thc. and early 13thc.). Gem suggested that an episcopal vacancy delayed the building of the crossing and nave arcades, but there seems no compelling reason to postulate two separate campaigns rather than a single, extended one. Where there is a marked break is between the nave arcades and the triforium and clerestorey above. The triforium includes waterleaf, stiff-leaf and moulded capitals, suggesting a date in the 1190s; while the clerestorey capitals are predominantly stiff-leaf with some moulded forms, pointing to a date in the first decade of the next century. The present author is happy to accept Gem's broad chronology.

Pevsner and Hubbard pointed out that the entry arch into the S choir aisle differs in design from the arches from the aisles into the transept in having heavy soffit rolls and angle rolls to the orders. This is true, but its northern counterpart appears to be entirely rebuilt, and may well originally have shared the extra elaboration. There is nothing in the capital sculpture to suggest a later date for this arch. Most of the loose stones find ready comparisons with sculpture in the fabric. The most tantalising exception is VI.i.6, here identified as an annulet. This identification is not secure - indeed the only even slightly similar feature known to the author is on a nook-shaft on the south chancel window of St Mary Magdalene, Cambridge, and this is not made in the same way.

Perhaps the finest capital sculpture of the earlier 12thc. in the county appears in the two ruined capitals and imposts surviving on the S chancel chapel arch. Parts of their carved surfaces survive in a very fragile state, but if they are left in-situ they will rapidly deteriorate to the condition of their counterparts on the other two chapel arches, which were clearly of similar quality when new. The loose capital with oxen (VI.i.1) appears to belong to this group of capitals. It would benefit from conservation and merits permanent display.