The first historical reference to York after the end of Roman government in Britain is in April 627, when Bede tells us that Eadwine (Edwin) of Deira/Northumbria was baptised in a newly built timber church in the city. Eadwine later started rebuilding the timber church in stone but was killed before it was finished, and the work was completed by his nephew Oswald:
The king's baptism took place at York on Easter Day, the 12th of April, in the church of Saint Peter the Apostle, which the king had hastily built of timber during the time of his instruction and preparation for baptism; and in this city he established the see of his teacher and bishop Paulinus. Soon after his baptism, at Paulinus' suggestion, he gave orders to build on the same site a larger and more noble basilica of stone, which was to enclose the little oratory he had built before/ The foundations were laid and the walls of a square church began to rise around this little oratory; but before they reached their appointed height, the cruel death of the king left the work to be completed by Oswald his successor.
--Bede, Ecclesiastical History, Book II ch.14
Oswald was killed in 642 (Bede Book III ch. 9), so the stone church in York must have been completed some time before then. Where might it have stood?
Eadwine's church was dedicated to St Peter (Bede, Book II ch.14). As this is the same dedication as the present York Minster, it is likely that it stood on or very near the same site.
The present York Minster is an enormous building , on the same sort of scale as the headquarters building (principia) of the Roman legionary fortress which it partly overlies, so Eadwine's small church would have occupied only a tiny fraction of the area of the current cathedral. When the Minster foundations needed underpinning in the 1960s an archaeological excavation was conducted as part of the work (partly displayed in the present Undercroft), which identified massive remains of the old Roman principia under the Minster. For obvious reasons this was the archaeological equivalent of keyhole surgery, but it was sufficient to identify some of the mighty columns of the principia basilica or cross-hall, one of which was later re-erected in the precinct outside the Minster (Ottaway 2004). It was also sufficient to locate the principia precisely in relation to the Minster.
Sketch plan showing the location of the present York Minster in relation to the Roman legionary fortress principia.
There is a much more detailed plan in Tweddle et al 1999 (see references), but this rough sketch should be enough for the purposes of this post. The Minster is aligned east-west and the principia is aligned north-east to south-west with the entrance on the south-west and the cross-hall occupying the north-east range. As you can see from the sketch, the present Minster sits diagonally across the north corner of the cross-hall, with the south transept overlying a corner of the principia courtyard. Most of the principia courtyard is not under the footprint of the current Minster. The excavations carried out during the underpinning work were in the north corner of the cross-hall and parts of the associated range of offices along its back wall (Ottaway 2004).
Excavations in and around the south transept located a graveyard of Anglo-Scandinavian date (ninth to tenth century) containing large numbers of re-used 'Anglo-Saxon' stone sculptures (Tweddle et al 1999). These graves were aligned north-east to south-west, rather than the usual E-W alignment of Christian graves, which suggests that they were aligned on the church associated with the graveyard and that this church was on the Roman principia alignment.
No trace of the church associated with this graveyard or of the small early church built by Eadwine and Oswald was identified in the 1960s Minster excavations, but this is not all that surprising. The excavation was undertaken in very difficult conditions, and the underpinning work on the foundations provided access to only a small part of the cross-hall. The rest of the area occupied by the principia remains unexplored. It is quite likely that Eadwine's church was near the present Minster, but not necessarily directly underneath it.
There have been at least three churches on the site since the one built by Eadwine (counting his timber church and its stone replacement as one): a larger 'Anglo-Saxon' church built in the 8th century and sometimes called the Anglian Minster; a Norman Minster built in the Romanesque style in the 11th century; and the present Minster built in the Gothic style between the 13th and 15th centuries. It is quite likely that some of the successive churches were in slightly different locations, partly because if the old church was still standing there is a logic in building the new one on a different site so the old church can still be used while the replacement is under construction, and partly because the degree to which the Roman remains constrained the site will have changed over time as the Roman buildings collapsed or were demolished and became buried by the rising ground level.
It is not known how long the Roman principia building stood or remained in use after the end of Roman government, but it may have been a considerable time. Say what you like about the Roman military, they could do engineering when it suited them. The west gateway of the Roman fort at Birdoswald on Hadrian's Wall was still standing until the fifteenth century, dated by the pottery found underneath its collapsed arch (Wilmott 2001 p.137-138), and part of the Roman Multangular Tower at York is still standing now. The principia was a massive building judging from the size of the columns, so it may well have stood for some considerable time if it was not being robbed for stone. Ottaway (2004) suggests that it may have been standing and in use until the Danish (Viking, if you prefer) invasions of the late ninth century. If this is the case, the Norman Minster would not have been constrained by its presence, which would explain why it disregards the Roman alignment, but the earlier 'Anglo-Saxon' churches would have been.
This being the case, one plausible location for Eadwine's early church is the courtyard of the Roman principia building. Bede makes it quite clear that the church was a new building, not a re-used or refurbished Roman one (and Bede understood the difference; he specifically says that Augustine in Canterbury at first used an old Roman church dedicated to St Martin, and then repaired another old Roman church in the city; Book I ch. 26 and ch. 33). The principia courtyard may well have been one of the few sites in the vicinity of the York principia in 627 that was not already occupied by standing Roman buildings or Roman ruins and was therefore available for a new build.
A story recorded in a Life of Pope Gregory the Great written at Whitby some time in the early eighth century is consistent with this. It describes Bishop Paulinus and Eadwine giving religious instruction in the royal hall, then hurrying across a public square or street to the church (Tweddle et al 1999). Unfortunately the location of the incident is not recorded (chroniclers can be annoyingly remiss about these things), but if it was in York, it is consistent with the principia cross-hall being in use as a royal hall and the church being somewhere in the principia courtyard.
Eadwine's church was built specifically for his baptism, and this may be another slight clue to its location. Roman forts often contain a well or a water tank in the principia courtyard (Bidwell 1997 p. 71). As the excavated examples are often quite shallow or of limited capacity, they would not have gone far as emergency water supplies during a siege. Paul Bidwell suggests that their purpose was to provide water for religious ceremonies conducted in the principia, and cites the example of the fort at South Shields where the well was built at the same time as the shrine of the standards, before the rest of the building was constructed (Bidwell 1997 p 71). If there was such a well in the principia courtyard at York, possibly there may have been a shadowy tradition that it was a 'lucky' or 'holy' or 'powerful' well (even if no-one remembered why) and therefore an auspicious site for the baptism of a king and the establishment of a new church.
So I would hypothesise that the seventh-century church in York was probably located somewhere in the courtyard of the Roman principia, outside the cross-hall and near the current Minster. Needless to say, other interpretations are possible.
According to a newspaper report in the Guardian, excavations carried out by York Archaeological Trust during building work under York Minster in the spring of 2012 have identified two large post-holes and a jumble of human bones that pre-date the building of the medieval cathedral, located just outside the walls of the Roman basilica. The bones and post-holes are undated; it is possible (but not proven) that they could relate to the seventh-century church. If so, this finding is broadly consistent with the hypothesis advanced above, that the seventh-century church was located in the principia courtyard.
Bede, Ecclesiastical history of the English people. Translated by Leo Sherley-Price.
Penguin Classics, 1968, ISBN 0-14-044565-X.
Bidwell P. Roman forts in Britain. Batsford/English Heritage, 1997. ISBN 0-7134-7100-X.
Ottaway P. Roman York. Tempus, 2004. ISBN 0-7524-2916-7.
Tweddle D, Moulden J, Logan E. Anglian York: a survey of the evidence. Council for British Archaeology/York Archaeological Trust, 1999. ISBN 1-902771-06-0.
Wilmott T. Birdoswald Roman fort: 1800 years on Hadrian's Wall. Tempus, 2001. ISBN 0-7524-1913-7.