Pevsner: Wells Cathedral

INTRODUCTION. Wells Cathedral is the work of two periods, c. 1180-1240 and c. 1290-1340. The earlier achieved the creation of a completely English yet completely Gothic style, the later - together with Bristol - represents the most original treatment of space in architecture of which any country at that time was capable.

Wells started early in the C8 as a collegiate church. It became the see of a bishop in 909. Of the Saxon cathedral nothing at all is known. A Norman replacement was undertaken by Bishop Robert de Lewes (1136-66) and consecrated in 1148. It may be assumed that the present church was in its plan to a certain extent influenced by the Norman building, that perhaps even outer walls and positions of piers remained the same; but there is no proof. The building with which we were familiar was begun by Bishop Reginald who died in 1191. The earliest documents referring to the fabrica Wellensis ecclesie, to operaciones, and to the optatam consummacionem are of between 1176 and 1188.¹ On the other hand there was a period of unrest immediately after Reginald’s death, and Bishop Jocelyn, who was elected in 1206 and died in 1242, says in a charter of that latter year: ‘Ecclesiam sancti Andree Wellensis, que periculum ruine patiebatur pre sui vetustate...edificare cepimus et ampliare. ’ This statement contradicts the earlier documents and other documents referring to grants of money for the ‘strucrura surgentis ecclesie’ (probably before 1191) and the construction of the novum opus (1195-1205). One might say that it would be only human of Jocelyn to minimize what had been done by his immediate predecessors, or one might suggest that what was menaced by collapse because of antiquity was only one particular part, say the nave. In any case if we assume that Jocelyn was honest we are forced to put as much as we can into the time after 1206. The cathedral was consecrated in 1239, which does not necessarily mean that it was complete.²

Reginald’s and Jocelyn’s work comprised an E end with a one-storeyed straight ambulatory, a straight-ended chancel, crossing with tower over, transepts, nave, and W front, and also the undercroft of the chapter house.

The work of c. 1285-c. 1345 comprises the chapter house, a new E end with an E transept and E Lady Chapel, and the completion and strutting of the crossing tower. Dates are as follows. The doorway from the chancel aisle to the passage and the passage itself belong to the last third of the C13. In 1286 it was decided to complete a ‘nova structura’, long since begun. That must be the Chapter House proper. It first appears under the name Domus Capitularis in 1319. That may mark its completion³; For the beginning of the Lady Chapel we have no date. The chapel is, however, called ‘newly constructed’ in 1326.⁴ By then work on the new campanile was finished. It had been begun in 1315 and a grant for roofing was made in 1322. In 1325 several documents mention a ‘novum opus’; most probably the chancel. A burial (Bishop Drokensford) took place in 1329 in the E transept S of the Lady Chapel. When the high vault of the chancel was put in, we do not know. In 1338 the church is called enormiter confracta and enormiter deformata. It is suggested that that refers exclusively to the effects of the new crossing tower on the crossing piers and that the strainer arches were the remedy. They can therefore be dated c. 1338-40.⁵ By 1345, at the latest, it can be assumed that the whole E end was complete. Later work is confined to the W towers and the cloisters, a remodelling of the crossing tower, a raising of parapets on the C13 part, and a thorough remodelling of the C13 lancets by introducing simple two-light Perp tracery, apart of course from the Precinct which will be dealt with separately.

EXTERIOR. It is comparatively simple to appreciate the work of the first period. The subtler points of development come out more clearly inside. Church building as a rule started in the Middle Ages from the E. We will assume that that was so at Wells too, and begin where we can see what remains of the CHANCEL. The best point is in the churchyard to the E of the cloister. The broad aisle buttresses remain and the corbel-table, and of the clerestory the narrower buttresses, the string-course which goes round the original windows like a hood-mould, and again the corbel-table. The break between what we may for the moment call c. 1180-c. 1200 and what was done in the early C14 is obvious enough, and it is equally obvious that it takes place in the clerestory one bay further W than in the aisle - proof of the fact already mentioned that the church of c. 1180-c. 1200 had a one-storeyed ambulatory.

The TRANSEPTS come next. They have both E and W aisles. But they differ in other ways. That on the S side is more easily visible than in monastic cathedrals, where extensive buildings usually rise to the E of the cloister. It has broad buttresses, three ground-floor windows, then a row of blank pointed arches with continuous roll-moulding - the hallmark of Wells, as we shall see - cut into by the lowered sills of three stepped lancets above. The gable again has three stepped lancets, the central one wider and blank. On the buttresses stand tall polygonal pinnacles like chimneys, their long shafts with blank arcading, again without any capitals. The E aisle windows were enlarged in conjunction with the new presbytery, i.e. are not of the simple early C15 forms that most other windows are.

On the N side the pinnacles on the buttresses are in three tiers, with blank panelling. The blank arcade below the main upper windows has eight arches of which the two .middle ones are higher. The window zone consists of three tall arches of even height and two narrow blank bays l. and r. Under the tall arches are three stepped lancets, and above the lower of them paterae of stiff-leaf foliage, the first appearance of another leitmotif of Wells. In the gable finally there are twelve stepped blank arches with continuous moulding. Above the middle pair stiff-leaf in the spandrel. This description will show that the N transept has more variety of motifs than the S transept, perhaps a sign of later completion.

The system of the NAVE is again very simple. Broad buttresses below, narrow buttresses above. The lower buttresses have on the N side a stepping of the set-offs - four little steps on the first, three on the second set-off - which Salisbury took over and made a speciality. Corbel-tables on aisle and clerestory. The window tracery again Perp and the parapet raised. More ornate only the doorways, and especially that leading into the church from the N under the porch. It is placed five bays from the crossing and six bays from the W front.

In the S aisle is a doorway at the E end into the E walk of the cloister. It is a tall rather narrow doorway with an inner continuous chamfer, then two pairs of columns on each side separated by a step in the wall of the jamb. The columns have one shaft-ring and stiff-leaf capitals. The capitals (cf. below) are more developed than in the chancel, less than in the nave Broad arch with filigree of stiff-leaf arranged in oblong shapes. The hood-mould on head-stops is damaged by the later cloister.

The N entrance is a twin doorway with a middle post or trumeau. The trumeau has three shafts and two hollows between them, and shaft-rings to the shafts. Each doorway has a continuous moulding, which is however stopped short by the trumeau. The whole entrance is enriched by one order of columns with shaft-rings, one outer continuous keeled moulding, and a keeled hood-mould on two small busts. The capitals of the columns are decorated with the standing figures of a monk (with a scroll) and a bishop, with stiff-leaf behind them – the first figured capitals which we meet. The part of the arch which they carry has, to one’s surprise, a Norman zigzag, set at an angle of 45 degrees.

The doorway is preceded by a PORCH treated so sumptuously that we cannot be in any doubt where the principal entrance to the cathedral was meant to be. The porch has externally completely plain sides (except for the Perp parapet of the same design as everywhere else, see below) but a highly enriched facade and interior. Two flat nook-shafted buttresses flank the entrance. They end in polygonal pinnacles, again with blank arcading without any capitals. The doorway has eight orders of columns with shaft-rings plus two facing into the interior. The arrangement of the columns follows an odd rhythm. From outside, the wall first comes out to form a polygonal attached shaft, then curves back and allows space for two columns. Then it comes forward again, and so on. The rhythm is shaft-column-column-shaft-column-shaft-column-column-shaft, and then the columns at the opening proper. The capitals are purely stiff-leaf on the r., but full of figures on the l. (five capitals represent the Martyrdom of King Edmund). The arch mouldings repeat the same rhythm: keeled rolls over the coupled columns, a frieze of the most puzzling ornament over the single columns. It is unquestionably again a paraphrase of the Norman motif of the zigzag and corresponds to a certain degree with what was done at the same time at Glastonbury. Two crenellations with triangular merlons at an angle of 45 degrees to the surface meet in the middle a similar zigzag set at an angle of 90 degrees, that is pointing straight at us. Stiff-leaf sprays inhabit this spiky frieze. The hood-mould is a filigree band of thin stiff-leaf scrolls. In the spandrels are two oblong panels with reliefs: a man and a lion, and the gable of the porch contains a group of six stepped blank lancets with continuous moulding. The highest pair in the middle has a short shaft on a corbel. Below it, inside this pair of blank arches, a group of three small stepped lancet windows. Four bits of stiff-leaf and minor figure work above. The conception of all this is developed from the N rather than the S transept. The inside of the porch is a masterpiece of the E.E. style, of a richness which is at the same time orderly and measured. Two rib-vaulted bays; on detached double shafts in the corners, on triple shafts in the middles. The shafts have shaft-rings and stiff-leaf capitals. The walls are covered with tiers of arcading, first a tier of four blank pointed arches per bay with continuous moulding and spandrels filled with symmetrical stiff-leaf arrangements, then the sill-moulding with hanging sprays of stiff-leaf. The moulding is cut off at the ends of each bay by tailed monsters biting into it (cf. Elder Lady Chapel, Bristol). The upper blank arcades are almost in two layers in so far as they have their own supporting shafts close to the wall but are separated from one another by two detached shafts standing behind one another. So there is much depth in this blank arcade, but depth in front of a clearly maintained back surface. That is E.E. at its best and most English, in the same spirit as contemporary work at Lincoln, the cathedral which was built by St Hugh of Avallon, Bishop of Lincoln. The shafts of this arcade again all have shaft-rings and stiff-leaf capitals. Arch mouldings with keels and fillets. The outer moulding of each arch intersects a little with that of the next immediately above the capitals. In the spandrels rings of stiff-leaf and also figure motifs. The lunettes above have blank windows with Y-tracery un-cusped - and detached from the wall – once more the desire for a distinction between front layer and back layer. This must be one of the earliest appearances of the Y-motif anywhere. The vault makes no difference in girth or mouldings between transverse arch and diagonal rib. Fillet on the middle member of each rib and arch.

Now the WEST FRONT. If you approach it, as it is proposed here, immediately after the N porch, the impression will at once force itself upon you that here is the work of another designer, and what is more, not simply another man from the Wells lodge, trained in Wells traditions. Here the spirit is entirely different, different (if I may anticipate) from the whole interior too. The Wells style so far has been one of amplitude, of firmly rounded forms set against nobly sheer surfaces. Now we find something spare instead, harsh uprights and horizontals, angular gables, long Purbeck poles almost, in their detachment from the wall, like steel scaffolding. These are facts. What must now be added is more personal and less capable of proof. I would suggest that the Wells nave in its way and the Wells porch are masterworks of the highest order, but that the front is not. Mr Harvey calls it ‘the most nearly perfect of any among English cathedrals’. I confess I cannot subscribe to this, not even if it were said in a less challenging form. The W front of Wells, as it is, seems to me unsatisfactory. This is, it must at once be admitted, primarily due to the unmitigated contrast between the C13 substructure and the late C14 and early C15 towers. Before condemning the facade one should perhaps make an effort to imagine what towers the master who designed the C13 front had in his mind. For there can be no question that he visualized his facade completed. What then would his parchment drawing look like, if we were as lucky at Wells as they were at Cologne and Strassburg, when they found the original drawings? What height does this tremendous breadth demand which the master had chosen to give his front by placing his massive towers outside his aisles? To achieve a height commensurate to the existing breadth the towers would have to rise to something unprecedented in Gothic cathedrals, and incidentally something that would have dominated the crossing tower completely. On the other hand the designer might have aimed at a screen in the sense of Lincoln and Salisbury. But could he? Does not his uncompromising stress on the buttresses call for towers? Here, in this quandary, lies to my mind the failure of Wells. There was an English way of dealing with fronts, and there was a French way. The French place their towers in front of the aisles and thereby achieve a logical interrelation between interior and exterior. At the same time they are led to a prevalence of height, because the facades are relatively narrow. In England towers had already in Norman times been set outside aisles (St Paul’s, Colchester, etc.). The screen-facade is the consequence of that. At Wells the designer wanted to have the best of both worlds. Once again, I may slander him and he may have dreamed of towers of unheard-of height. Short of that, the facade could not have been made wholly successful, even if the designer of the S tower had been more considerate.

Now for the details, which are necessary in this case to an unusual degree, because apart from its architectural aspects the facade of Wells is the richest receptacle of C13 sculpture in England.⁶ This display of statues - nearly 400 have been counted as still existing - of course has also its bearing on the architectural argument. It makes one suspect that the master did not after all dream of crags but of an image screen, a reredos as never reredos had been seen before.

The C13 front is divided into five parts by six mighty buttresses. At the angles they connect with the equally broad and deep N and S buttresses by broad diagonals which actually hide the real angles of the towers. Horizontally there are five tiers below the zone, where the towers and the middle gable rise independently, the gable incidentally in three more tiers. Perhaps it is this grid of very strong verticals and very insistent horizontals which gives that odd feeling already voiced of some kind of steel scaffolding. The horizontal zones are a plinth, a tier of gabled blank arcades, two tiers linked vertically by the three tall nave W lancets, only slightly stepped, and by two blank lancets of equal height in the W faces of the aisles and between the tower buttresses. Their blankness and the utilitarian little windows in them are a serious blemish, as if pygmies had come to inhabit this mighty rock. The top tier is a frieze of low trefoiled niches. That is the general picture.

The base is the zone of the portals, although the middle one reaches up into the next zone. Nowhere else is the contrast between French and English more apparent than here, and nowhere else does England stand so defeated. Wells has its glorious porch. Why these niggly doorways, just too reminiscent of France to let one forget the wonders of Reims (or of Paris, to choose a more moderate front). The portals have Purbeck shafts, the nave doorway has a trumeau in addition. The stiff-leaf capitals are renewed. The arches are firmly moulded, and there are also inner continuous mouldings and hood-moulds on head-stops. In the tympanum of the middle doorway, in a deeply sunk quatrefoil, the seated Virgin with angels l. and r. The figures are badly damaged, but must have once been amongst the best at Wells. Now all the motifs so far met are motifs familiar from the work further E. But, as we go on to the next zone, we find two novelties: the use of so much Purbeck shafting, and the excessive use of gabled arcading.

We must now examine this first tier of blank arcading. It runs above the base all along the facade, arched and gabled, except immediately above the arch of the nave portal, where there is an incomprehensible little niche with a Coronation of the Virgin, absurdly small. The niche has a broad shouldered gable, or a trefoiled head with the middle lobe transformed into a gable. The two badly mutilated figures also must once have been enjoyable sculpture. The sharp long parallel folds have a good deal of vigour. Above the aisle entrances unfortunately instead of the gabled circles, there are windows – Perp now. As for the rest, there are trefoiled arches with their own little gables under each main gable. Under each sub-arch stood a statue on a bracket. Few survive on the W front itself. But as the towers project so far beyond the aisles the whole system of the facade is carried round them, and on the N tower, for some reason, plenty escaped the vandals of the C16 and C17, - fifteen in the row we are looking at now. They are, it has repeatedly been said by scholars, the best of the Wells statues. That is true, but it is not saying very much. One need not go to Reims, the obvious source of many, or to the Chartres porches, the source of others, to see that Wells is not and never was of that calibre. The angels of the Westminster transepts are sufficient proof of England’s capability, if not of reaching the highest mastery of France, at least of doing better than Wells. And Westminster, though the rebuilding here started more than fifty years after Wells, is hardly later in its sculpture. For the style of the Wells statuary in its relation to France makes it clear beyond doubt that this W front was not given its figures before c. 1235. The sources, to say it again, are Chartres at its most advanced (Ste Modeste, St Theodore, Visitation) and the W portals of Reims. Now at Reims the comparable figures (South portal, and Queen of Sheba and Solomon, also certain Kings under canopies high up, along the nave) were not carved or put up before c. 1230.⁷

So much for the statues. In the spandrels between the sub-arches are sunk quatrefoils, and these were filled with three-quarter figures of angels. Again, what remains of them is good sculpture. In the spandrels between two main arches is another series of such quatrefoils, but here the master chose stories told in detached, rather than single, figures, stories again told on too small a scale. They are from the Old Testament on the S, from the New Testament on the N side (e.g., from l. to r. , the Transfiguration, Christ among the Doctors, the Creation of Adam, Adam and Eve, Adam delving and Eve spinning, the Building of the Ark). The quatrefoils incidentally, instead of stopping at the corners, break round them which, as they are sunk, results in strange dents in the outline. It is worth speculating on this apparent solecism (which also occurs at Salisbury). It seems to imply a denial of the stony solidity of a front. What the designer has done was to him so much a screen, that like embossed cardboard or leather it could be pressed round corners.

Then the window zone. There is enough space between the three nave lancets for two tiers of canopied niches with figures. Excessively thin Purbeck shafts run up l. and r. of them. The figures are of normal proportions below, of exaggerated length above. If this proportion was chosen because the figures would always be seen from below, as has been suggested, that device has certainly not come off and they look as weird and gaunt as if they stood right in front of us. Again, who could deny it, although they are also among the best at Wells, there is not one amongst them that could move us like a figure at Chartres or indeed at Bamberg or Strassburg. The blank windows in the aisle parts and the outer parts are also flanked by the Purbeck ‘stove-pipes’. They have here a vertical row of crockets behind - a motif first used at Lincoln.

So far nothing has been said about the buttresses. But at this stage they take over the chief display of sculpture. Corresponding to the lancets and blank lancets there are two tiers of statues, one on each tier against the front of the buttress and two against each side. At this height the destroyers could do less damage, and so here, though the figures are defaced, the intended impression of the statuary screen can still be obtained. The figures against the front are seated, those against the sides stand. Artistic inspiration seems to come from Reims and Chartres still. The niches are again trefoiled and gabled and flanked by Purbeck shafts with vertical stone crocketing behind them. The shafts carry arches high above the upper image niches and, where there is space for a pair of niches, the arches are treated as sub-arches only and gathered under a main arch (that is the Y-motif which we have seen in the N porch) with intersected mouldings (which again we have seen in the N porch).

Above these arches, which are once more gabled, runs a much less conspicuous zone, no more than a top frieze. This consists of low trefoiled niches with scenes from the Resurrection of the Dead in small figures. Much stiff-leaf in the spandrels.

The stepped gable above the nave is in three tiers, the lowest of cinque foiled arches, the rest of trefoiled arches, and the topmost with an elongated cusped oval niche in which Christ sat in Judgement (the lower half of the figure remains) and two trefoiled niches, no doubt for angels, l. and r. At the angles again quatrefoils breaking round. Pinnacles l. and r. of the gable, circular with conical roofs. The images of the Apostles in the niches of the second tier are not C13 work. They look c. 1400 and were probably carved when the S or the N tower were built. The odd C14 or C15 pinnacle set on the middle of the gable requires a further remark a little later.

Before leaving the W front, two more things must be said. One is that the restoration of 1872-3 has certainly added a hardness to what was somewhat wiry from the beginning, which perhaps warps our judgement. The other is that to visualize the facade as it was when completed about 1250 or 1260 it must be remembered what part colour played in it. Enough remains have been found to make it certain that the backgrounds of the niches were painted in strong colours and the figures as well, and we can assume from evidence in other places that gold was extensively used too. It is hard for us to imagine what that must have done to the facade and harder still to decide whether we would have liked it. It was without any doubt very much like a painted Gothic polyptych.

That was the end of the first period. There is one piece of transition to the second: the STAIRCASE TO THE CHAPTER HOUSE. The windows here have Geometrical tracery, two circles with sexfoils below one large circle with three small encircled sexfoils. The spandrels are not yet pierced as in bar tracery, but a little sunk. The buttresses have the many steps in the set-offs already referred to. The date of this may be c. 1260 - c. 1275. At Salisbury bar tracery begins to be used about 1260. At that time the substructure of the chapter house must of course also have been under construction. However, the chapter house proper, as is proved by the details of the tracery of its windows, cannot have been completed until thirty or forty years later and indeed until the new Lady Chapel was completed E of the E end of the cathedral.

This was begun at an unknown date, at first detached from the old E end.⁸ There is reason to assume that it was complete by 1319. The very original design of the window tracery looks earlier than that; for it contains no ogee forms. The motif of the window tracery is arches upon arches - a kind of pre-reticulation - and a set of arches upon arches running up the main arch so that their apexes point concentrically to the group of the inner arches. The boldness of placing arches in another position than the upright should be fully appreciated, although there was of course the precedent of rose windows, or arched niches in the voussoirs of French cathedrals and also in the flying buttresses of Chartres. The parapet of the Lady Chapel has an openwork frieze of cusped triangles, and it may well be that this motif was created here, a motif which was then extended to the presbytery and its clerestory, the crossing tower, and more or less the whole cathedral, to the chancel of Bristol Cathedral, and in the end to a whole group of Perp Somerset churches.

The RETROCHOIR AND EAST TRANSEPTS, for reasons which the interior reveals, must have gone up at about the same time as the Lady Chapel. However, the window tracery is more varied and unquestionably later than that of the Lady Chapel. The E chapels of the retrochoir e.g. have reticulation, a mature ogee motif, and also a kind of straightened-out reticulation which, though in accordance with certain occasional and interspersed perversities of the early C14 at Bristol, seems to point, in this consistent presentation, to the proximity of the Perp style. The E transept E windows have a motif of intersected ogee arches, familiar from Bristol Cathedral, the end and W windows are again reticulated. The chancel aisles and transept E aisles carry on with intersected ogees or reticulation.

Ogee arches are frequent also in the splendid large windows of the CHAPTER HOUSE, but otherwise the tracery here is, if anything, a little closer to C13 traditions. A date c. 1310 would seem fitting. Workmen were obviously not the same as at the Lady Chapel. The windows are splendidly broad, of four lights, divided into two-plus-two. Each light has an ogee-head and a pointed trefoil above. Each two lights have a circle with an ogee sexfoil, each four lights a large circle again with an ogee sexfoil. Buttresses with gables decorated with ball-flower. Big gargoyles. Top frieze of sunk blank arches of two lights with Y-tracery, all in continuous mouldings. Then a later pierced parapet with diagonally set pointed quatrefoils and diagonally set pinnacles.

At the E end of the cathedral work on the new retrochoir was continued at the same time. The clerestory is of the same stage of tracery design as the lower parts, but then, when we come to the W bays (i.e. the C13 bays remodelled), curiously bleak, as if money or zest had given out. The most extraordinary piece of the design is the E end of the CHANCEL. The great seven-light window here is undoubtedly close to the Perp style and cannot be earlier than 1335 or 1340. The main division is into two-plus-three-plus-two lights. It is not done by sub-arches but by running the two main mullions right up into the main arch - a hallmark of the Perp. The three-light middle part then has again the new Perp type of panelling, the side parts a painfully lopsided arrangement of two ogee-lights with one reticular unit above the one foiled circle squeezed next to it towards the middle and crying out aloud for a companion. This wilfulness is Dec, in the Bristol spirit, but the rest, to say it again, must be called Perp. The gable over the window is as curious. It has what normally would be two two-light windows with Y-tracery and an almond-shape above, subdivided into four almond-shapes. But every curve is rigidly straightened out, and the whole made into a pattern of lozenges echoing the pitch of the gable.

The CROSSING TOWER belongs to the years 1315-22. However, its exterior was altered c. 1440.⁹ It is a noble design, calm and peaceful, and deeply satisfying from wherever it is seen. The C14 work stands on a short storey left of Jocelyn’s time - with the slim blank arcades with continuous moulding that his master mason liked so much. This low lantern of Jocelyn was raised in 1315-22 by a tall storey with, on each side, three pairs of very elongated lancets. The master of c. 1440 then filled in the lancets and decorated them in his own taste. The arrangement now appears of three two-light bell-openings with a transom, repeated below by blank two-light windows without transoms. There the proportions of the whole composition are wonderfully felicitous. Each light of the bell-openings is given a small foiled circle and on it a gable; the blank openings below have a short piece of quatrefoil frieze above each light. The buttressing etc. is brilliant. The clasping buttresses of the C14 are continued and end in pinnacles with their attached sub-pinnacles. Little niches for statues in the middle of each pinnacle. Between the buttresses are three panels separated again by (much finer) buttresses. They turn diagonal higher up and end in shorter pinnacles above the top parapet. This parapet has the same frieze of pierced cusped triangles as we have already found on the Lady Chapel. All buttresses have at half-height little ogee gables.

The WEST TOWERS were built later, the S one by the executors of Bishop Harewell after 1386 (by William Wynford), the N one by those of Bishop Bubwith after 1424. Both have the same design and are, like the crossing tower, the source of a group of Somerset towers. They lack the fine harmony with the previous work which the crossing tower achieves. They stand bold and bare on the mid-C13 front. Each C13 buttress is continued by one with two thin diagonal buttresses. In the N tower the two towards the W have richly canopied niches with images. These C14 or C15 main buttresses end with gables and pinnacles. At that point they throw back an inclined plane up to the bell-stage, like a huge final set-off. Out of that situation develop big diagonal buttresses divided into two. These are panelled and have thin diagonal shafts at the angles. The bell-openings are pairs of two lights, with diagonally set shafts between and with simple Perp tracery. The distinguishing feature is their very long blank continuation below, by descending mullions, as it were. The whole can be seen as huge, immensely long windows, left blank up to their transoms. As compared with the crossing tower, all uprights are stressed much more vigorously. The intended effect is unmitigated verticality. It was balance in the crossing tower. In the W towers it is therefore fatal that no crown whatever was put on. The whole slancio is abruptly cut short, and there is nothing but a low parapet with blank arcading and a silly frill of the smallest battlements. May it be suggested that the equally silly single pinnacle on the apex of the stepped gable of the nave was a trial pinnacle for one of the W towers? Pairs of them might have been set on the twin buttresses at the angles.

INTERIOR. The nave is in its interior completely of Jocelyn’s time, and it is rightly taken as E.E. Wells par excellence. But we do not know when it was begun, and whether it was begun to the plan of c. 1185-90. Once again, if we assume that work started at the E end and proceeded to the W, it is the CHANCEL that we must look to, to see what was planned at Wells under Reginald, a few years before the saintly Bishop Hugh started work on his chancel at Lincoln. Now of the late C12 chancel only the piers and arches survive. The rest was replaced or remodelled, and we have to consider the transepts at the same time to arrive at a composite picture of the original design. The situation in England was this. The Gothic style, as created in France about 1140, had reached England in a first fragmentary and later complete but much simplified form in the buildings of the Cistercian order about 1150-60. As a reflexion of this e.g. the pointed arches of c. 1160-70 at Malmesbury can be explained. Then the first complete building moulded on the pattern of the French Early Gothic cathedrals had gone up at Canterbury in 1175-85, directed at first by a Frenchman. Wells and Lincoln followed ten to fifteen years later. By then France had moved on, from the Early to the High Gothic. The great creative genius was the master of Chartres who began work in 1195. His design is characterized by far more determined verticalism than had been ventured upon before, by slender circular piers with four thin attached shafts, by vaulting shafts standing immediately on their capitals, by the substitution of a triforium for the gallery, and by oblong quadripartite vaults.

Now the master of Wells was in almost complete disagreement with these most recent French tendencies, the growth of which we can follow in France from cathedral to cathedral until Chartres found the final solution. His piers in the chancel are broad and complex in section, spreading generously rather than pulled together, and his vaulting shafts do not stand on their capitals. Horizontals are not sacrificed to verticals. The piers in particular are without doubt a demonstration – an anti-Canterbury demonstration soon to be an anti-Lincoln demonstration as well. They consist of a solid core of Greek cross shape and attached to it twenty-four shafts in groups of three in trefoil grouping, three to each main direction, three to each diagonal. The abaci are square over the diagonals, polygonal in the main directions. The arches have a variety of roll-mouldings, three for each diagonal, four under each arch, two plus two for the outer mouldings, and a hood-mould in addition (chipped off in the chancel, on head-stops in the transepts). The capitals in the chancel are the most telling feature of the earliest phase. They are of the stiff-leaf variety, but the leaves are small, in one or two rows, still rather like the crockets of French capitals. Only one at the NW end, by the crossing, is livelier and has a small head peering out. The aisle walls are divided by triple shafts as well. Above the arcade in an English cathedral of the C12 or C13 one would expect a gallery. At Wells there is a triforium instead. In the chancel it cannot be seen by the normal visitor, because the C14 has placed its curious grille of stone in front of it, but John Bilson, the most observant of English architectural historians of this century, found it under the aisle roof. It was a low and small triforium grouped in pairs and detailed with continuous mouldings, exactly, in point of fact, as it can still be seen on the E and w sides of the transepts.

So we can safely go to the TRANSEPTS for an impression of the whole of the design of Wells in c. 1185-1200. The only important difference between chancel and transept is a step forward in the development of stiff-leaf foliage. The treatment now is both bigger and freer. The leaves curl over more boldly, the motifs are larger, the carving deeper. One other difference should at once be noted - as it is also a difference within the transepts indicating probably the way they were built. The E aisle capitals are wholly foliage, the W aisle capitals introduce with some gusto a good many pretty figure motifs: a bald head, a man pulling a thorn out of his foot, fights, a grape harvest, a bridled woman, etc.¹⁰

Looking now at the triforium and trying to analyse the aesthetic difference it makes as compared with the galleries at Canterbury or Lincoln, we must first dispose of one point of historical interest. To have a triforium instead of a gallery may sound an extremely progressive touch, but in fact the triforium at Wells is more like that of the C11 at the Trinité at Caen than like that of Chartres. For at Caen and Wells it is primarily a horizontal band, at Chartres it is subordinated to the verticals. The Wells triforium in the transepts is grouped into two twin openings per bay, each bay separated from the next by the vaulting shafts, which start, not on the capitals below, but on corbels above the triforium sill. So the sill is drawn as one uninterrupted horizontal. The triforium openings have continuous mouldings (a roll and a chamfer), a favourite motif of Wells, as has been observed a propos of the exterior, but a motif which can be traced back to the Late Norman of e.g. Bristol. The corbels of the vaulting shafts incidentally again have foliage on the E side, seated figures on the W. The profile of the aisle vaults is three rolls arranged in a trefoil fashion, of the high vaults two rolls flanking a triangle or spur. The aisles have no bosses, the high vaults have bosses with soberly-treated stiff-leaf.¹¹

A fact worth much emphasis is the oblong shape of the high vaults. It is true that Durham had already had oblong vaults, but the Wells vaults are much more understood in the French Gothic way. They are the one really French Gothic motif at Wells. For the memorable fact remains that Wells, though in nothing else French, is yet wholly Gothic. At Canterbury (and at Chichester, where the Gothic retrochoir is exactly contemporary with the start at Wells) round arches are not eliminated. Nor are they at Glastonbury, a few miles from Wells. Wells is the first Gothic building in England in which the pointed arch is used exclusively.

The next detail to be observed refers to the end walls of the two transepts. They are identical inside, though, as we have seen, they differ outside. There are windows below, separated by triple shafts, and windows above with depressed two-centred rere-arches. But between these two zones the triforium runs here as a blank frieze of five arches, not divided into groups. This must be stressed; for it contributes the main innovation of the nave and the most memorable feature of the Wells interior.

In the NAVE for the first time one can sit down and absorb enough of the C13 to be able to decide what the master of Wells was after. To do that one must try to forget about what is without any doubt the most obtrusive motif at Wells, the grossly detailed strainer arches. They will have to receive their share of attention in due course, and they deserve it. Meanwhile the crossing ought to be visualized as it would be without them. And also, if I may digress, without the fan-vault in the tower. The crossing has its details of c. 1200. But they can now only be seen by ascending above the fan-vault: two tiers of arcading, both now blank, but the upper perhaps intended to let light in. They have shafts with moulded capitals, and the lower tier yet smaller blank arches inserted oddly in their lower halves. That view is now doubly barred.

To return to the nave, it is perhaps best to turn W to see it, although the interior design of the front, though less demonstratively different from that of the nave walls than are the strainer arches, is yet not in harmony with it either. Even so the purity of the nave will win; for here - this cannot be stressed enough - is a design of great intrinsic beauty and of supreme consistency preserved without any later interference, not even in the vaults.

The designer of the nave - he could be the man who had designed chancel and transepts, but at a more mature age - drew his conclusions from what has already been called Wellsian as against French: the stress on horizontals. The nave is ten bays long. Piers, aisle shafts, aisle vaults, high vaults are as in the transepts (except that the piers now have the subtle touch of keeling the shafts pointing to the diagonals and that the aisle vaults have small bosses of stiff-leaf).¹² But the great innovation is this. The grouping of the triforium in twins is given up, and it runs through from E to W as one seemingly interminable band. To achieve that, the designer pushed up the corbels for the vaulting shafts to just below clerestory level. That removes one more interference of the vertical with the horizontal.

These are the principal features. They had a second effect, besides horizontal stress, and one which is equally important. One feels sheltered between walls at Wells, and that also, it is my experience, irritates the Frenchman. Perhaps he is not wholly wrong, because horizontality and solidity of the walls are indeed presented with almost too blunt a directness at Wells. Lincoln also allows the horizontal its English place. But one does not feel horizontal strips, as one feels at Wells, even on the ground-level. For so massive are the piers at Wells and so finely subdivided that they also, in perspective, form one strip or band.

Some more differences between the nave and its aisles and the transepts and their aisles must now be pointed out, although they are subordinate to the ones described so far. The triforium has paterae of stiff-leaf immediately above each arch and in the spandrels between the arches. This and the bosses of the aisle vaults point in the direction of increased richness, and we shall see presently that within the nave such a development also took place.

Meanwhile the inside of the W front must first be looked at. It is clearly designed to a different scheme. The doorway has a depressed two-centred arch and is flanked by trefoiled blank arcading. Purbeck shafts are again used as they are so excessively outside. The aisle doorways place their depressed arches on vertical springers. The zone of the big W windows was reshaped in the Perp period. The horizontal ribs are patent, also above, where the Perp capitals carry E.E. arches with dog-tooth decoration. Stiff-leaf also reappears in the top spandrels.

As the towers stand out to the N and S, there are separate chapels under them, with eight ribs leading to a big circle (for the bell ropes). There is a certain restlessness in the design owing to the fact that the capitals of the shafts of the windows or blank arches sit up higher than the capitals of the vaulting shafts. In the S wall of the S tower, up in the lunette, is a prettily cusped stepped blank arcade such as are to be found also at the E end of Ely Cathedral.

In 1928 John Bilson published a paper on Wells in which, with his customary precision, he showed a break in style running through the nave along a joint clearly definable and, as building operations go, naturally lying further W on the ground floor than on the upper floors. It runs up six bays from the E below, five bays from the E above. Some of the differences are not easily visible: vertical instead of diagonal tooling, and larger ashlar blocks. But the abandonment of head-stops at the ends of the hood-moulds e.g. will be noticed by everybody.

The most telling change, however, is that in the character of the stiff-leaf capitals. Wells is the best place in England to enjoy and study stiff-leaf. Stiff-leaf foliage was an English speciality anyway; C13 capitals of stylized foliage as beautiful as those of Wells and Lincoln do not exist anywhere outside England. They make the crocket capitals of France look dull. Wells presents us with the whole gamut of stiff-leaf, from the timid beginnings of the chancel to the classicity, as it were, of the E bays of the nave and to the Baroque effusiveness of the W bays and the tower chapels. The capitals here assume indeed a lushness, a depth of carving, a fullness in their overhang which goes quite beyond what older men had attempted before. But what date does this end of the work represent? There is no certainty, whatever answers have been attempted. All that can be said has already been said, namely that the cathedral was consecrated in 1239 and that the sculpture of the w front was not begun before c. 1235-40. Is it not most probable then that the chancel is the work before Jocelyn, and that transepts, crossing and nave grew from about 1190-1200 to 1239? ¹³

As an appendix to the C13 work the UNDERCROFT of the chapter house must be mentioned. The doorway from the N chancel aisle into it has one order of Purbeck shafts, two capitals rather like those near the W end of the nave (one with birds and grapes), a depressed two-centred arch the inner mouldings of which die into the jambs, and a gable on large and excellent head-stops, and with bar tracery: a quatrefoil in a circle. This is the first bar tracery in the cathedral, and bar tracery was not introduced into England until 1245-50 (at Westminster Abbey). That provides a convenient terminus post quem. The passage to the undercroft is low, but impressively treated, with wall-shafts, stiff-leaf capitals which look earlier than those of the doorway, rere-arches of the windows dying into the jambs, and with transverse arches, ribs, and a ridge-rib, again the earliest in the cathedral. Ridge-ribs are a Lincoln innovation, but had by 1260 also been used at Westminster, at Worcester, at Ely, etc. The rolls of ribs and arches have fillets. Bosses with figure work. The undercroft itself has sturdy circular piers with plain circular moulded capitals and single-chamfered ribs. The middle pillar has eight attached shafts against an octagonal core with concave sides. The barred windows have rere-arches. The Perp STOUP or lavabo with a carved pet dog inside gnawing a bone should not be overlooked. How enviably free from considerations of propriety the Middle Ages were!

The STAIRCASE TO THE CHAPTER HOUSE, though not using bar tracery, is yet of the same date as the passage below. Plain Doorway from the transept E aisle with a depressed pointed head and continuous moulding. The staircase with its delightfully uneven steps is vaulted in two bays with ridge-ribs. The two S vaulting shafts are placed on two charming figures. The capitals now turn noticeably more naturalistic (all upright leaves, somewhat like the one transept capital renewed after the earthquake) - again a sign of the second half of the century. That goes with the tracery of the windows, as we have seen and can now see again; for the former N window is now above the doorway to the bridge which leads to the Vicars’ Hall. It has four lights with two sexfoiled circles and one large eight foiled one above. At Salisbury such kind of tracery appears in the 1260s (as bar tracery).

With the ENTRANCE TO THE CHAPTER HOUSE we are firmly in the C14. The walls are very thick and so the wall shafts and trumeau and tracery, all very light and transparent, are duplicated in two identical layers with a little vaulted space between. The W as well as the E entrance has above the two doorway arches a large spheric triangle. Large cusping; no finesses. The reveals or side walls of the little intermediate room have blank arcading with a curious top consisting of two sexfoiled circles side by side and two eight foiled almond-shapes above one another, the whole like a paraphrase on the theme of the elongated quatrefoil. The vault again has ridge-ribs.

The CHAPTER HOUSE is one of the most splendid examples in England of the style which might be called the tierceron style and which originated at Exeter in the last quarter of the C13. The first tiercerons were again used at Lincoln, but it is only where they are multiplied and three pairs spring from one springer that that palm tree effect results, which makes Exeter so memorable an experience. But the Wells chapter house goes beyond Exeter. From the central pier of the Westminster Chapter House spring sixteen ribs. The number was twenty at Lincoln. At Exeter, if one continued the scheme of the vaulting shafts full circle one would arrive at twenty-four. At Wells thirty-six ribs rise from the pier. Clusters of five wall-shafts support the vault from the corners. The walls have the usual seats under blank arcading below the windows, seven bays on each side. The cusping of the arches undulates. The gables rest on head-corbels and have crockets. Pointed trefoils in the spandrels and slim little buttresses and pinnacles between the gables. The whole panel of each of the seven bays is framed by a frieze of ball-flower. The inner window surrounds are also decorated with ball-flower. The vault has at its apex an octagon of ridge-ribs. This does not stand in line with the sides of the outer octagon, but side against corner, and corner against side. The same had already been done at Lincoln. The bosses are still fairly near to nature in their foliage.

The chapter house, as the tracery indicates, was probably designed in its upper details after the LADY CHAPEL, yet in its interior the Lady Chapel is so much part of the general rebuilding of the E end that the one cannot be described without the other. It is indeed their great fascination that they are so inseparable, and the way in which this great master has interpenetrated Lady Chapel and retrochoir is a feat of spatial imagination as great as that of the master of Bristol. What actually happens spatially at Wells has never yet been sufficiently analysed. The Lady Chapel at first sight seems easy enough to understand. It seems a room of one bay with an apse of three sides of an octagon. Trefoiled wall shafts with knobbly capitals, windows of that ingenious tracery which has already been described. Fleuron band below the windows. The S doorway and the sedilia are of excellent quality, with ogee arches, crockets and finials. It deserves very special mention that the sedilia have nodding or three-dimensional ogee arches; for unless the sedilia are an addition, made when the chapel was ready, they are amongst the earliest nodding ogee arches on record. But as soon as one looks back or up into the vault, one realizes that one is standing in a room meant to be read as an elongated octagon. The W sides of the octagon stand on two clustered shafts and merge into the retrochoir. The vault has the first liernes of Wells, liernes about contemporary with those of Bristol choir. They are made to appear a star within a star. From each shaft rise diagonal ribs and three pairs of tiercerons. The inner star is formed by the third tiercerons and not by liernes, but the outer consists of liernes. Bosses and sprays of knobbly foliage at the rib junctions. The only aesthetically unresolved problem is that of the difference in height between the Lady Chapel and the retrochoir. Here the master of Bristol, having a hall church to play in, was in an easier position. At Wells the three W arches are lower than the others and the space above them is disappointingly filled by blank arches.

The RETROCHOIR with its chapels l. and r. of the Lady Chapel and the E transept are all part of the same ingenious spatial conceit. Again all looks square and rectangular and normal from outside - like Salisbury e.g. But inside the sensitive visitor is at once thrown into a pleasing confusion. There are six clustered shafts in the retrochoir plus the two intermediate piers at the E end of the chancel itself. It takes time to realize why they are placed as they are. The two easternmost shafts we have already located as part of the Lady Chapel octagon. Then there are piers to mark the W ends of the two eastern side chapels, the chapels which flank the Lady Chapel. They naturally stand in axis with the chapel walls. Moreover they are in axis with the arcade between chancel and chancel aisles, and they form part of the initial rectangularity. But the next two clusters a little to the W of these are again in axis with the Lady Chapel W piers. That is unexpected, in fact unnecessary, but it is where the designer reveals his genius. For by this means, floating in the open space of the retrochoir, an elongated hexagon is formed - at r. angles to the elongated octagon of the Lady Chapel. That the two intermediate piers of the E arcade of the chancel are not placed so as to be in line with the W piers of the hexagon is an additional complication which, instead of heightening the aesthetic significance of the whole, only involved the designer in unnecessary difficulties as soon as he had to invent and set out vaults. It will be realized that in several places odd triangular spaces would have to be vaulted, and the master makes his appreciation of this known by giving the W pier of either E chapel a triangular shape. The Dec style always liked the diagonal and so also the triangle. Yet however one sorts it out, there was bound to remain Restrἅume, as Dehio called them when he wrote his classic analysis of Vierzehnheiligen.

It will no doubt by now be obvious to anyone who has tried to follow this description that English Dec space can be as intricate and as thrilling as German Rococo space. The vaults here are all of the new lierne kind except for the E chapels which are easily disposed of by diagonal ribs, ridge-ribs and one set of tiercerons. The E transepts, being square, could again receive a lierne star without much difficulty, though one should remember that liernes were still a very new toy. But the centre was the problem, and the combination here of lierne stars of various patterns with the triangles. How it is done cannot be described and only be drawn by the expert. But one should not shirk the effort of understanding it. It is like penetrating a piece of complicated polyphonic music. Nor was the master wholly successful. In one place indeed he has broken down, or someone took over who was incapable of understanding the original plan. From the massive NE and SE piers of the chancel rise, amongst other arches and ribs, three to the E which must have turned out to be so useless that lions are called in to bite them off. This is a ruthless procedure and one which the less naive C18 would no longer have allowed itself.

The CHANCEL AND CHANCEL AISLES offered less scope for spatial play. In the aisles the difference in the shafts (e.g. the use of Purbeck marble) and in the capitals from the work of c. 1200 is at once seen. The vaults are lierne stars of yet a different pattern (without any diagonal ribs or ridge-ribs), and they were substituted for the original, simple, quadripartite vaults in the W bays also. There are again no real bosses but rather sprays of foliage, where the ribs meet.

The chancel, especially in its upper parts, is the foremost piece of design in England of the few but interesting years between Bristol and Gloucester. If the Wells Lady Chapel (and no doubt the retrochoir) was complete by 1326, we have every reason to assume that the high walls and vaults of walls belong to the 1330s. They are therefore not earlier in date than the Gloucester S transept, though they are earlier in style. For while Gloucester, in spite of certain motifs and attitudes which are inspired by Bristol and must be called still Dec, is in its essentials Perp, the chancel at Wells, in spite of certain motifs which are in the Perp spirit, and of which some have already been commented on, is in its essentials Dec. This must be demonstrated.

The contrast to 1200 is of course evident everywhere: more finely divided piers, the principal shafts of Purbeck marble with foliage capitals, the less stressed parts without any capitals – a Bristol device. Thus a continuous moulding runs up all the way to the vault and frames arcade bay and clerestory bay together (shades of the Glastonbury transept?). But the memorable feature of the Wells chancel is the grille of delicate stone forms between arcade and clerestory, the kind of Vergitterung, as German art historians call it, which was the exaggeration of classic French mid-C13 ideas. The W front of Strassburg above the parts of 1275, etc., is the best known example. But no interior example is as intricate as this Wells stone filigree. Verticals are resolutely emphasized, and between them are three canopied niches in each bay with brackets for images, and two narrow spaces to the l. and r. The verticals are paired thin buttresses, again with narrow ogee niches between. The wall-passage in front of the clerestory windows breaks forward in a significant little triangle above every one of these buttresses: 1-2-2-2-1. Moreover, the wall-passage runs through the piers with diagonally set, ogee-headed entrances, and there are ogee-headed blank panels above them.

As far as possible the same grille was laid over the early C13 upper wall further W. That all is a little flatter could not be avoided. Yet an attempt is made even here to obtain a feeling of air between transparent front and solid back - by means of a canting forward of the narrow side pieces l. and r. of the outer buttresses.

So far the Dec elements only have been mentioned. But the panelling of a wall-space as such is a Perp idea, the favourite Perp idea, and the fact that the buttresses - glorified mullions - stand straight on the arcade arches below, is also Perp. In connexion with that the tracery of some of the windows must again be remembered with their unmistakable Perp motifs.

The vault is a tour de force. Though it keeps transverse arches, it has no intention any longer of stressing bays. There are e.g. no diagonal ribs, only diagonal ribs tying together two bays at a time. There are also no ridge-ribs, and instead a lierne pattern crystallized in squares at the ends of a saltire cross - a rectangular, wholly arbitrary pattern. The squares incidentally are cusped, an innovation at Wells, but in fact in imitation of the Bristol chancel (cf. also the Tewkesbury chancel vault of c. 1340). Again no bosses, but leaf sprays at the junctions.

The crossing tower of 1315-22 had to be strengthened very soon after its completion. The way in which this was done is sensational and incidentally further proof of the dependence of Wells on Bristol. One is almost tempted to suggest that to secure the tower an outside expert was called in and that he came from Bristol Cathedral. The strainer arches are huge, and they are - according to how one looks at them - two intersected ogee curves or an arch standing on its head on a normal arch. Exactly this is what some tracery and the sedilia had done at Bristol. Also the huge gaping eyes in the spandrels are the direct reflexion of the mouchettes in the spandrels of the aisle bridges at Bristol. The arches rise straight from the ground, without responds to carry them, and a thin triple-chamfer with quadrant moulding is carried on all up the crossing piers.

A tower can be shored up in other ways. That this is the way chosen here shows once again, with no punches pulled, what the Dec style is concerned with in England in matters of space. No smooth vista along the nave and up into crossing tower. Let it be filtered through these gargantuan meshes. The obstacle to the eye is worth more than the vista.

The Perp style added almost nothing to Wells except pieces of furnishing. The only architectural contribution is the fan-vault in the crossing tower which dates from c. 1480 (L. S. Colchester).

FURNISHINGS. LADY CHAPEL. STAINED GLASS.¹⁴ E window by Willement 1843 except for parts in the tracery which seem genuine. In the other windows a jumble of fragments, quite effective, including much original figure work. This must date from c. 1315-20. The SE window is the most complete. Most of the glass in the N windows is contemporary but comes from other parts of the church. Amongst details to be looked for is a panel with two of the Three Magi, a trumpeting angel, and several canopies. Also bases for figures; and it is said that this is the earliest occurrence of such bases in England. Another innovation is the use of silver stain which at York appears for the first time about 1310.

RETROCHOIR, EAST CHAPELS, EAST TRANSEPT. STAINED GLASS. Of about the same date as in the Lady Chapel. In the SE Chapel S window heads of bishops. Above a Christ in Majesty. In the E window a Christ apparently from a Coronation of the Virgin. In the NE chapel N window more heads of bishops, and above another Christ in Majesty. The SE transept S window is from Rouen, early C16, and of the school of Arnold van Nijmegen. - In the NE transept glass of 1902 by Powell, designed by G. P. Hutchinson. - SCREENS. Fragments of a wooden screen, now N side of SE chapel. - Pretty if sentimental new screen S side of NE chapel, c. 1935 by Sir Ninian Comper. - Elizabethan Screen, NE transept, with unfluted columns. - SCULPTURE. C15 stone relief above the altar in the NE transept: Ascension of Christ. Only his feet are visible in the clouds. - MONUMENTS. SE chapel. So-called Bishop Bitton d. 1264. Perp tomb-chest. Canopied niche to the E. Three canopies with steep, concave-sided gables with tracery. Panelled vault inside. - SE transept. Bishop Drokensford d, 1329. Tomb-chest, with low ogee-headed arches. – Dean Gunthorpe d. 1498. Big tomb-chest without effigy. – Brass Plate to Humphrey Willin d. 1618. Probably by Robert Haydock, Fellow of New College Oxford. Leaning figure with a remarkable display of inscription. On the r. his hat, sword, violin, etc., on the l. the Armatura Dei. He looks up and says: ‘Da Me Domine’. Two cherubs answer: ‘Pletenti dabitur’ and ‘Vicisti recipe’. - NE transept. Dean Godelee d. 1333. Tomb-chest like Drokensford’s with low ogee-headed arches. Of the effigy no details are recognizable. - Bishop Creyghton d. 1672. Bulgy sarcophagus. Big recumbent alabaster effigy. - John Milton d. 1337, recumbent effigy. - Bishop Berkeley d. 1581. Tomb-chest with cusped circles and shields. No effigy.

SOUTH CHANCEL AISLE. STALL-BACKS. Only the entrances into the chancel are of the C14; crocketed ogee arches flanked by buttresses. - STAINED GLASS. In the tracery heads some of the most beautiful early C14 glass, e.g. a Christ Crucified, a Virgin accompanied by angels, and a St Michael. These figures have all the sophistication and fragility of the architecture of the same moment and also its leaning towards the excessive. In one window heraldic and figure panels of the C17. - MONUMENTS. The interesting monuments of Wells Cathedral are those of the Saxon bishops all made about 1220-30. Most of them differ in detail more than style. They are all firmly modelled with deep rounded regular folds in the draperies and a deep rounded treatment of the features as well. Yet by the canopies at their heads the original stone block is still felt, and much of the carving seems sunk rather than raised. Canopies vary from trefoiled pointed to a curious shape with lobed sides and straight head. The first from the W has wilder drapery than the others. But they are all work of one workshop, and a workshop traceable in several of the capitals as well. The first from the E is quite different, and must be a little later. The effigy is far more independent of the coffin-lid. He really seems to lie on it. The pillow instead of a canopy especially helps to create that verisimilitude. The figure now wears a low mitre, and there is a frieze of small stiff-leaf along the edge of the lid. This must be by the W front men. Other monuments in the S chancel aisle from the E: Bishop Bekynton, dedicated 1452, but made some years earlier. Cadaver below in the opened winding sheet; six low oddly bulbous shafts with little ogee gables and above them demi-figures of angels attached to them. The wings of the angels spread into fern-like leaves. Recumbent effigy on a slab carried by the six shafts. E wall behind the effigy, depressed arch with openwork tracery, demi-figures of angels as cusps, straight top. Intricate little three-bay vault with pendants. Round the tomb or chantry an IRON RAILING, sturdy and unrefined, with coarse little heads as occasional decoration. – Bishop Hervey d. 1894. By Thomas Brock 1897. - Bishop Bitton II, d. 1274. Incised slab, coffin-shaped, the figure under a trefoiled gable. It is the earliest incised slab in England (cf. Chelvey). - Bishop Harewell d. 1386. Alabaster effigy.

NORTH CHANCEL AISLE. MONUMENTS. Three more of the C13 series, two of the earlier ones, the third again of c. 1250. The earlier have a cinque foiled canopy, and a trefoiled pointed canopy on stiff-leaf corbels with two angels in the spandrels, the later again lies free of any canopy and wears the low mitre. - Bishop Ralph of Shrewsbury d. 1363. Good alabaster effigy. - STAINED GLASS. In the tracery again some original figures: St Michael, Christ Crucified, St John the Baptist. - In the window E of the doorway to the Chapter-House Undercroft glass by Westlake, 1885.

CHAPTER HOUSE. DOORS. To the Undercroft; excellent with elaborately scrolled ironwork of the C13. - In the Undercroft COFFIN LIDS with foliated crosses. - STAINED GLASS. Several roundels with early C14 glass.

CHANCEL. MISERICORDS of the STALLS.¹⁵ They must be of c. 1340. They are among the best in England, carved with a delicacy as if they were for a less menial purpose. Amongst the subjects represented are mermaids, a pelican, Alexander carried up by griffins, a hawk and a rabbit, a lion and a griffin, a monkey and an owl, a horse and a wyvern, a griffin and a lamb, a puppy and a kitten, a cat playing the fiddle. – STAINED GLASS. The E window with the Tree of Jesse is one of the best monuments of mid-C14 glass painting in England. The date is probably c. 1340. Large standing figures; much yellow, much green, hardly any blue. - Clerestory. The two NE and the two SE windows have large impressive figures of saints, also original. - In the tracery heads on both sides small figures belonging to the Resurrection. W of the old windows stained glass on the S side by Willement 1846, on the N side by Bell of Bristol 1851. - WALL HANGINGS. Wool embroidery, made about 1930-40 by local ladies. Designed by Lady Hylton after consultation with Sir Charles Nicholson.

CROSSING. CHANCEL. SCREEN or Pulpitum. Made as good as new by Salvin in 1848. He also moved the whole centre forward. But basically the pulpitum must be work of the time of the completion of the chancel, and it is indeed still Dec rather than Perp. Stone, two-storeyed. Tall ogee-headed niches below with brackets for images; low ogee-headed niches above; battlements. The doorway is ogee-cusped and sub-cusped, with openwork cusping and encircled quatrefoils in the spandrels.

NORTH TRANSEPT. SCREENS, to the E aisle: stone, Perp. - To the chancel aisle: remarkable piece, presumably of the time when the chancel had just been finished. Ogee doorway with thin buttresses and pinnacles like that from the aisle into the chancel, but flanked by a vertical band of lozenges, consisting of four concave-sided arches in the four directions - a conception more in the Bristol than the Wells spirit. - DOOR to the chapter house staircase. C14, with tracery. - CLOCK. The famous Wells Clock, far more visited than the cathedral. Made c. 1390. There is no justification for the name Peter Lightfoot as that of its maker. On the dial the heavenly bodies are represented as they seem to move round the earth in twenty-four hours and thirty days. The small bell in the centre represents the earth (‘Sphericus architypus hic monstrat microcosmum’). The two larger circles to the l. and r. are the moon (l.) and her age (r.). The three outer circles of the dial show the day of the month, the minute, and the hour. Above the dial a procession of four small carved figures, on horseback. To the r. of the clock a seated figure known as Jack Blandifer. It may also date from c. 1390 and would then be the oldest clock jack in the country. On the outside of the transept wall a simpler dial and two ‘quarter-jacks’, figures in late C15 armour. - SCULPTURE. St Andrew, E aisle. Small figure above the entrance to the chapter house. Of a curiously wild style. Is it C17? And is it English? - Wise Virgin (?). Very good small figure by one of the W front masters. In the W aisle, N wall, over the entrance to the stair-turret. - STAINED GLASS. Clerestory E side. Decapitation of St John. Probably Arnold of Nijmegen; dated 1507. The inscription is in French. Renaissance details in the ornament. - N lancets. By Powell, 1903; highly praised at the time. - MONUMENTS. Bishop Cornish d. 1513. Tomb-chest with shields on cusped fields. Tudor arch, straight top. No effigy. Against the E wall mutilated figures of Christ and the kneeling Bishop. - Bishop Still d. 1607. Recumbent alabaster effigy flanked by two black columns. Shallow coffered back arch, spandrels with shields and ribbon work. - Bishop Kidder d. 1703. On the base semi-reclining figure of his daughter, rather daringly dressed and looking up to the two urns of her parents. Two columns, wide open segmental pediment. Back wall with a cloth on which three cherubs’ heads.

SOUTH TRANSEPT. FONT. Circular with blank arches and something unrecognizable in the spandrels. Dating is no longer possible. - FONT COVER. Naive Jacobean. - SCREENS. As in the N transept, but with C15 IRON GATES probably from the Bekynton Chantry. - STAINED GLASS. E side, clerestory, two Saints by W. R. Eginton 1813. - Two E windows by A. K. Nicholson. - MONUMENTS. Some of the most important ones in the cathedral. So-called Dean Husee (E aisle). Alabaster monument of c. 1400, from Nottingham, the tomb-chest with two panels of the most familiar themes: an Annunciation and a Trinity, but of an expressive power rarely achieved in English alabaster. Between the two panels three standing statuettes. Effigy also of good quality though no match for the two panels. Heavy canopy of stone; big superstructure with blank arcading. - William Byconyll d. 1448 (E aisle). Tomb-chest with ogee arcading. Back wall and coved ceiling with plain panelling. Heavy straight top. - Bishop William de Marchia d. 1302 (S wall). To the l. of the monument a separate chantry altar. The historical importance of the monument depends on whether one is entitled to assume that it was - as was usual - made shortly after his death. If so, it represents Wells at the beginning of the great work about the E end. The tomb is in a recess with a canopy. The effigy, of excellent workmanship, lies on a low base with a frieze of detached heads on it, a weird, as yet unexplored conceit. Three arches, ogee-cusped ogee gables with crockets and big finials. If this is indeed c. 1302-5, the ogees are the first at Wells. Vault with ridge-ribs and bosses. On the back wall three figures in bad condition. Two of them are angels. Against the E and W walls just one head each, of a grossly exaggerated size: one man and one woman. What can their meaning be? (cf. frieze above the reredos of the Lady Chapel, Bristol Cathedral). The touch of the sensational that turns up so often in the style of the early C14 is certainly present. – The surround of the chantry altar is livelier. One arch only with pierced ogee cusping. Panelled spandrels, straight top. The panelling in the spandrels is so much a Perp motif that it makes one consider whether the altar was not set up a generation later. Back wall with three ogee-headed niches, the middle one wider. Openings like small windows towards the big S window behind. Brass plate to commemorate the burial here of the Countess de Lisle d. 1464.

NAVE AND AISLES. PULPIT. An extremely interesting stone pulpit is attached to the Sugar Chantry (see below) and can only be reached from its inside. Its interest lies in the fact that it is of solid, monumental and very plain Italian Renaissance forms, handled without any hesitation and without any hankering after prettiness, and yet is as early as the time of Bishop Knight who gave it, and whose arms appear on it. Bishop Knight died in 1547. He was an able politician and a valued adviser of Henry VIII. In 1527 he had been to Rome in connexion with the divorce case, and he had spent many years in the Netherlands. His pulpit is one of the earliest attempts in England at a serious understanding of the Renaissance, as early as Lacock Abbey and Old Somerset House. It is circular with broad projecting piers and an inscription still in black letter. - LECTERN. A magnificent piece, the gift of Bishop Creighton, 1660, and so similar to the lecterns at Canterbury and Lincoln that it is probably by the same master: William Burroughs of London. Big bulbous stem and reading desk with symmetrical foliage in the triangles between the two reading surfaces. - STAINED GLASS. W window. 1670, also given by Creighton. An important document more than an enjoyable work of art. The centre light by A. K. Nicholson 1931. - Three s aisle windows, two by Kempe 1900, the third by his successor Tower. - In the NE clerestory window C15 Annunciation in the tracery head. - MONUMENTS. The two only chantry chapels proper. Bishop Bubwith d. 1424 and Treasurer Sugar, 1489. Of identical shape, namely hexagonal. The choice of the shape dictated simply by the wish to gain more space between two piers of the nave arcade. So the sides were canted out. In the architectural details much difference. The E wall in the earlier with quatrefoil-panelled coving, in the later a fan-vault instead of coving and demi-figures of angels in the frieze. The tracery of the earlier contains an odd reminiscence of the C13, rounded trefoiled lights though with Perp panels above. In the later the windows are arranged in four lights with two two-light sub-arches. The doorways of the earlier four-centred, of the later ogee-headed. There seems little stylistic significance in these changes. They are no more than variations within the same style. - On the altar of the Bubwith chapel small alabaster panel of the Ascension. What else the nave and aisles may have contained, has all been cleared out by restorers.

- PLATE. Two Chalices and Patens and a Flagon 1573; two Patens and a Flagon 1667; Almsdish 1675; two Candlesticks given in 1789; two Candlesticks by John Schofield 1793; two Maces 1798 and 1823.

THE PRECINCT

CLOISTERS. The cloisters were rebuilt in the C15. Of Jocelyn’s cloisters only the outer walls, the plain doorway towards the Bishop’s Palace, and two small doorways in the E wall. E of the E bay the Lady Chapel of the C12, excavated in the C19 but covered up again. It was replaced by Bishop Stillington’s chapel (1477-88) who was buried in it. This had a nave and transepts and a chancel and fan-vaults, but it was destroyed in 1552. Only the stone panelling of its former W side against the cloister E wall survives. The cloister EAST WALK was built by the executors of Bishop Bubwith, i.e. c. 1425, etc. Six-light windows with transoms and two-centred heads, divided into three-light sub-arches. Much pretty cusping. The vaults start on solid springers like fan-vaults, with two nodding ogee panels. The pattern of the vault is built up with liernes and a square centre with concave sides. On the upper floor the LIBRARY, a splendid room originally 160 ft long and intended as the cathedral library from the beginning. It must be the largest C15 library in England. The windows are small and of two lights. The book presses and the fine panelling of c. 1690. The Perp Wooden SCREEN at the N end comes from the Vicars’ Close.

The SOUTH WALK was begun by the executors of Bishop Bekynton d. 1465 and continued gradually to c. 1508. The WEST WALK was done for Bekynton. The Singing School is over the W walk. The design was not changed, except in one detail in the vaulting. The S and W walks straighten the sides of the square centre of the lierne-vaults. The singing school and part now belonging to the Girls’ High School have the same small two-light windows as the Library. The roof has collar-beams on arched braces and one tier of wind-braces. Attached on the W side, i.e. towards the Cathedral Green, various appendages, one with the same blank-arcaded parapet as the W towers of the cathedral. This contains a straight, wide staircase which leads at the bottom through a narrow arch with transverse ribs to a room which was there before and indeed before the present cloister; for it had to the W a large two-centred arch the moulding of which appears to be early C14. To the W again and on its own the ruin of the QUERISTERS’ House, built by Bishop de Shrewsbury (d. 1363). It has a two-light unmistakably Perp N window under the gable, and that furnishes a date for the introduction of standard Perp forms to Wells.¹⁶

MONUMENTS. From NE to NW. Thomas Linley d. 1795, also his daughters Elizabeth Ann Sheridan and Mary Tickell, by Thomas King of Bath. Urn on a base with musical instruments. - Peter Davis d. 1749. By Benedict Bastard of Sherbome. Obelisk and standing in front of it the solitary and silly figure of a putto reversing a torch which looks like a cornucopia. - George Hooper d. 1727. By Samuel Tuffnel of Westminster. Standing wall-monument. Two attached columns with open scrolly pediment and achievement. Two putti standing outside the columns. - John Berkeley Burland d. 1804, by John Bacon jun. Relief medallion with the dying man held by a woman. - John Philips d. 1834. By Chantrey, 1837. Big, vacant seated figure. - Many tablets with urns.

¹ The date is earlier than that accepted hitherto on the strength of Dean Armitage Robinson’s research. Its justification will be contained in a forthcoming book by L. S. Colchester and John Harvey. Mr Colchester most generously made some of his results available to me in time for the necessary alterations to the above description. Mr Colchester is inclined to place the beginning of the new building as close to 1176 as possible; for the piers of the Canons’ Barn according to a document of the time of Reginald are of the same (unusual) Chilcote stone as occurs spasmodically in the transepts and the E bay of the nave which would therefore be of before 1191.

² In 1248 the ‘tholus’ of the building fell in an earthquake and did damage to battlements, capitals of columns, and the like. This most probably refers to the crossing tower, but remains obscure, for there is little of repaired damage that we can trace.

³ Mr Colchester draws attention to a document of 1306 referring to the great expenses incurred in the building of the Chapter House (propter magnos sumptus et expensas quos ipsi ...fecerunt circa constructionem ...capituli). However, this document does not necessarily imply completion.

⁴ Mr Colchester has refuted Dean Armitage Robinson’s argument in favour of a glazing of the Lady Chapel windows in 1302 or 1305. He does however draw attention to the re-burial of Bishop Bytton I in the Lady Chapel in 1319, and suggests convincingly that that date may mark its completion.

⁵ Mr Colchester quotes a document about the reparacia of the magnum campanile in 1356 and assigns the strainer arches to that date. On stylistic grounds this seems improbable – see the similarity to Bristol work of the 1330s at the latest.

⁶ A detailed Guide to the Statuary has been published by the Friends of Wells Cathedral.

⁷ Dr Anderson in a recent book has stressed and in my opinion over-stressed connexions with the portals of Notre Dame in Paris which may have been begun about 1210 but are probably a little later. The Chartres sculpture in question is of c. 1225-35, that at Reims of before 1243 (the date of the beginning of the Sainte-Chapelle in Paris) if not 1237 (the date of a consecration at Bamberg in Germany), but not much before.

⁸ Mr Colchester points out that above the roof of the present retrochoir there are toothed ends of former flying buttresses. They prove that, even if only for a short time, the Lady Chapel must have stood detached except probably for a temporary narthex.

⁹ Mr Colchester quotes a document referring to 1439.

¹⁰ The first pier of the N transept E aisle from the crossing has a peculiarity worth noting. The triple-shafts to the E are cut off, and there is a long stiff-leaf corbel instead, with a lizard on it - the forerunner of so many long foliage corbels in English cathedrals. The first capital from the crossing on the E side differs from all others and seems the one proof of renewal after the damage to the ‘tholus’ in 1248. It is similar to capitals on the chapter stair.

¹¹ In this connexion we must also remember the stiff-leaf of the doorway from the E bay of the S nave aisle to the cloister. Here also the capitals are still reminiscent of crockets.

¹² The last bay leans with a fifth rib against the W front.

¹³ For Mr Colchester’s argument in favour of work on the transepts and the E bay of the nave before 1191 see footnote 1.

¹⁴ On Stained Glass in Wells Cathedral see the excellent booklet brought out by the Friends of Wells Cathedral and written by Mr L. S. Colchester (2nd ed. 1956).

¹⁵ At the time of my visit photography wasn’t, for reasons that elude me, allowed in the Choir and Chancel.

¹⁶ Mr Colchester regards this window as a later alteration.

Flickr.

Pevsner: Exeter Cathedral

ARCHITECTURE. The English cathedral as a rule is an assortment of architectural styles. Salisbury and Exeter are the most notable exceptions. Both are of a piece throughout, the work of two generations in the case of Salisbury, of three in the case of Exeter; and in both cases the younger masters kept essentially to the plans of the older. Exeter was begun at the moment when Salisbury was complete, i.e. c. 1275. Salisbury had taken some fifty years to build, Exeter was to take eighty or ninety. The fact that the basic plan at Salisbury dates from 1220, at Exeter from 1275, makes all the difference to the style of the two buildings. The hallmark of Salisbury is concision, of Exeter abundance. Salisbury is economical in its means, Exeter profuse. The full impact of this contrast can be obtained only inside; for as to the exteriors, that of Salisbury is dominated by the one vital later addition to its structure, the C14 spire, while that of Exeter is oddly unimpressive throughout. It has no height, and the two massive towers are characteristic but hardly beautiful. They do not dominate the city on the hill as the towers of Lincoln or Durham do.

They are the earliest part of the present building, Norman obviously, but not exactly datable. The South-West English see was transferred from Crediton to Exeter in 1050, no doubt because the city afforded better protection against attack. The beginning of the Norman cathedral is attributed to Bishop William Warelwast, a nephew of the Conqueror. It had a main apse, in line with the present E transepts. This was polygonal outside and semicircular inside (cf. e.g. the chapel at the E end, Gloucester). It was no doubt flanked by subsidiary apses at the E end of chancel aisles (cf. Old Sarum, Hereford, etc.). Of the Norman nave flat buttresses survive in the present N wall, and also indications of the responds inside which suggest cross-ribbed vaults. Fragments of such ribs with the plainest roll-mouldings were indeed found after the air raid of 1942. They prove rib-vaults for the aisles. As to the nave we know nothing, except that also in 1942 circular scalloped capitals of impressive girth were found, belonging evidently to circular Norman piers such as are typical of the W country: Gloucester, Hereford, Tewkesbury, Oxford, etc. The Norman nave must have been as long as the present nave, as in 1280 a chapel was completed just N of the W end of the latter (see below). The unique feature of Norman Exeter, however, is the position of the TOWERS over the N and S transepts. Parallels known point to the ‘Burgundian’ orbit: Murbach in Alsace of C11, and the cathedrals of Lyons and Geneva, the choirs of the one built from 1174 onwards, of the other after 1185. Their elevations are severely unadorned and indeed as inaccessible as keeps. The N tower is the earlier. It has its lowest band of tall blank arcading only above the height of the Norman nave and chancel roofs. The S tower starts the first band lower down. Above the first (on the S second) stage of arcading is a frieze of intersected arches on the N, of circular windows on the S. Also the S arcading on this stage has zigzag decoration which was not common yet when the N tower had reached that height (introduced into English architecture c. 1110-20). The next stage blank arcading with zigzag N and S, the last stage with alternations of arched bell-openings and blank arcades. The battlements and square pinnacles are late C15. The original tops must be reconstructed with low pyramidal spires. Inside, the openings towards the crossing were of course low and round headed (cf. Crediton, etc.). They were opened to their present size c. 1285.

But before that time, i.e. between the completion of the Norman and the beginning of the C13 cathedral, additions were made to the S for the clergy of the cathedral - Exeter being an establishment of secular canons like Hereford, Chichester, Salisbury, London, etc., and not of Benedictine monks like Worcester, Winchester, Canterbury, Ely, Durham, etc. The lower parts of the CHAPTER HOUSE belong to the time of Bishop Bruere (or Brewer, 1224-44), as proved at once by its ‘wind-blown’ stiff-leaf capitals. The chapter house is a rectangle of four bays, on the E side (as usual) of the former cloisters. It is entered by a doorway with a finely moulded arch on shafts with Wind-blown stiff-leaf capitals. Inside, it has a ground floor stage, plain except for the main Purbeck marble vaulting shafts, groups of three, meant for transverse arches and diagonal ribs. At the sill-level of the next stage they are divided by shaft-rings. On this stage the wall is enriched by two blank arcades per bay, again with grouped shafts. The higher parts were re-done from 1413 onwards, the main shafts replaced by shafts of the standard A profile leading up to pretty niches for figure sculpture, not surviving, and then the timber braces supporting the tie-beams of the low-pitched charmingly decorated roof. The decoration is a rendering in two dimensions of small fan-vault effects. The roof belongs to the time of Bishop Bothe (1465-78). The E.E. windows were replaced by four-light (E end seven-light) Perp windows with standard tracery. In the E.E. blank arcades remains of wallpaintings: figures under ogee canopies. The CLOISTER originally also belonged to the early C13. Nothing survives except a doorway in the E bay of the S aisle, round headed of two orders outside, with a depressed pointed arch inside. The capitals are of the same wind-blown stiff-leaf type as in the chapter house.

So much for the cathedral before rebuilding began. The exact date is not recorded, but it must have been in the last years of Bishop Branscombe (or Bronescombe). The general shape of the building is broadly speaking this: nave of seven bays with aisles, nave transepts with square E chapels, straight headed chancel of one narrow bay (for the pulpitum or rood screen) and seven bays (four choir, three presbytery), chancel aisles, low and narrow E transepts N and S of the 4th of the seven bays of the chancel aisles, each E transept with two shallow E chapels, straight E ambulatory, Lady Chapel of three bays flanked by E chapels of the N and S chancel aisles. The ambulatory and Lady Chapel are much lower than the nave and chancel clerestories. The building is throughout of grey limestone, mostly from Beer. The name of the original designer, the Master of Exeter, is unknown, although some later master craftsmen are recorded by name. The principal building dates can be more precisely fixed than in any other English medieval cathedral, thanks to the survival of a unique number of original fabric rolls.

Bishop Branscombe died in 1280 and wished to be buried in the chapel S of the Lady Chapel. The chapel is then called ‘fere de novo constructa’. In 1279 money was spent for windows in one of the chapels on the E side of the SE transept (the chapel was destroyed in 1942), in 1280 for windows in the chapel immediate N of the W end of the nave. In 1285 carpenters worked in the chapel N of the Lady Chapel. At the same time windows were put in in the main transepts, remodelled to make them go with the new work. Carving and glazing in the upper parts of the Presbytery, i.e. the E bays of the chancel in 1301-02. The roofs of the E end leaded 1303. The bosses in the vault of the choir, i.e. the W parts of the chancel, carved in the same year. These vaults painted 1309. By then, after about thirty-five years the E parts of the new cathedral were complete, and the Norman E end around which the novum opus had gone on had disappeared. The choir stalls (see Misericords below) were put in from 1309 onwards. On the High Altar workmen were busy from 1300 to 1322. The Bishop’s Throne (see below) dates from 1313 to c. 1317, the chancel screen (see below) from 1317-26.

If 1275-1309 comprises everything as far W as the crossing, the first bay of the nave had its bosses painted in 1317. Then in 1327 work on the nave was resumed in earnest, and probably also the W front started. The figure in the W gable of the W facade was in position in 1342. For the screen in front of the W facade with its figures and embattled top exact dates are missing; for the completion of the nave also. It was all done under Bishop Grandisson (or Graunson, 1327-69). The nave bosses look c. 1340 in style, but apparently the novum opus of vaulting the nave began only in 1353. Maybe the bosses had been carved in the masons’ lodge ten years before, and the Black Death had interrupted the intended progress. The W screen also is probably work of the forties and fifties.

Stylistic evidence confirms the evidence of the rolls. In the piscina and sedilia of the LADY CHAPEL (and one boss of the chapel E of the S transept) stiff-leaf foliage is still used. Then we find at once everywhere bosses of enchanting realism in the rendering of foliage: vine, rose, hawthorn, oak, etc. Particularly elaborate the corbels which carry the vaulting shafts of the presbytery vault. The closest possible study of the Exeter bosses and corbels, preferably with glasses and a strong torch, will amply repay the effort.¹ The Lady Chapel has a vault with longitudinal and transverse ridge-ribs and two pairs of tierceron ribs in the N and S cells of each bay, one pair more than, for example, at Lincoln, a little earlier. The Lady Chapel is divided from the NE and SE chapels by quatrefoil piers. The tracery of the Lady Chapel, NE and SE chapels, chancel aisles, and chancel clerestory (four-light and five-light) is also essentially in advance of Lincoln or Westminster or Salisbury. Its elements are (apart from the usual trefoils, quatrefoils and sexfoils in circles which are the characteristic element of geometrical tracery) trefoils and quatrefoils un-encircled, pointed trefoils and quatrefoils, pointed cusping of the tops of the individual window-lights, intersected mullions, spheric triangles and quadrangles - all in inexhaustible variations, with only the one limitation that windows which correspond to each other (e.g. clerestory same bay N and S, or NE chapel E window and SE chapel E window) have the same patterns.² The elevation of the Lady Chapel and the chapels flanking it shows blank arcading (pointed trefoil heads) on the ground floor with pointed trefoils in the spandrels.

The presbytery set the pattern for the whole rest of the cathedral: piers of lozenge-shape with sixteen c1ose-set shafts, more than anywhere else in England, correspondingly multiple mouldings of the arches, vaulting-shafts on corbels starting immediately above the piers, but not connected with them, large clerestory windows and comparatively low broad vaults with one pair of tiercerons in the E and W cells of each bay, and three pairs (again more than anywhere before) in the N and S cells. The luxuriant palm-branch effect is unforgettable. In only one respect does the present appearance differ from that of c. 1300. There was apparently originally no triforium at all - a very unusual thing in England (but cf. Pershore) - and the triforium, a low trefoil-headed arcade of four arches per bay with a pierced quatrefoil parapet for the clerestory passage above, was put in only in 1316-18, when this change in design had already been carried out in the choir.

Another significant if slighter change of style occurred even before the presbytery was complete, i.e. c. 1302. A comparison between the N and S corbels of the presbytery will show it. Leaves begin to lose their freshness and tend to display a tendency to undulate or congeal into a stylized nobbliness. This was a typical development all over England. It marks the end of classic Gothic perfection and the coming of the sophistication of the Dec style.

But Exeter, at least in its architecture as against its fittings and furnishings, did not go far on the way into Dec vagaries. The style as set before 1300 remained valid even under Bishop Grandisson. Evolution can be seen only in such details as the first and only appearance of lierne ribs in the vault of the chapel just N of the W end, and in the window tracery, where now more complex motifs occur such as large spheric triangles filled with smaller ones, lozenges with two sides concave and two convex, circles with a wheel of curved ‘fish-bladders’ (‘mouchettes’) and various spoke motifs in circles. Ogee arches, the standard motif of the earlier C14 in most other work in England, are used only very sparingly. Where change of style is most marked is again in bosses and corbels. There were many different hands busy on them. But the best of the nave (towards the W end) are as characteristic of c. 1340 (cf. Ottery St Mary bosses) as the best of the chancel arches and chapels had been of c. 1280-90.

Of special architectural pieces added in the C14 the charming balconies on the W walls of the transepts must be mentioned, which here take the place of the triforium. They rest on little demi-vaults just like the covings of later Devon screens. A little later, though also under Bishop Grandisson, the N PORCH was added. Its front is late C15, but too much renewed to serve as evidence. The W FRONT of the cathedral is at first puzzling to the student. It consists of a facade in place of the Norman one and a screen with two tiers of figures in front of and structurally interconnected with it. The screen was, it seems, an afterthought, but one that came only a few years after the facade had been completed. The facade itself has a large central window (nine lights), steep lean-to aisles without windows but with rows of climbing slender blank arches. Aisle and nave tops are embattled, the battlements climbing up the aisle slopes and then running horizontally above the nave window. Above and behind these battlements is the steep-pitched nave roof with a window of spheric-triangular shape and mouchette motifs in the tracery. The screen is like a reredos crowned by elaborate pierced battlements. The lower tier of figures represents kings seated on pedestals which are enriched by demi-figures. The upper tier has figures, put in, according to their style, in the late C14 if not early C15. The figures in both tiers are under canopies with three-dimensional or nodding ogee arches. Their faces and the folds of their garments are characteristic of c. 1340-50 and very similar to some of the nave bosses: long bodies, tortuous and yet stiff attitudes. The central doorway with angels in very shallow relief in the spandrels. Inside this doorway and the N and S side doorways porches to fill the space between screen and W front proper. In addition there is off the central porch in the thickness of the wall Bishop Grandisson’s tiny funeral chapel, with a pointed barrel-vault carrying an unusually large boss. The arches on the W and E have no capitals between jambs and voussoirs, and there are charming leaf-scrolls and fleurons in the arch mouldings. Both motifs were later going to be much used all over Devon. The porch of the S doorway of the W front has specially attractive figure-sculpture on the walls: Nativity and Adoration of the Magi in the 1330-50 style (note ox and ass below the Nativity), the N doorway a fan-vault, a C19 alteration (see Britton’s engraving). The three inner doorways between porches and church have also no capitals between jambs and voussoirs. The same applies to a small door from the S aisle into the former cloister just E of the W front. Here again there is fleuron decoration.

But such symptoms of a style in advance of that adopted in 1275 are unimportant. The main design remained unaltered from 1275 to the death of Grandisson and the completion of the building. It was a very personal design, of much character. One would wish to be able to ascribe it to a man known by name and circumstances of life. The Master of Exeter felt driven to the nec plus ultra of an existing style, the style of 1250-60, rather than to the creation of a new style. Such artists exist at all times. In his vaults he uses more tiercerons than anybody before, in his piers more shafts. This multiplication makes for richness, and the Master handled it with a sense of luxuriance, the epitome of late C13 tendencies. His work makes even the Angel Choir at Lincoln appear restrained. But he did not abandon the accepted type of vault or type of pier. He did not conceive the lierne, an essential innovation over and above the tierceron, nor the pier leading up without complete break into the vaulting zone, nor the pier wholly or partially without capitals, nor curvilinear tracery. His tracery is particularly personal. It is composed of motifs of immediate French derivation. The un-encircled trefoils and quatrefoils and the spheric triangles and quadrangles and elementary mouchettes in spandrels were the favourites of French designers about and after 1250 (Ste Chapelle, Paris; S transept, Notre Dame, Paris, etc.). They became the main stand-by in the cathedrals of the late C13 and early C14 (e.g. Nevers, Carcassonne, Bordeaux). The Master of Exeter must have seen them in France. There were not enough of them in England (e.g. spheric triangles, Westminster triforium, Hereford N transept, Lichfield). Yet he was emphatically an Englishman. The proportions of his nave are utterly un-French, generously broad, and the weight of the profuse palm-branch ribs from l. and r. makes the interior appear decidedly low. The short piers are also wholly against French C13 ideals, and the abundance of sculptured bosses has no parallel the other side of the Channel. Given these English characteristics of the building and its impressive scale and richness of architectural conception it is extremely surprising that, with the exception of Ottery St Mary and a few odd C14 windows here and there, it has virtually had no influence on the diocese, i.e. either Devon or Cornwall. The SW English parish church is of great consistency of style, but its style is not that of Exeter.

Additions after the death of Grandisson do not amount to much. The E window of the PRESBYTERY was replaced by one in the early Perp style in 1389. It is of nine lights and not specially original in its tracery. The renewal of the upper parts of the Chapter House has already been mentioned. The CLOISTER was rebuilt from 1377 onwards. Building went on into the C15. It was demolished during the Commonwealth, but its SE angle rebuilt in 1887. The details could largely be verified by surviving fragments. The transverse arches are panelled, the cells have lierne vaults starting in all four corners from mouchette-shaped comer-pieces. The bosses are original. The windows of four lights have standard tracery. The reconstruction is by Pearson, the C19 restoration of the cathedral by Scott 1870-7.

FURNISHINGS. ROOD SCREEN. 1320-4, called La Pulpyte, the Pulpitum, in the fabric rolls. The screen as well as all the other furnishings and fitments provided in the early C14 are characterized by a far more fanciful treatment than the masons allowed themselves in the architecture itself. The rood screen is of the ‘verandah type’, not just a partition wall or arcade. It consists of three depressed ogee arches on stone piers with moulded capitals, the spandrels above them decorated with fantastically curved elongated quatrefoils filled densely with nobbly foliage. At the outer angles niches placed diagonally, a typically Dec conceit. The parapet with cusped and crocketed blank ogee arcades, now filled with C17 paintings of little value. Elaborate ogee cresting. Fragments of a C17 timber gallery displayed in the nave aisles. Behind the three main arches little vaults. E wall with tracery panels not of specially interesting shapes.

OTHER SCREENS. The oldest is the SCREEN to ST EDMUND’S CHAPEL (N of W bay of nave), plain, solid timber construction, mullions only, no tracery, the only decoration a top band of pierced circles with a mouchette wheel - the motif taken over from the architecture of the W end. - Timber, otherwise, only the short PARCLOSE SCREENS between the chancel aisles and chancel in the bay which has also the E transepts. They are supposed to be late C14, and have four-light doors with standard Devon screen tracery, but the l. and r. sections of two lights only, one band of decoration above, and pierced cresting. - The other screens are of stone. The oldest is now fitted up against the N wall of the NE transept. It is not known where it was originally. It consists of three sections with steep crocketed gables, the sections separated by large projecting figures of angels above them. Between the gables two well-carved faces. The arches and the pierced trefoils in the gables all ogee-shaped: C14 certainly and probably of the time of Grandisson. It may have been a canopy of a tomb or a sedilia and not a screen. - The other screens are all of one and the same design with very minor variations: Screens to the transept E chapels (1433-4) and to the Lady Chapel and the chapels N and S of it (before 1419). They have very slim four-light sections with standard Devon tracery and entrances headed by a boldly cusped arch below the tracery of one four-light section.

SEDILIA. (1) In the LADY CHAPEL (cf. above). Three seats stepped up and two-light piscina with six-foiled circle in top. The sedilia with simple pointed trefoil heads. (2) In the PRESBYTERY, far more elaborate, and much restored, but still the most impressive proof of the development from 1275 to c. 1320 (unfortunately no exact date known). They possess all the intricacy and unrestrained spatial play which characterize Dec say at Ely or Bristol, but are absent in the major architecture of Exeter. The seats with tall, plain concave backs painted to represent cloth hung up. Tiny heads at the top of each of these curved back walls. Above the back walls the sedilia are open towards the aisle. Each seat is crowned by a seven-sided canopy, two sides for the back, one each for the l. and r. sides, and three for the front. The back and sides supported by shafts or outer walls, the front unsupported. Minute star-vault inside each canopy. The gables ogee-shaped and sumptuously crocketed. On top of the canopies another set of tall, fragile three-sided canopies, with the point of the triangle facing the aisle. From the aisle one sees the elaborate shaft-work, with unexpected shafts and gables peeping through a gossamer of finely carved forms. The Pointe of the composition is that the back has three and a half not three gables, so that every time a shaft between two gables corresponds to the centre of a gable in the front. It is more easily seen than described.

MINSTRELS’ GALLERY. The most popular fitment at Exeter, but neither architecturally nor sculpturally of a high order. It is placed high up in the bay of the nave which corresponds to the N porch, and takes the place of the triforium and the parapet above it. Its precise purpose and its date are unknown. It can, however, not be earlier than c. 1330 and looks in style rather 1350. Fourteen angels in tiny flattish recesses, all with different musical instruments.

CHOIR STALLS. 1309-10, probably by john of Glastonbury. Replaced by Gilbert Scott’s stalls in the 1870s. The original seats must have been older; for the MISERICORDS remain, the earliest set in England, dating from c. 1230-70 according to their style. Some have still stiff-leaf foliage; in others the leaves are naturalistic. Of figures, note, for example, the following: an elephant, a lion, a mermaid, a siren, a centaur, Aristotle saddled, ‘Lohengrin’, a knight fighting a bear.

BISHOP’S THRONE. 1313-c. 1317, the most exquisite piece of woodwork of its date in England and perhaps in Europe. Probably by Thomas of Winton (= Winchester) who appeared at Exeter in 1313 for a month and then returned versus patriam suam, where incidentally wooden choir stalls had been made c. 1290. The basic shape is square with large crocketed gables on the lowest and largest stage. The arches underneath the gables have cusped nodding ogee arches (fine carved human heads), amongst the earliest in the country. The upper stages recede and are fantastically pinnacled.

FONT. Tall, white marble, on pillar, 1692.

CLOCK. On N wall of N transept, large, square and originally of 1376. Not over-ornamented. Much restored and redecorated, notably in 1760. Much of the detail is Gothic of that period. The medieval works are now on show inside the Sylke Chapel. The clock shows hours, minutes, days, and phases of the moon.

SCULPTURE. St Ann, the Virgin and Child: Flemish group, C16, in Speke Chantry, N chancel aisle.

LECTERN. In the shape of an eagle; brass, probably late C15 and made in East Anglia. The figure is hollow and may have been used originally to collect Peter’s Pence.

WALL PAINTINGS. (1) Chapter House, see p. 133. (2) Above Sylke Chapel, N transept: Ascension of Christ, looks Flemish, late C15.

STAINED GLASS.³ In 1644 one Richard Symons noted down 150 coats of arms in windows. Less than twenty of these remain. In order of time the earliest surviving glass is the grisailles in the SE chapel and the NE chapel, very neat and delicate geometrical work. It dates from the first years of the C14, as borne out by the fabric rolls (glass bought 1303, and again, and this time from Rouen, 1317).⁴ The early windows have the same naturalistic foliage as the earliest bosses (oak, hawthorn, ivy); see the small fragments in the s choir aisle. Also early C14, but a little later, the clerestory window facing the Bishop’s Throne, also grisaille, but with a row of figures of saints (headless) under tall canopies. Similar figures and canopies in the six side-lights of the E window. Its centre lights renewed 1389 by Robert Lyen of Exeter. Lyen used yellow stain, still absent in the early C14, but his canopies are in imitation of the Dec predecessors. – William Peckitt of York, the most celebrated mid-Georgian glass painter, did much restoring in the 1760s and a new W window in 1766. Fragments of Peckitt’s predominantly yellow windows are now in the cloister. - E window, NE transept, by Sir Ninian Comper, 1948.

PLATE. Very damaged Chalice from the tomb of Bishop Bitton, 1307. - Four Chalices given 1628, two Flagons 1629. - Two fine fluted Candlesticks, 28 in. tall, 1693 (?). - The rest Georgian, notably Cradle Chalice, London, 1771, two square Almsdishes 1771, large oval Dish (German?) with inscription, 1772.

MONUMENTS. As there are not over-many of them at Exeter, a chronological arrangement may be preferable to a topographical.

1100-1400. Spurious monument to BISHOP LEOFRIC, d. 1072, N chancel aisle. Made up in 1568 and interesting for its date and the antiquarian attitude which its making implies. - BISHOP BARTHOLOMEUS ISCANUS, d. 1184 (?). SE recess Lady Chapel, coffin-shaped Purbeck marble slab, very flat relief obtained by carving away the background between the figure and the arcade in which he stands. Only slightly curved triangular arch, angels in the spandrels. The draperies still with the few ‘telescoped’ folds usual in France fifty years before. – SIDE SLAB or TOMB-CHEST, c. 1200. S side N chancel aisle, Purbeck marble, with three seated figures in quatrefoils. They are Christ in Majesty, St Peter and St Paul. - BISHOP MARSHALL, d. 1206. S side N chancel aisle. Tomb-chest with effigy on coffin-shaped slab. Much higher relief. The sides of the chest may not belong. They have also three seated figures in quatrefoils with small quatrefoils with heads (cf. Walter Monument, Canterbury, 1205) between the larger ones and a fine display of stiff-leaf sprays. - BISHOP SIMON OF APULIA, d. 1223. SW recess Lady Chapel; still coffin-shaped, same composition as Bartholomeus, but now with the higher relief, the easier attitude, and the finer folds of the garments characteristic of the E.E. style. - Another SIDE SLAB or TOMB-CHEST with cusped quatrefoils in circles, late C13, also S side N chancel aisle. - BISHOP BRANSCOMBE, d. 1280. In the interval between Lady Chapel and SE chapel, as he desired it in his will. Black basalt, of the most delicate workmanship. Still the same composition, though no longer of coffin-shape. Capitals and leaves of the arch of the arcade still pre-naturalistic. - The draperies with their fewer and heavier folds and the position of the hands should be compared with Simon of Apulia. Canopy, angels at the foot, etc., added 1442. - BISHOP QUIVIL, d. 1291. Slab with foliated cross on the floor in the Lady Chapel. - BISHOP STAPLEDON, d. 1326. S side N chancel aisle, i.e. close to the High Altar, erected during Stapledon’s time. The first of the canopied tombs, but only with a low canopy (was it originally higher?). To the chancel depressed, cusped, crocketed ogee arch, fleurons in jambs and voussoirs, to the aisle three trefoiled openings. Hovering angels in the frieze. The effigy recumbent and looking up at a contemporary painting in the roof of Christ displaying his wounds. C18 iron railings to the aisle. - SIR RICHARD DE STAPLEDON, d.. 1320 (?). Opposite the bishop, his brother, N side N chancel aisle: recess with slightly taller ogee arch. The knight cross-legged, and at his head and feet two enchanting figures of pages, the one holding a horse. Angel heads as cusps of the arch. If all this survived in its original state, we should know better how exquisite English carving of the time could be. - Two CROSS-LEGGED KNIGHTS in modern recesses in the S chancel aisle, also early C14 and supposed to be Sir Humphrey de Bohun, Earl of Hereford, and Sir Henry de Ralegh. - BISHOP BERKELEY, d. 1327 (?). S chancel aisle; Purbeck slab with circular matrix for brass demi-figure in a trefoil shape. – BISHOP GRANDISSON, d. 1369. See p. 138. - KNIGHT, later C14. Rubbed-off stone slab on the floor of the S chancel aisle. - SECOND EARL OF DEVON, d. 1377, and his Countess. S transept. Free standing tomb-chest with recumbent effigies. Originally in a chantry chapel near the E end of the nave, erected in 1381. The Countess’s badge, the Swans, interlocked at her feet.

1400-1550. SIR PETER COURTENAY, d. 1409. Brass at E end chancel aisle, nearly life-size, the thin architectural framework reconstruction. Very much like the Hawley brass at Dartmouth and another at Hurstmonceux, Sussex. - CANON LANGTON, d.1413. Brass in N chancel chapel; kneeling and with a scroll coming out from between his praying hands. - BISHOP STAFFORD, d. 1419. Between Lady Chapel and N chancel chapel. The earliest surviving monument of a type which became a pattern for the whole C15, and was, for example, copied in 1442 for Branscombe’s monument opposite. Tomb-chest with cusped quatrefoils, cusped Tudor arch, traceried spandrels, cornice with hovering angels, broad quatrefoil cresting. One composition with the stone parclose screen which here as well as opposite has contemporary paintings of saints (contemporary also the screens to the three chapels of the E end; see p. 140). - CADAVER (‘Gisant’) in recess with elaborate little, tipped-forward vault, N side, N chancel aisle; said to be Bishop Lacy’s brother, but probably a little later. - Another Cadaver in the front of the SYLKE CHANTRY, d. 1508, in the N transept. The inscription says: ‘Sum quod eris, fueram quod es, pro me, precor, ora.’ The chapel has no altar. It is surrounded by Perp stone screens with small figures against the uprights and crisp tracery. The wooden door comes from a screen in the chancel and is C14. - SPEKE AND OLDHAM CHANTRIES, built out to the N and S of the E ends of N and S chancel aisles. Sir John Speke died in 1518, Bishop Oldham (founder of the Manchester Grammar School and of Corpus Christi College, Oxford) in 1519. In design the two chapels are very similar outside, but quite different inside. The front walls of four sections with outer uprights decorated with small statuary. Broad parapet as horizontal top ending. Tracery and details of doorway as in contemporary stone screens. The large openings guarded by contemporary iron bars. Contemporary wooden door to the Oldham Chapel. Quatrefoil tomb-chests inside both chapels. Inside, the Speke Chapel has fully panelled walls, figures of angels in niches above, and a gently vaulted ceiling with pendants; not an inch left unadorned. The Oldham Chapel on the other hand has an unusually elaborate reredos with three figure scenes (Annunciation, Mass of St Gregory, Nativity), but less decoration otherwise.

1550-1860. Not one monument of the first rank. – A Harvey, d. 1564. N chancel aisle, the earliest with Italian motifs; balusters instead of Gothic uprights, Renaissance instead of Perp scrolls in the voussoirs of an otherwise still Perp depressed segmental arch. Still much quatrefoiling, and still block-letter inscription on the plain tomb-chest. - The following four are Later Elizabethan: Sir Peter Carew, d. 1589: S transept, epitaph with kneeling figure. - Matthew Godwin, d. 1586: N aisle; relief, representing an organist of Exeter cathedral who died aet. 18, an ‘adolescens’, kneeling in profile with musical instruments, organ, etc.; charming and full of feeling. - Carew family monument, 1589: NE chancel chapel; the most ambitious of the later monuments. Sir Gawen and his wife, recumbent behind low columns carrying an entablature with achievement. Below as in a truckle-bed their nephew Sir Peter - represented cross-legged! It must be one of the earliest instances in sculpture of a conscious medieval revival. - Sir John Gilbert, d. 1596, and wife: S transept; recumbent figures not behind columns but behind low arches (cf. Poltimore). - The following Jacobean and Carolean: Lady Dodderidge, d. 1614, Lady Chapel effigy propped up on elbow in C13 recess; a cartouche behind the effigy. - Sir John Dodderidge, d. 1623, recumbent, in robes of office, also Lady Chapel. - Then two very similar monuments of a usual type: Bishop Cotton, d. 1621, S chancel aisle, and Bishop Carey, d. 1626, N chancel aisle. Both alabaster, With bearded recumbent figures under shallow arches with columns l. and r. In the spandrels of Cotton, Fame and Father Time, of Carey simply putti. - Again a pair: John Bidgood, d. 1691, and Jacob Railard, d. 1692, both in E ambulatory, and both quite grand, sombre, black and gold epitaphs of a type not infrequent in Exeter parish churches. Quite a different mood: Benjamin Dillen, d. 1700, N transept: epitaph with portrait bust in medallion and a naval scene as a tail-piece. - BISHOP WESTON, d. 1742: at entry into S chancel aisle, with a dully carved angel blowing a trumpet and pointing at an inscription. - Finally C19: Flaxman’s General Simcoe, damaged in 1942 and to be re-erected in Canada. - DEAN PALMER, d. 1829, S transept, standing figure, with reversed torches, cold; by Hopper. - James Northcote, d. 1831: N transept, life-size seated marble figure by Chantrey. - Mrs Howe, d. 1834: S aisle by Gaffin of Regent Street. - Officers and men of the Ninth Lancers who fell in India, 1860, by Baron Marochetti: N aisle, very popular, with two relief bronze figures on horseback.

¹ They have all been photographed by the late C. J. P. Cave, and many of them will be included in a forthcoming King Penguin Book.

² In the Lady Chapel one small S door was added under Bishop Lacy, i.e. between 1420 and 1455.

³ Not restored to its original positions at the time of writing. Some of this glass was damaged during the Second World War.

⁴ In the E window of the NE chapel a C15 figure of a kneeling canon is inserted.

Flickr.