Part 15

In the course of last autumn,[28] after having spent three weeks in climbing Swiss mountains, I was able to devote a few days, on my way home, to a district which, as far as I had been able to gather from books, appeared to contain a mine of interest for the architect, not less than for the geologist and the lover of natural scenery. From Lyon I went by Monistrol to Le Puy, which was the grand object of my tour; thence by Brioude into Auvergne, and through Issoire, Clermont-Ferrand and Nevers, to Bourges and Paris. I was so much struck by what I saw, that, though I am well aware that my visit was too hurried to be at all exhaustive, I think I cannot do better than give you the results of my journey, in the trust that what was full of interest, novelty, and instruction for myself, may be of some use also to others who have not been able to make this journey for themselves. The complete-Gothic architecture of Velay and Auvergne is not, it is true, to be compared to the best work in the north of France. I am not, however, going to tell you about it, but about an earlier style, which, as I hope to show, has special value as illustrating, among other things, the way in which French Gothic was developed from Romanesque and Byzantine buildings; and our attention will, therefore, be almost entirely devoted to buildings which are either Romanesque or Romano-Byzantine in their character, or belonging to the period of transition from those styles to first-pointed. The complete-Gothic buildings are comparatively few, and have no special value; and I shall, probably, not have time now to refer to them even in the most cursory manner.

I will begin with Le Puy, the ancient capital of Velay. The city is crowded up the side of a volcanic rock, one end of which is crowned by the picturesque mass of its Eastern-looking cathedral. It consists of a network of narrow streets not passable by carriages, and reminds one forcibly of some such city as Genoa. Above the rock on which the cathedral is perched rises another, called the _Corneille_, on which are some old fortifications, and which has just been crowned by a monstrous image of the Blessed Virgin, made of the metal of guns taken at Sebastopol, to whose charge I may fairly lay much of the imperfection of my account of the buildings beneath her feet; for I had the ill-luck to arrive at Le Puy only three days before the inauguration of this statue, and I found the whole city so entirely occupied with the preparations for the fête, that it was with the greatest difficulty that I examined the cathedral at all, and into some portions of it I was quite unable to penetrate; whilst the only condition on which I could obtain rooms at an inn was that I should not stop for more than two days, and should make room for some bishop, prince, or cardinal (of whom there were a legion on the road), before the great fête-day. I had to work very hard, therefore, to do as much as I did, and I make no doubt that a more leisurely and uninterrupted examination would have enabled me to discover and do much more. Separated from the great volcanic rock I have already mentioned by one or two furlongs only, is the smaller, but even more striking rock, called the _Aiguille de S. Michel_, and crowned with a little chapel dedicated to that Archangel. It rises, in the most abrupt and precipitous manner, to a height of about 265 feet. The distant background includes a series of truncated conical hills, evidently ancient volcanoes, and from almost every point of view a landscape of the most picturesque and extensive description is seen. Rarely have I enjoyed a more charming ride than that which, for the last twenty miles into Le Puy on the road from S. Étienne, made me generally acquainted with the remarkable physical formation of this mountain district; beautiful throughout, it was at its best just when, some twelve or fifteen miles before I reached the city, I first saw the “angelic” church, as it is styled, standing up boldly on its rock, the centre of an almost matchless landscape.

The story of its claim to this style of “angelic” is this. Bishop Evodius, at the end of the sixth century, on being made first bishop of Le Puy, wished to construct a church; the Virgin, who had before shown to S. George the place where she wished one to be built, appeared to a sick woman on the Mount surrounded by a crowd of angels, and desired her to tell Evodius to proceed at once with his work. After much prayer he went to Rome, and the Pope sent back with him an architect and senator named Scutarius, under whose auspices the church was soon built, and whose tombstone is still to be seen near the transept door. Evodius and Scutarius then started for Rome again, but on the way met two old men, who gave them two boxes of relics, and desired them to return to Le Puy, saying that as soon as they arrived with the relics before the church the doors would open, the bells would ring of themselves, the whole interior would be bright with torches and candles, and they should hear divine melodies, and smell the sweet perfume of the heavenly oil which had served for the consecration of the church by the angels. Everything happened just as had been foretold, and Evodius felt it unnecessary again to consecrate his church, which from that time to the present day has been called the “angelic” church. No doubt you all know how curious a parallel to this legend the history of our own Abbey of S. Peter at Westminster affords.[29] But in searching for information about the churches of Auvergne, I came upon a continuation of the Le Puy legend, to which the Westminster story affords no such parallel. This second legend tells how, when the seraphic basilica of Le Puy had been thus dedicated, S. Anne descended from heaven to visit the palace of her daughter. Not content with this human work, she seized the hammer of the master-mason, and, taking wing, descended on the summit of a hill, and, turning towards Auvergne, which to her mind offered no church worthy of the Queen of Heaven, she threw the hammer, saying as she threw it, “On the place where the hammer falls a church shall rise.” The hammer fell on the right bank of the Allier, and immediately there rose from the soil like a flower the church of Les Chases, which was dedicated forthwith to S. Mary.[30]

Let us now leave legends, and direct our attention to the ground-plan of the cathedral. Its architects have ingeniously contrived to cover the whole of the summit of the rock on which it stands. It consists of a nave with aisles, transepts, a choir, and choir-aisles, and a steeple at the east end of the north choir-aisle. To the south of the cathedral is the modern bishop’s palace, whilst to the north are the cloisters, two grand halls, some ruins, and to the north-east a chapel dedicated to S. John and other buildings. There are entrances in the east walls of each of the transepts, but these were rather intended, I suppose, for the exit than for the entrance of the people, and the mode in which they were admitted forms one of the most striking features of the whole scheme. I said that the church was built on a rock, and its western face, forming one of the principal streets of the city, is so steep as to consist alternately of steps and inclines, until, at a short distance in advance of the west front, it is changed to an almost interminable flight of steps. The grand west entrance is an open porch, like an enormous crypt, beneath the three western bays of the nave and its aisles, whose walls and piers it reproduces in its plan. The steps[31] formerly rose in a straight line, until they came up in the very centre of the church, in the fifth bay of the nave, and in front of the roof-loft, and of the miracle-working image of the Blessed Virgin, which, brought from the East and given to the church by S. Louis, was, until its destruction in A.D. 1789, the greatest attraction for pilgrims in France.[32] This singular entrance, and the mode of exit by the eastern doors of the transepts, gave rise to an old saying, “In Notre Dame du Puy one entered by the navel and went out by the ears.” Unfortunately, however, the central entrance has been diverted, and after ascending a hundred and two steps, and arriving at the Golden Gate, as it was called, the passage branches right and left--on the left ascending into the cloister, and on the right winding round the south side of the church, until the hundred and thirty-fifth step lands the weary pilgrim in the south aisle, near the transept.[33] This, then, is the general scheme of this most singular church. Let me now go on to describe it in detail, beginning with the oldest portion. This comprises the choir, the transepts, and crossing, and the two easternmost bays of the nave. The choir is completely modernized, and I am unable to say whether any portion of the internal arrangement is old. It presents the peculiarity of a square exterior and a circular interior. This is a not uncommon arrangement in the earliest Italian examples of the apse, and is seen at St. Mark’s, Venice, and elsewhere. The arches opening into the choir-aisles are old, and I believe that we may venture to say that the original plan must have been very nearly the same as that of the church of S. Martin d’Ainay, at Lyon, in which the choir-aisles are shorter than the choir, and all are terminated with apses.[34] I shall have other occasion to point out that at a later date the architects of Ainay and of Le Puy must have been the same. The date of the foundation of Ainay is some time in the ninth century, and it was carried on until the end of the eleventh; but the apse and capitals of the columns of the crossing--for the columns themselves are Roman--cannot, I think, be later than about A.D. 940 to A.D. 1000, which latter would, I think, be the date generally accepted for this portion of the work at Le Puy. To proceed with my notice. The crossing is surmounted by a quasi-dome, carried up as an octagonal lantern, much of which has been modernised in restorations, whilst much is quite new; though the universality of the raised central lantern in the churches of the district makes it probable that it is, to some extent, a proper restoration.[35] The transepts are covered with barrel-vaults, strengthened by transverse ribs of a square section below them; the small apses in their end walls have semi-domes, and the tribunes which cross them are groined with quadripartite vaults without ribs. The whole of the nave is covered in the same way as the crossing, each bay being divided from the next by bold transverse arches, and having a quasi-dome, supported by arches across the angles of each compartment, and all of them, in truth, being not domes, but eight-sided pointed vaults, springing from the octagonal bases thus contrived. There are no pendentives, properly so called, and the construction is, I should say, that of men who desired to erect domes, but had no knowledge whatever of the way in which they were constructed in the East; or--to take a more favorable and, perhaps, juster view--of men who, desiring to give a small building the greatest possible effect of space, to roof it with stone (not knowing anything yet about flying buttresses), and to light it from a clerestory, actually solved all these points in a successful way. Where this kind of roof was first attempted I am quite unable to say. Certainly the central lantern at Ainay is so identical in character with some of those at Le Puy, that the same workmen must have executed both; but there seems to be no other example in the same district as Ainay, whereas at Le Puy, and in Velay and Auvergne, everything is more or less roofed on the same principle. The second portion of the cathedral at Le Puy consists of the third and fourth bays of the nave, and the third portion of the fifth and sixth bays.[36]

The latest portion is of early pointed character, and not later in date than _circa_ A.D. 1180 to 1200, and it was at the same time that this was erected that the greater part of the enormous substructure forming the porch was also completed. The aisles throughout the church are vaulted with quadripartite vaults, the three western bays alone having ribs. In the two western bays there are engaged shafts both in the porch and above it in the nave, but the rest of the piers are of the simplest plan, large and generally cruciform in their section, save at the crossing, where the arches are carried on coupled detached shafts. There is much elaborate sculpture introduced in the capitals of the pilasters and columns of the nave, but it is nowhere of any very high merit, and is so inferior in delicacy and beauty to the sculpture of the same age to be seen on the banks of the Rhone, that I should attribute it to a native school of sculptors acquainted, probably, with none but inferior Roman sculpture, from which they endeavoured to develope a style for themselves. A clerestory of wide and rude round-headed windows, one in each compartment, lights the series of domes in a very effective manner.

The arches across the nave are very bold, and, in the wall above them, an opening is pierced under each of the cupolas. As is generally the case, however, in churches covered in this way, very little is seen of the real vault in any general view of the church, these transverse arches only, with the quasi-pendentives above them being seen. The pendentives are true semi-domes, constructed in alternate courses of dark and light stone, and the difference between their plan and the square angle in which they are placed is skilfully concealed by detached shafts, with capitals placed under the pendentives.

I think you will agree with me that considering its early date (no part probably later than _circa_ A.D. 1150 or 1180), it would be difficult to find a grander or more nervous scheme, or one which, with such small dimensions, conveys nevertheless so great an impression of size and importance. The choir-aisles were altered at various times. That on the south has been rebuilt in second-pointed of poor character, and is now a mere passage-way to the modern sacristy, and that on the north was probably interfered with not very long after its first construction, when the great steeple which now abuts upon it was commenced. M. Mérimée,[37] in his very interesting description of the church, suggests that the base of the tower was originally a baptistery, but I see no reason whatever for this suggestion and it is impossible to doubt, when we carefully examine the whole design, that though the steeple was long in building, the main feature in its design was from the first just what we now see it to be. Moreover, the chapel of S. Jean close by is said to have been the baptistery for the whole city until within the last sixty years. The design of the steeple is very bizarre and unusual. It consists of a long series of no less than nine stages on the exterior, and it diminishes rapidly in diameter, and is, perhaps, on the whole, more curious than pleasing in its outlines. If you look at the ground-plan you will see that its construction is most remarkable. The internal diameter of the tower at the base is twenty-four feet six inches, but this is reduced to only twelve feet by four detached piers, one foot ten and one-half inches square. These piers are carried up from the base to the very summit, detached in the three lower stages, and forming part of the thickness of the wall in the portion above. The highest stage of the steeple, twelve feet in internal and sixteen feet in external diameter, is therefore, as nearly as possible, carried up on these four piers, and the rapid decrease in the external dimensions, from thirty-six feet to sixteen feet, was only rendered possible by this very ingenious mode of construction. So far as I know there is only one other example of the same scheme, viz. in the steeple of the cathedral of S. Étienne at Limoges. Here, however, the base is the only portion remaining of the original work, and the columns are cylindrical in place of being square, but it is evident that the intention was the same as at Le Puy. The steeple at Limoges is probably the first in point of date. M. Viollet-le-Duc dates it at about A.D. 1050, but the Abbé Arbellot, in a learned paper on the cathedral, in the _Bulletin_ of the _Société Archéologique et Historique du Limousin_, maintains that it was certainly built before A.D. 1012, when the Bishop Arnaud de Périgeux, after assisting at the consecration of Bishop Gerald at Poitiers, accompanied him to Limoges, and put the cords of the bells into his hands. The lower part of the steeple at Le Puy may, I think, safely be referred to the end of the eleventh century, and its completion to the end of the twelfth, whilst the planning appears to me to be thoroughly characteristic of a Byzantine artist, the construction of the piers in the lowest stage being almost identical with that of the main piers under the domes of S. Mark’s, Venice, and S. Front, Périgueux.

The arrangement of the belfry stage, with its gable on each face, is very noteworthy, and is, perhaps, one of the earliest examples of a type which was developed afterwards into the well-known arrangement of the belfry of the south-west tower at Chartres, and this, with the influence of the churches of the Rhine,[38] developed in almost all subsequent modifications of the spire with its gabled spire-lights; one of the windows under this pediment is planned in a most ingenious manner, presenting externally a semi-dome pierced by two pointed arches; another window is pierced with a trefoil head, the diameter of which is much larger than that of the light it surmounts. This is a favorite form of cusping throughout this district. I have seen it in Lyon, at Vienne, often at Le Puy, at Brioude, at Notre-Dame-du-Port, Clermont, and in the south porch at Bourges; and there can, I think, be little doubt that it is somewhat Eastern in its origin, and analogous to the horseshoe form of arch.