The incomparable apse of the second cathedral

Summary. A large, basilica-shaped foundation to the second cathedral was found under the choir area, 1854. It is unlike any east end of known English cathedrals, though it has some resemblance to early Continental churches.

     In 1854, a foundation for the second cathedral was found under the choir and presbytery floor. Its size and shape were drawn by John Hamlet, who had been commissioned to build brick ducts under the floor to carry hot air. The foundation was basilical in form with an east end in the shape of a semi-circular apse. By every measurement it was extraordinary.[1]

 

Hamlet’s 1854 drawing of the foundation together with the published drawing by Willis, 1861. [2]

 The large apse was hemispherical without any side chapels or apparent buttresses which made it unusual. It was also as wide as the lateral walls of the choir-presbytery. The entire shape resembled early Roman basilical buildings.

A comparison of apse sizes of early churches.

 

A comparison of apse sizes of early churches.

 

Large size and basilical shape do not intrinsically indicate a specific time of construction, unless it was an Early Medieval Romanesque church. Willis published the discovery without giving a precise date for the foundation. It was believed Anglo-Saxons[3] did not build large in stone, so he defaulted to thinking it was Norman. This lax thinking has persisted and surprisingly is still claimed the second cathedral was Norman, although there is no supporting written evidence or any other Norman stonework existing. Several reasons for it being Early Medieval and not Norman are given in the post, ‘The second cathedral has to be Early Medieval.’

A review of other Norman and Anglo-Norman cathedrals that have had at some time a round apse emphasises the incomparable shape of this foundation. The following diagrams are not to scale and dates are approximations, but show large cathedrals, abbeys and churches which have had a distinct, round apse.

 

Conjectured east end of Canterbury 1030. The apse has a north and south porticus. It was only 9.2 m across at the chord. The whole of the east end was changed in 1095 by extension with towers and a chapel.

 

Ely 1083, but whether it had an apse is uncertain, it was soon squared off.

    

 Peterborough 1118.The hemispherical apse was built on polygonal columns, later retained and enclosed in a squared end, 1500.

 

 Norwich 1090 with site of rectangular Lady Chapel added on.            

 

 Old Sarum, Salisbury, late 12th-century.

 

Westminster Abbey 1245–69. A Lady Chapel was added to give 5 apsidal chapels.

 

 Bury St Edmunds Abbey 12th-century.                                            

 

Gloucester 1089.       

 

 Wells 1175, squared in the 14th-century with a polygonal chapel.  

 

Norman cathedrals only having square-ended chancels are St Albans, Chichester, Hereford[4], Lincoln (?), Old St Pauls, Rochester, Southwell, Winchester (?) and York. Durham had an apse for Cuthbert’s shrine, but then the east end was squared off; the line of the apse can be seen in the pavement. Exeter had a five-sided apse c. 1133, but by 1292–1308, the current squared east end was built containing several side chapels.  

                        

It is clear Canterbury, late-10th or early 11th-century, comes closest to being like Lichfield, if its floor plan is true, but the overview is Lichfield apse has no obvious equivalent.

           When Rodwell investigated the foundation in the south and then the north choir aisle[5] he made a comparison with Norwich Cathedral and conjectured the existence of apsidioles off the apse, but then failed to find any. Instead, he suggested the absence of chapels or apsidioles off an eastern apse was comparable to early and small Jumiège Abbey, begun 1037 and consecrated 1067. However, the simple round apse with an ambulatory at Jumièges was a conjectured layout after a 1935 excavation and is part of a church floor plan which has yet to be accepted. The dimensions of its putative apse, c. 70 x 70 feet, was close to the Lichfield apse. Lichfield apse could also have a resemblance to the slightly earlier apse at Bernay Abbey,

 

East end of St Pierre, Jumiège Abbey, showing reconstructed floor plan for 1037–67. Soon after, c. 1040, it had a series of small apsidioles.

 

Bernay Abbey 1010–55, the oldest surviving Norman Romanesque church with a simple round apse. It might be how the east ends were truly built in early Norman times.

 

 

A search of all Romanesque churches, 800–1200, as described by Krautheimer[6] and Conant[7] has failed to find a close affinity with the apse at Lichfield with the early build of Jumièges Abbey coming closest. There is a resemblance to St Denis Abbey, located in a suburb of northern Paris[8]. Abbot Fulrad built a basilica church, dedicated c.775, with many features modelled on St Peters in Rome. Partial excavation in 1938 by Crosby[9] revealed a wooden roofed columnar basilica with a spacious adjunct extending a little beyond the aisle walls, a lantern tower, a new kind of west end, and a simple, short, apse extending from the east end.   

 

Reconstructed Basilica of Saint-Denis, the earliest Carolingian Romanesque church. Nave and apse were 30 feet wide. The alignment of the crypt is unexplained and shape conjectured.

 Conclusions

•      A simple, very wide, hemispherical apse is extraordinary and there is no equal with any of the British Norman churches.
•      Comparable churches are either very early Norman (early conjectured Canterbury, Jumiège and Bernay) or early Romanesque churches such as the abbey of St Denis.
•      Dating this foundation has been extremely contentious. A current cathedral pamphlet states it was built around 1085 and is Norman. This has been given without any evidence. Other later dates have also been given without any supporting evidence. Why there is unsubstantiated dating in the history of the cathedral is inexplicable.
•      The dimensions of the foundation have an Anglo-Saxon metric of a short perch (15 feet) and not a Norman rule of a long perch (18 feet). See footnote 1.
•       The mortar of the foundation needs to be carbon-dated. It would be a relatively simple and cheap project. Four top historians have called for this work to be done. Others have expressed surprise it is not a priority.

---

[1] More detail is given in the post, ‘It is short perch: historians, please note.’

[2] R. Willis, ‘On foundations of early buildings recently discovered in Lichfield Cathedral.’ The Archaeological Journal, (1861), 18, 1--24.  

[3] There is a problem with naming the people Saxon or Anglo-Saxon (the preferred title in academic publications). Based on surviving texts, early inhabitants of the region were commonly called englisc and angelcynn. From 410 A.D. when the Romans left to shortly after 1066, the Anglo-Saxon term only appears three times in legal charters in the entire corpus of Old English literature and all are in the tenth century. It is used here because Anglo-Saxon is understood despite being inaccurate.

[4] At Hereford, there were three supposed apses in 1079 which changed between 1226 and 1246 to square chapels, but it is very uncertain.

[5] W. Rodwell, W. Revealing the history of the Cathedral., Cathedral report in the Cathedral Library. (1992). 24–34;

W. Rodwell, An interim report on archaeological excavations in the south quire aisle of Lichfield Cathedral. Lichfield Cathedral report in the Cathedral Library (1992) 1—8 and W. Rodwell, Revealing the history of the Cathedral. Lichfield Cathedral report in the Cathedral Library, (1994), 20–31.

[6] R. Krautheimer, R. Early Christian and Byzantine architecture. (Harmondsworth: 1965).

[7] K. J. Conant, Carolingian and Romanesque architecture, 800–1200. (New Haven and London: 1978).

[8] The old church, c. 475, resembled Wilfrid’s church at Hexham according to Clapham (Oxford: 1930).

[9] S. M. Crosby, The Royal Abbey of Saint-Denis from its beginnings to the death of Suger, 475–1151. (New Haven and London: 1987).