Today I am delighted to be able to share with you a guest post written by my colleague Dr Nicholas Groves, in which he discusses the evidence for the most persistent of Ecclesiological myths, that of the ‘Devil’s door’. Many medieval doors in churches are blocked up, some of the blocked doors are to the north of the church and some of them to the south. Blocked doors on the north have been subject to a mythical explanation and Dr Groves points out that the myth has no evidential basis whatsoever.
‘Devil’s Doors’ Revisited
Many, if not most, mediæval churches have a north door into the nave. In church guides and other popular literature, this is frequently called ‘the Devil’s door’, as it was supposed to have been left open during the ceremony of baptism so that the Devil, when he left the child, could escape from the church. Most of these sources say merely that it was ‘traditionally’ or ‘formerly’ left open: but adduce no hard evidence to prove that it ever was. My suspicion is that this is a piece of unintelligent post-Reformation antiquarianism resulting from a lack of knowledge of mediæval liturgy, and like the so-called ‘leper squints’ (low side windows), needs re-examining.
– I –
First, this theory ignores the fact that in many churches, the north door is the principal entrance. This is the case at, for example, St Peter Mancroft in Norwich, where the north side of the church faces the Great Market: the north porch is a grand affair of two storeys, while the south one is much plainer and smaller. St George Tombland in the same city, though now entered from the south, was originally entered from the north, as the north porch was built directly onto the nave, with the north aisle being butted up to it later, while the south porch is built onto the south aisle. St Julian, conversely, appears to have changed its main entrance from south to north, where it remains. Which door of the two is the principal entrance will be determined by where the majority of the houses in the parish lie, or possibly (in country parishes) by the situation of the manor house – this seems to be the case at Paston, where one approaches the church from the north, but finds the main entrance on the south; in town parishes it is usually determined by which side of the church the street runs past. Of the greater churches, Durham Cathedral is entered by its north door, as the monastery lay to the south (and the west door was blocked by the Galilee chapel), and Westminster Abbey also has a grand entrance on its north side as does Wymondham Abbey (illustrated above).
Second, this supposition shows a lack of knowledge of the theology and ritual of baptism in the Middle Ages. Although it was not formalized until the eleventh century, the Sarum rite, which came to be the dominant rite in England, was merely a variant and development of older rites, and thus its outline may serve. Churches had of course been in existence in Great Britain from a very early date, but stone ones, which are what concerns us, did not come into general existence until after the conversion of the Anglo-Saxons, initiated by Augustine of Canterbury in 597, and most were rebuilt (with or without north doors) after the eleventh century in any case.
The Sarum rite of baptism was in two parts: the first, ‘the making of a catechumen’, took place in the church porch (or at the door if there were no porch). This consisted of the following elements:
The priest breathes on the baby.
The baby is named, and signed with the cross.
Salt is placed in the baby’s mouth.
The baby is exorcised, including signing with the cross by the acolytes.
The priest signs the baby with the cross and then exorcises it.
The gospel is read (Matt 19:13).
The ears and lips are anointed, with the word Ephphatha.
Lord’s Prayer, Hail Mary, and Creed said by priest and godparents.
The baby is signed with the cross on its right hand.
At which point the baptismal party moves into the church for the baptism proper. Although the formal renunciation of Satan took place just before the immersion, the actual exorcism (i.e., driving him out of the child) had already taken place during the catechumen ceremony – outside the church proper. So, by the time the child is brought into the church, it is already free of the Devil. One then wonders why the Devil, when he was exorcised, would need to flee through the consecrated building from south to north in order to escape (assuming a south main entrance): were the ceremony performed at a north door, then he is already outside it! This may, of course, reflect an imperfect understanding of the baptismal ceremonies: that it was not generally understood what was going on in the porch (one does wonder how many rural priests could have explained it coherently), and that the actual baptism itself was seen as the moment when the Devil left the child.
A less coherent version of the myth is to be found on the BBC’s Domesday Reloaded website:
St. Nicholas’ Church in Boarhunt is made of flint and stone with a tiled roof. The North Door, known as the Devil’s Door ,was never used by common people, only witches and evil spirits used it. It was said that if a possessed child was christened the exorcised spirit fled through the North Door. After the civil war between the Royalists and the Roundheads, many doors were bricked up to stamp out superstition about witches. An ordinary person could walk through that door and instantly gain the repute of being a witch or warlock. These people were either mobbed or if a minister was near, put to the test: either way, they died. Then again it could be the door was blocked up for some completely mundane reason!!!!!!
One may be excused for asking exactly why a witch or evil spirit would wish to enter a church, and the information on baptism is garbled to an extreme.
– II –
So, if these doors were not constructed for the Devil to make his escape, what was their purpose? The answer lies, I suggest, elsewhere in the liturgy. On all Sundays, a procession was held before the Sunday mass, which perambulated inside the church itself. But on some occasions, the procession went round the outside of the church: the best-known of these occasions is Palm Sunday, when the entire parish joined in, re-enacting the triumphal entry of Christ into Jerusalem. It left the church by the north door, proceeded in a clockwise direction round the east end of the church, and re-entered by the south door. It has to be borne in mind that the geography of many churchyards has been drastically altered by the number of post-Reformation burials and that they now stand far higher than they did in the Middle Ages, so processional routes are not always now apparent. In some cases, where the church has expanded to the very edges of its churchyard, processional ways were constructed under the chancel (examples are to be found at St Peter Mancroft and St Gregory Pottergate in Norwich,  and at Walpole St Peter in west Norfolk) or under the tower (examples at St Peter Mancroft and St John Maddermarket in Norwich).
This, of course, does not mean that north doors were not left open at baptisms, but if they were, then (as with many other things) it was a piece of folk-religion, and definitely not part of the official liturgy. Some remembrance of this may perhaps still be found in the belief that a baby should cry when it is baptized, as that means the devil has departed from it; if it does not cry, then he has not done so: this, of course, casts the popular view of the efficacy of baptism itself in an interesting light. On the other hand, while this custom of leaving the door open seems to be ‘common knowledge’, and is regularly repeated, actual contemporary records of it happening are rare, if not non-existent: as noted above, the words ‘traditionally’ and ‘formerly’ are loosely used, with no actual evidence. Folk memory is a notoriously fickle thing and much given to embroidery, and one does wonder if an odd occurrence found by an antiquary has been extrapolated to universal use.
Likewise, the fact they are often blocked up is said to be to stop the Devil returning. We need to ask exactly when they were blocked up. This seems to happen in the sixteenth century, judging by the buildings themselves, and it is often explained (as above) that it was done to prevent the superstition lingering. More likely is that, once processions were banned, the extra door, on whichever side, was not needed, so by blocking it up an extra area of wall against which the newly-fashionable box-pews could be erected was provided.
However, the church guide to Wretton All Saints says: ‘During the 13th century all churches were ordered by Pope Innocent III to block off the north doors because he decided the practice of leaving them open was mere superstition’. This statement, as is to be expected, is unsourced, and thus far no credible source for it has been found. Its main weakness is that, as noted above, not all north doors were blocked – and many could not be, without making their churches impenetrable. In any case, most of the blockings are of much later date than the thirteenth century.
One further explanation remains, which may well have some truth in it: they were blocked up to stop draughts. In a country where north winds can penetrate the smallest gaps, this may have been a practical consideration!
– III –
How does the north get this association with the Devil? The north was associated with spiritual darkness: the gospel at mass was chanted by the deacon facing north. This has its roots in the passage in Isaiah 14:12-20, where the ‘Shining One, the Son of the Morning/Dawn’ has been taken in Christian exegesis to refer Lucifer, and the passage to refer to his fall from heaven. Verse 13 refers in some translations to setting up his throne ‘in the north’, although more modern ones use the translation ‘on the summit of Zaphon’, which appears to have been the abode of the gods. (The whole Isaiah passage (14:1-23) is in fact a mock lament over one of the Assyrian monarchs who had been oppressing Israel.) But the north became associated with Satan, and in the plan for setting out the playing-space in the early-fifteenth century morality play The Castle of Perseverance, God has his seat on a scaffold in the east, while Belial has his in the north.
That there was a popular perception that the north side was in some way undesirable is borne out by attitudes to the northern sides of churchyards. They were generally avoided for burials, usually being reserved for undesirable members of the community (suicides, unbaptized or stillborn babies, etc): but one also regularly hears that the north side was not consecrated, which is pure nonsense. This partly underlay the famous burial dispute at Akenham in Suffolk in 1878, where the unbaptized child of a Baptist farmworker was to have been buried on the north side of the churchyard, without any ceremony: the child’s mother was quite convinced the north side was not consecrated ground. The affair became a cause célèbre, and led to the Burial Laws Amendment Act of 1880.
There is also an association of women with the north side of churches: it is the side anchoresses’ cells were usually placed, and also Lady Chapels. A remnant of this association may possibly linger in the fact that the ‘bride’s side’ of a church is the north, with the groom’s family using the south side. The suspicion of women entertained by most mediæval churchmen does not need expanding upon here.
The north is also on the left-hand side as one faces the altar – the Latin for left-hand is sinister. As noted above, the baby was signed on the right hand at the exorcism ceremony, but not the left.
– IV –
So the north is associated with Satan/Lucifer/Belial …, and little-used doors on the north side of churches came to be associated with him too, most probably after they had been blocked up as redundant. The story of their being left open (where they were not blocked) for the Devil to fly out at baptisms probably has it roots in this concept of the north side of a church being associated with the Devil, but one would like to know where and when the open door idea started: did anyone ever leave the north door open at baptisms – either for this reason or another? No hard evidence, either in liturgical texts or in diaries, etc, for this has yet been found, and so it can safely be consigned to the dustbin of romantic but untrue ‘facts’ about churches.
 Of the thirty-four mediæval churches existing in Norwich until 1942 (when three were destroyed by bombs), nine have (or originally had) their main entrance on the north side. Without exception, all thirty-four have two doors, one south, and one north. Of course, many of the 20-odd churches lost before 1600 may have had the north door as the main entrance, also.
 Wooden churches of course had doors, but none of them survives except as archæological remains.
 There were, of course, differences in the baptismal customs of the already established British church which Augustine encountered, but it is not known exactly what they were. Certainly the Roman version was imposed and used after 664.
 I.e., a candidate for baptism. This has its roots in ancient practice, where the adult catechumens were instructed in the faith before being baptized. The reading of the gospel is the remnant of the instruction. The need for exorcizing an apparently innocent baby is explained by the doctrine of original sin.
 Proctor, F, revised Frere, WH: A New History of the Book of Common Prayer, with a rationale of its offices; Macmillan, 1958; p 566. The seventh-century Roman version is to be found on the same page.
 This must refer to the full Use as practised in the cathedral. It is doubtful that any but the very biggest parish churches ran to the luxury of formally ordained acolytes (the fourth of the minor orders), and probably not even them. The second of the minor orders was, of course, exorcist, so this may be a fossilization of that function.
 This also may be significant – see below.
 http://www.bbc.co.uk/history/domesday/dblock/GB-460000-108000/page/20. Spelling and punctuation corrected; multiple exclamation marks in the original.
 In some cases, they left and/or re-entered by the west door, if there was one. West doors and their settings form a distinct liturgical space.
 There is also in towns significant encroachment on the churchyards in the way of buildings: most of the London churchyards are built over; in Norwich, the shops along the east side of St George Tombland stand on the churchyard. Nineteenth-century street-widening schemes also reduced their size.
 St Gregory has an image niche on the north face of the way under the chancel (now known as St Gregory’s Back Alley), thus indicating that the procession followed a clockwise path. There is no corresponding niche on the south face. The way under the chancel at Mancroft was blocked in 1983 by the construction of a coffee bar.
 See Fewins, C: The Church Explorer’s Handbook: Canterbury Press, Norwich, 2007; p 46.
 Jewish Study Bible, note to verse 13: see also Psalm 48:3. The New Oxford Annotated Bible (NRSV) gives Zaphon, with ‘or assembly in the far north’ as an alternative reading, and under Psalm 48 gives ‘Mount Zion, in the far north’, but a note says ‘better “the heights of Zaphon”,’ and explains it as the abode of the storm-god Baal.
 See frontispiece to The Macro Plays, ed Mark Eccles, EETS, 1969. This also contains the delightful stage direction, ‘And he that shall play Belial look that he have gunpowder burning in pipes in his hands and in his ears and in his arse when he goeth to battle’.(spelling modernized NWG).
 See Ronald Fletcher, The Akenham Burial Case, Wildwood House, 1974.
About: Dr Nicholas Groves, MA, BMus, PhD, FRHistS, FBS
After initial degrees in music (Aberystwyth) and in mediæval English (London), he read for the MA in Local & Regional History at UEA, where he worked on the phenomenon of shared churchyards. I then read for the MA in Celtic Christianity at Lampeter, with a dissertation on St Fursa, a seventh-century Irish missionary. This was followed by a PhD at Lampeter, examining the growth of Ritualism in Norwich, 1857-1910, supervised by the late Nigel Yates. (‘The Restoration of Popery’: the impact of Ritualism on the Diocese of Norwich. 1857-1910, with special reference to the parishes of the City of Norwich and its suburbs.) He is a fellow of the Royal Historical Society and is a freelance historian, lecturer and writer. He is the Director of the Centre for Parish Church Studies.