“ON a visit to the cathedral church at Whithorn my attention was arrested by a peculiar feature in the ruins at the west end of the Nave.1 Little more can be seen than a long confused mound of earth and stone, from 4 to 5 feet high. It begins a few feet from the gable of the cathedral, and extends for a considerable distance towards the boundary wall of the churchyard. The foundations of walls lie buried in this mound.2 Thirteen feet from the cathedral the lower courses of a rubble wall are visible, lying parallel to the gable. The opening in the centre, on the medial line of the cathedral, is 6 feet wide, the jambs are plain and square, and the wall is 3 feet thick. The two walls which project at right angles towards the west are 10 feet apart. The sketch plan in the margin shews all that is ascertainable at present. This is the foundation on which I have now to build. And my hope is that I may find here the site of St. Ninian’s Church – the first Christian church in Scotland erected in stone.3
King David I. re-established the diocese of Candida Casa, or Whithorn, in the period from 1124 to 1130.4 Fergus, lord of Galloway, during David’s reign founded a priory at Whithorn for canons of the Premonstratensian order. The church, which was dedicated to Saint Martin of Tours, to whom had been dedicated the original church by Ninian, doubtless formed the cathedral church of the diocese.5 Only the nave of the cathedral remains – a thing of patches, to which the builders of many generations lent their hands. The beautiful Romanesque doorway at the west end of the south wall is the one fragment preserved of the early twelfth-century foundation by Fergus. Its position may be held to indicate that this fabric did not extend further to the west than the line of the present gable. It is known that the three walls whose ruins have been described above are parts of the foundation of the steeple or tower which was is existence when Symson wrote his description in the year 1684.6 This tower was of considerable height and formed a striking landmark on the coast.7 As the ruins shew that there was a building of some kind to the west, and as it is clear, from an examination of its western face, that walls extended westward from the cathedral gable to join the wall forming the easter face of the tower, it is evident that the tower occupied the centre of some building. Notes and studies on the plans of other structures, to which reference will now be made have led me to conclude that it stood between the chancel and nave of an earlier church than that founded by Fergus. If it is possible to bring the plan to light, the spade may prove that there is an opening in the west wall of the tower, similar to that in the east wall, with a door, perhaps built up, in the south wall.
The opening in the east wall is sufficiently wide to serve the purpose of a chancel arch, for the chancel arch in St. Blaine’s Church, Bute, a building of the twelfth century, is only 5 feet 3 inches wide.8
The chancel arch of the early Romanesque church of Upper Denton, near Gilsland, Cumberland, is illustrated in the margin. It is 6 feet 6 inches wide, the jambs are built of high courses of masonry, and no mouldings are present. The arch springs from a simple chamfered abacus, and at the springing the width is greater than between the jambs. The walls are 3 feet thick. At the ground level they rest on a square projecting plinth.9
The tower of Monkwearmouth Church, County Durham, stands at the west end of the present nave. The open archway in its western wall is 4 feet 9 inches wide. The arch is plain, with only a flat head on the angle, and it rests on a chamfered and beaded abacus, supported on each side by two turned stone baluster-pillars. The doorways in the north and south walls, which are built in high courses of masonry, have no architectural features, and are 6 feet high by 2 feet 2 inches wide. There is a broad opening in the east wall giving access to the present church. Many attempts have been made to solve the difficulties in connection with the plan of this tower. My own opinion is that the church consisted originally of a nave with an eastern semi-circular apse, and a western porch. The entrances were in the north and south walls of the porch and the entrance to the nave was as at present, the west wall of the porch being solid. At a later time, and owing to a change in ritual or to want of accommodation, the nave became the chancel, and a new nave was added to the west. The foundations of this building have been found by digging. The walls of the old porch were retained, and on these the present tower was erected. The west wall, which was solid, was opened up, and the existing archway formed and decorated as the eastern arch of the new nave.
The church at Monkwearmouth was built by Benedict Biscop in the year 674, in the Roman manner, by masons whom he had brought home specially from France. There is every reason to believe that part of the west wall of the old nave and the lower part of the present tower are the remains of this early work.10 It is possible to form an approximate estimate of the date when the additions were made. They must have been undertaken after the beginning of the ninth century, for the earliest historical reference to the erection of a bell tower is in the year 780, in connection with St. Peter’s in Rome.11 As there is no evidence of the influence of the foreign art which accompanied the Norman conquest, they cannot have been built later than the middle of the eleventh century is probably the true date. The presence of finely moulded and turned stone balusters (and many are carefully preserved in cases) seems to indicate that the masons were not foreigners, trained in the traditions of stone construction, but natives, applying the art of timber structures to stone.12
In the old church at Jarrow, County Durham, so closely associated with the name of the Venerable Bede, the tower stands in the centre, between the present chancel and the nave.13 A figured plan and a sketch of the arches in the interior are shewn in the margin. The tower is peculiar in being oblong instead of square, the measurements over the walls being 22 feet by 13 feet 5½ inches. As the east end of the north and south walls of the old nave, recently removed, was for over 10 feet of later date than the rest of the nave, I have no doubt that the tower was square when first designed, although the reason for the alteration cannot now be fathomed. The walls of the tower are 3 feet 1 inch thick, and the jambs of the openings are, in part, built of high courses of masonry. Like the arches, they are plain and square. The arches spring from chamfered abaci, and the width at the springing is greater than between the jambs.14
The church at Jarrow was founded by Benedict Biscop in the year 681. As this was seven years after the foundation of Monkwearmouth Church, we may reasonably assume that the foreign masons were employed for that length of time on the earlier fabric. The present chancel is probably Benedicts work. I am convinced, however, that this is the old nave, and this opinion is supported by the presence of the lines of the apse on the east gable. No part of the original western porch remains. There was, apparently, the same need for extension at Jarrow as at Monkwearmouth, and a new nave was projected to the west of the old church. As the narrow openings in the central tower at Monkwearmouth Church, due to the retention of the walls of the old porch, must have proved inconvenient, the architect at Jarrow overcame this difficulty by removing the porch entirely, and by erecting a tower, new from the foundation. The opening in the south wall is now built up – the one on the north leads to a modern vestry. That a western porch existed originally, and that the present chancel is the old nave is, I think, proved by this fact that the north and south walls of the chancel are not bonded into the tower walls, and that the quoins or corner stones of the gable, i.e., the western gable of the old nave, still remain. The many turned stone balusters – fragments of the added nave – which are carefully preserved in glass cases, indicate that this work was of the same date as the addition to Monkwearmouth Church.15
An old tower, now surmounted by a stone spire, stands at the south-west corner of the thirteenth-century choir of the priory of Restennet, near Forfar. The plan is shewn in the margin […] The door in the south wall is 2 feet 3 7/8 inches wide. The tower measures 10 feet 5½ inches by 10 feet 4½ inches inside, and the walls are 2 feet 8 inches thick. The arch in the east wall is 8 feet 5 inches wide. The jambs, which are built in high courses of masonry, are, like the arch, plain and square. The arch springs from chamfered abaci, and the width at the springing appears to be greater than between the jambs. The sketch in the margin shews this arch from the east side. The walls of the tower, at the ground level, rest on a square projecting plinth. It is evident that this work is of the same character as the work at Upper Denton, Monkwearmouth, and Jarrow. The lower part of the tower former the western porch to a church which doubtless consisted of a nave and semi-circular apse. The entrance to the porch was by the door in the south wall, and the west wall was solid.
Restennet has been associated with the name of Nectan, King of the Picts, who, in the year 710, requested the abbot of Jarrow to send him architects to assist him in building a church after the Roman manner.16 Following the example of Monkwearmouth and Jarrow, the church of Restennet was transformed. The nave became the chancel and a new nave was added to the west. Then it was that the western archway was cut out of the tower. It is probable that some weakness became manifest during the progress of the work, for a part of the wall on the south side was rebuilt, leaving the opening a considerable distance off the centre line.17
A plan of the tower of the Romanesque church of St. Regulus at St. Andrews is shewn on the margin. In the interior the tower measures 14 feet 9½ inches by 14 feet 7 inches. It is 5 feet 1½ inches narrower than the chancel.18 The east wall is pierced by a tall arched opening. The jambs and arch are moulded, and the arch springs from an elongated capital of convex outline and in proportion after the Corinthian order. The arch, which rises flush with the face of the jambs, is very slightly of the horse-shoe form. The entrance doorway in the south wall is now built up. The west wall was solid. The walls of the chancel are still almost perfect, the height from the base to the wallhead being 33 feet 8 inches.19 The squared stones are carefully cut and laid in horizontal courses, measuring in some cases 21½ inches high. The arch in the east gable is of the same design as the arch in the tower. The semi-circular apse has disappeared. The lines of its walls are to be seen on the gable, however, and the foundations were visible in the year 1787.20 It is manifest that we have here a church on similar lines to those already described. The tower formed the western porch to a nave with an eastern apse. But the parallel is closer still. A new nave was added at the west end. An opening was made through the west wall of the tower, and the present archway inserted, of greater height, but with jambs, capitals, and arch of the same design as the other two arches.21 The lines of the side walls of this nave are preserved on the face of the tower, as well as the raggle cut in the stone for the roof, which was higher than the roof of the chancel. The foundations of the nave have been met with in digging. To prove that a nave was attached to this building, I need only refer to the Seal of the Chapter of St. Andrews, of the year 1450, which gives a faithful picture of the church, shewing the tower between a nave and chancel, the roof of the nave being higher than that of the chancel.22
The advanced character of the art of St. Regulus Church is apparent. Couple this with the fact that it was designed and erected as a nave with apse and western tower only, and there is presented a most important piece of evidence bearing on the date of the naves at Monkwearmouth, Jarrow &c. St. Regulus Church must have been erected before these works, as it is impossible to suppose that it would have been designed as it was if a more suitable and elaborate plan was already in general use.
Many attempts have been made to determine the date of St. Regulus Church and tower. The earliest opinion, now abandoned, as it received no support from the testimony of History or Art, was that they were erected in the seventh or eighth century. The more generally received opinion now is that they belong to about the middle of the twelfth century. But the only evidence which has ever been offered in support of this late date is a single historical reference to building operations of some kind on the part of Bishop Robert, in the years between 1127 and 1144,23 a historical reference which has been wholly misunderstood and misapplied. The cathedral of St. Andrews, one of the greatest structures in Scotland, was founded about the year 1162.24 Its art belongs to a period of transition between the twelfth and the thirteenth centuries, while its plan shews the perfected arrangement of transepts and aisles. The distance in time which separates St. Regulus Church from the cathedral must be much greater than eighteen years. The most recent opinion which has been expressed regarding the date is that it may be found in the tenth, the eleventh, or in the twelfth century.25 The following brief historical outline of the See may help towards narrowing this too wide a field.
The reputed relics of St. Andrew were brought to Scotland sometime between the years 731 and 761. The primacy was transferred from Abernethy to St. Andrews in 908. In the same year an assembly was held at the Moot-hill of Scone at which Bishop Cellach appeared, called the first bishop of St. Andrews. Fothadsucceeded Cellach, but was expelled in 954. Malisius was bishop from 954 to 963. Doubtful notices then occur, until reference is made to Cellach who was bishop for twenty-five years (970 to 995). Alwyn was bishop from 1025 to 1028, and was succeeded by Maelduin, who was bishop for twenty-seven years – from 1028 to 1055. Tuthald was then bishop for four years. He was followed by Fothad, who saw the Norman Conquest and solemnised the marriage of Malcolm and Margaret in 1069. He died in 1093. After a fourteen years’ vacancy, the first bishop of the Margaretan church appears in 1107 in the person of Bishop Turgot. He was at the founding of Durham Cathedral, and he died there in 1115. Eadmer of Canterbury was succeeded by Bishop Robert, who was elected in 1128, established a priory at St. Andrews in 1144,26 and died in 1159.27
As the art of the building shews no trace of Norman influence, the date will be earlier than the time of Bishop Fothad (1060-1093). But it is necessary to find two dates. One is for the added nave, which should correspond in some measure with the dates of the naves and towers at Jarrow and Monkwearmouth. The other is for the chancel and tower; and the two dates must not be widely separated from each other, since the art of the added nave, as it remains in the western arch of the tower shews that no appreciable advancement in style had taken place. The tow dates I suggest are, the period between 1028-1055, with Bishop Maelduin the builder of the nave added to the west of the tower, and the years between 970 and 995, with Bishop Cellach the builder of the tower and church which still exist. It is almost impossible to imagine that the designers of the other buildings which have been described and illustrated could have designed this structure. It is in harmony with the Continental art of this period rather than with British art. And the probability that a foreign architect was employed is strengthened by the information, supplied by Bower,28 that Cellach was the first bishop who went to Rome for confirmation.
The date now suggested removes St. Regulus Church far from the period of St. Regulus at the end of the sixth century, and from the time when the relics of St. Andrew were brought to Scotland about the middle of the eighth century. St. Regulus has been identified as St. Raigail, Regulus, or Rule, who was with St. Columba near Sligo in the year 573.29 It is possible that the site of the church he established may be found in the choir of the little church of Kirk-hill or St. Mary on the Crag, outwith the precincts of the cathedral. The cathedral, St. Regulus Church, the nave of St. Mary’s, and the north wall of the provost’s lodging which stands a few yards to the south of St. Mary’s, and was erected about the year 1500, are all nearly parallel, having the same orientation. The choir of St. Mary’s, however, – alone of all the many structures in its neighbourhood, – points considerably more to the north of east. This is a peculiarity to be noted in early church foundations and it merits the closest investigation. The cell on the north side of Iona Cathedral points more to the north than the cathedral. The same is true of the little cell which stands a few yards due west of the abbey of Inchcolm. And it is only a reference to a very early foundation that will explain the fact that the tower of Dunblane Cathedral points several degrees more to the north than the cathedral with which it has been incorporated.30
It is probable that the Northumbrian clergy, when they came to this spot, dispossessed the Columban church and entered into possession of the old church on the crag. It would become the Cathedral of the see of St. Andrews in the year 908, and this dignity it doubtless retained until the time of Bishop Cellach, when the present church of St. Regulus was erected as more befitting the Bishop of Scotland. Bishop Arnold followed this example in 1162 by building the present cathedral. It is evident that St. Regulus Church must have served as the cathedral for the Margaretan Church for some sixty years. St. Andrews therefore, in all probability, presents the unique spectacle of three cathedral churches alongside of each other. It is known that the representatives of the Northumbrian church, with whom the name of Culdee is mysteriously associated, were allowed to retain possession of the old church on the crag. Here an independent position was maintained in a collegiate foundation, although under the shadow of the great cathedral, until the sixteenth century.31 It is recorded that when Bishop Robert died in 1159, after founding the new priory, he was buried in the “old church.”32 The old church must have been this church on the crag. It would be interesting if it could be proved which of these three churches was visited by the pilgrims who came to venerate the relics of St. Andrew.
It is now accepted as a reasonable suggestion that the reputed relics of St. Andrew were brought from Hexham to Scotland by St. Acca, on his banishment from Northumbria about the year 731.33 This date harmonizes with some of the more important statements in the legend of St. Andrew. But a most interesting piece of evidence has now to be noted, which may still further support the received interpretation of the legend as a narrative of two separate and distinct events which happened at periods widely apart from each other, yet related as one. According to the legend, Regulus erected a cross which he had brought from Patras.34 This statement is a fair example of the curious blend of fact and fiction in early legends. The fiction is in attributing to Regulus the erection of a cross and in bringing it from Patras in Greece. The fact appears to be that a cross was erected. And a great part of it is still in existence. The stone is a dark red sandstone, and measures 8 feet long by 10 inches broad. The original length was probably about 16 feet. The side is carved with a beautiful scroll of vine branches and leaves, and I have no doubt that it will be found, when it is removed from its present position – as I trust it will be before very long – that the front and back faces exhibit a series of panels which may rival those on the famous cross of Ruthwell, carved with scenes from the life of Christ. At present the stone is built into the basement of the interior face of the east gable of the cathedral. The legend describes the stone as coming from Patras. It probably came from a quarry far distant from St. Andrews, but experts may be able to determine whether or not it came from Dumfriesshire. The Ruthwell Cross, to which this example seems to bear a strong resemblance, is closely associated with the Northumbrian Church, and the date when it was carved must have been about the beginning of the eighth century. The greater portion of the cross, which was erected for St. Acca at Hexham, is still preserved in the library of the Cathedral of Durham.
A sufficient number of examples of early churches have now been examined in detail. The old churches of Muthill and Dunning, in Perthshire, need only be referred to in passing, in order to direct attention to them as worthy of careful examination. The names are suggestive of great antiquity. In the tower at Dunning the entrance door in the south wall has been modernised, but the archway to the nave remains, a beautiful and richly sculptured example of twelfth century work.
To return to Whithorn. Saint Ninian, according to his biographer, “chose a site for himself in the place which is now called Whithorn. This place is situated on the shore of the ocean, and, running far out into the sea, is enclosed by the sea itself on the east, west, and south, the way being open to those who would approach it only on the north. Here, therefore, by the command of the man of God, the masons whom he had brought with him, built a church, before which, they say, no other had been built in Britain in stone. And inasmuch as he learned that the most holy Martin, whom he always venerated with wondrous affection, had now passed away from the earth to the heavens, he was careful to dedicate it to his honour.”35 The church was founded in the year 412.36 The walls were doubtless plastered over in the Roman manner with adamantine cement, as was done in St. Wilfrid’s Church at Hexham.37 From this circumstance, and not from the use of white material in the building of the wall, the name Candida Casa arose. St. Ninian died on the 16th September, 432.38 He was buried, says his biographer, “in the church of the blessed Martin, which he had himself built from its foundation, and was placed in a stone sarcophagus beside the altar. At his most sacred tomb the infirm are healed, lepers are cleansed, the wicked are terrified, and the blind receive their sight.”39
The Church Council which was held in Britain in the year 692 was assembled by King Ina for the union of the Britons with the Saxons.40 It is not until the year 723, on the establishment of English bishops, that we learn of any successor to St. Ninian. Pichthelm was consecrated in that year. The succession of bishops was interrupted about 794, probably because of the anarchy in the country following the assassination of Æthelbert,41 and nothing further is heard of the bishops of Whithorn, until the see was re-established by King David I. in the early part of the twelfth century.42
The history of the early fabrics referred to above leads me to suggest that the history of this church whose site is under consideration followed on similar lines. Erected in the first place as a nave with apse and western porch, the church was extended by the addition of a new nave to the west, the porch being reared as a tower. If this was the case, then the presumption is that the site of St. Ninian’s Church has been found. The building was probably erected at the beginning of the eighth century, on the establishment of the Saxon see, And yet it is not altogether impossible to suppose, as Bede mentions the existence of a stately church in his time,43 that part of Ninian’s work remained until the last. It was probably planned on the same model as the churches of the early part of the eighth century, for Ninian brought his masons from the renowned abbey of Marmoutier in France.44 The preservation of the church as a separate building, when the early twelfth-century foundation was laid, was an example followed at St. Andrews and other places.45
The latter-day name of this early church is known. It was called the Outer-kirk, or Cross-kirk;46 and it is possible to give a sketch drawing of it here. As mentioned above,47 the seal of the chapter of St. Andrews, of date 1450, presents an interesting view of the early church of St. Regulus. Two seals of Whithorn are preserved,48 and on both a picture of the old church appears. One of the seals is of the year 1613; the other, of which a drawing is given […],49 is probably, judging from the character of the lettering, of about the middle of the fifteenth century. It will be noticed that the tower stands between the very short chancel and nave. Its high pitched roof resembles that on the tower in the seal of St. Andrews. The pinnacles, and probably the upper part of the tower, are late works. The panel shewn in the base of the tower, extending upwards to the wall-head, is evidently intended to represent the deep recess between the chancel and nave, due to those parts of the building being broader than the tower. We have seen that this feature existed in all the examples referred to above, with the exception of the church at Jarrow.
The fame of the miraculous powers possessed by the relics of St. Ninian spread over Europe. It was evidently to this church – this Outer-kirk – the pilgrims came. There is no more striking feature in the strangely romantic character of King James IV. Than his great veneration of this saint, and many were the visits he paid to the shrine. James V. took staff in hand in the years 1532-1533; and pilgrimages to the shrine had become so much a part of the life of the people that they were continued after the period of the last Reformation. They were only abandoned when prohibited and made punishable by an Act of Parliament in 1581.50
This is neither the time nor the place to enter on a discussion of religious movements. It ought to be borne in mind, however, that there have been three great Reformations in Scotland. The Reformation at the end of the sixteenth century was, perhaps, more of a revolution than a reformation. It led to no new development in the planning of fabrics. An attempt was made to get back to primitive forms of church government, and it may be said that the new religious order instituted was of secular clergy. The Reformation at the end of the eleventh century; with which the name of St. Margaret will for ever be associated, was necessary, because the early church had become petrified as a local church, having for long ceased to regard the new developments and reforms emanating from the great centre at Rome. Monastic orders to a large extent took the place of secular clergy. The influence exerted on the fabric of churches may still be seen in our great cathedrals and abbeys, with their transepts and aisles, and wide ranges of conventual buildings surrounding a garden. The altar now became the focus of the church. The Reformation at the end of the seventh century is associated with the name of St. Wilfrid of York. The Columban Church, which was monastic in its establishment, and whose buildings were of the most primitive type, and, as a rule, of timber, gradually lost touch with Rome. The early saints were held in great reverence, and little or no attention was paid to the advancing tide of thought and organization. Fresh from his Roman school Wilfrid found the clergy at home preaching and practising heresy. Ardent and eloquent, the Columban Church gave way before him, and soon his influence was exerted over Northumbria and the lowlands of Scotland. As a revolt against the evils of monasticism he instituted orders of secular clergy. Wilfrid’s Reformation brought with it a model church plan. As described above, it consisted of a nave, with a semi-circular apse at the east end and a porch at the west. This plan probably came from Rome, through France, but I find the original, not so much in the Roman Basilica, as in the smaller and simpler Roman Curia.
To many it may seem remarkable, yet it is true, that the table or altar held a more or less subordinate place in the arrangement of the interior of the early church. Wilfrid was not revolting from early Roman custom when he made the focus of his church the bishop’s or abbot’s seat or cathedra.51 The seat was placed on an elevated platform in the centre of the apse behind the altar, and the clergy were accommodated on each side. This was a suitable and satisfactory arrangement, designed for a church which looked to the bishop or abbot as the apostle of Christ. The change which was made at a later time may be traced to the Council of Rheims, held on the 3rd October 1049, when it was forbidden for any bishop, except the Pope, to assume the title of “Apostolic.”52 The church still maintained its logical position in that the Pope alone retained his place behind the altar, as he does at the present day. Two bishops’ seats associated with the northern church and with the name and period of St. Wilfrid still remain. One is Wilfrid’s own seat at Hexham. The other, in Beverley Minster, is the chair of St. John of Beverley, who preceded Wilfrid in the see of Hexham. What I presume was Wilfrid’s chair at York is referred to in a charter of the time of Henry VII.53
Development was not impossible with the plan adopted by Wilfrid. It has been noted above that as necessity arose the fabric was enlarged. But it is important to observe the method of enlargement. In every case referred to a new nave was added at the west end. The old porch was retained in the centre, and on this a central tower was reared. With the exception of Jarrow Church, in the examples cited, the tower is of less width than the chancel and nave. At Monkwearmouth and Jarrow there are openings in both the north and south walls; in Restennet and St. Regulus the opening is in the south wall only. When church plans were brought to this stage it was but a short step farther, and the transeptal form was attained. The side arches were already in existence, and all that was necessary in order to secure rudimentary transepts was to utilize the unsightly recesses on the exterior elevations by building walls across in line with the walls of the chancel and nave.54 Development was slower in North Britain than in South Britain or the Continent, and there was certainly no opportunity for the gradual and full realization of the transept in Scotland because of the inflow of the great tide of advanced foreign art towards the end of the eleventh century. But it is interesting to notice that there was a movement in that direction. If in this the architects of Scotland were but following in the wake of their brethren on the Continent, a reasonable and practical theory of evolution has been established.
Here endeth the first chapter. Setting out with the expressed hope of finding the site of St. Ninian’s Candida Casa [White House], to that search I have added the desire to place the study of early church fabrics on a broader basis than I found it.
But a second chapter is still required, and for that the spade is more fitting than the pen. The task of digging must be left for others more conveniently situated. If, on the ground which I must assume to be the site of St. Ninian’s Church, the spade brings to light all that has been anticipated, my leap into the dark will be amply justified. But if, on the other hand, the facts are not as I have ventured to predict, the reader may yet be pleased to think that this note has not been written altogether in vain.
P. MACGREGOR CHALMERS.”