On the Crypts and Transepts

[Editors notes- To read a footnote, pull the cursor over it. Section titles and outline below added by editor- JV]

Outline of chapter- click on title to go to that section
A. Summary of Building Styles
B. The Building of the Transept

1. General Style
2. The Two Building Campaigns
3. The Crossing Piers
C . The Building of the Crypt


HAVING in the last chapter brought down the history of the building to the end of the twelfth century, it is time to compare it with the building itself. The present cathedral is in the form of a cross. The transepts, of a rude and plain Norman, manifest themselves at first sight as the oldest part of the edifice. A central tower, also Norman, stands upon four piers of great and unusual magnitude, and of singularly close jointed masonry. A superficial examination is sufficient to shew that this tower is of subsequent workmanship to the transepts. The eastern arm of the cross is of a mixed work, including portions of Early English, Decorated, and Perpendicular; the eastern extremity is bounded by a high gable, and beyond this is a low Early English structure, which consists of three aisles of nearly equal height, each of which terminates eastward in a chapel. The central one of these chapels, known as the Lady-chapel, is Early English in its western compartment, but has received an elongation of Perpendicular work, which is the most easterly part of the whole church. Beneath the eastern arm of the cross are crypts of the same rude Norman work as the transepts. They (like many others) serve to shew us the original plan of the Norman church, and it thus appears that its presbytery was terminated eastward by a round apse, at the point where now the flat Perpendicular gable stands.

The aisles of the Norman presbytery were continued round this apse, and a small round-ended (Lady-?)chapel extended as far as the western arch of the present one; also two small eastern towers flanked the apse of the presbytery. A crypt seems to have been also erected under the Early English Lady Chapel, but the crypt of the Perpendicular addition to this chapel is extended westward, and thus occupies a greater space in that direction than the Perpendicular compartment above it, by which means its vault intrudes upon and cuts off the original Early English vault, the remains of which, denuded of ribs, which now lie between the end of the Norman Lady Chapel crypt, and the later Perpendicular crypt, have induced some observers to believe that they had found traces of a very ancient building indeed.

The nave, or western arm of the cross, was originally Norman, and is now early Perpendicular; the transformation from one style to the other, for it was not rebuilt, presents one of the most curious studies imaginable. The present western termination is a flat front, but from foundations that still exist, and have been excavated and examined in the last year, it appears that the Norman cathedral extended about forty feet farther westward, and had enormous western towers. I will now proceed to examine the different parts, thus concisely indicated, in detail.

1. General Style

The general style of the transepts may be understood from the elevation, fig. 1. This represents the western wall of the south transept. The architecture is of the plainest description. The compartment of the triforium is very nearly of the same height as that of the pier-arches, and the clerestory is also nearly the same height. In this respect the distribution of Winchester resembles that of Ely, Peterborough, and Norwich, (Click here for illustration-JV) and the actual altitude of the walls in the three first examples are very nearly the same, being 75 ft., 72 ft., 76 ft. respectively, but Norwich is only 64 ft. At Winchester each pier-arch is formed of two orders or courses of voussoirs, the edges of which are left square, wholly undecorated by moldings. This is the case with the pier-arches of Ely transept, but in the arches of the triforium of Ely, and in every other Norman part of that cathedral and of the other two above cited, the edges of the voussoirs are richly molded. In Winchester transept, on the contrary, the arches of the triforium and clerestory are square edged like those of the pier-arches below, and hence arises the peculiarly simple and massive effect of this part of the church.footnote m

These transepts (as at Ely) have both eastern and western aisles. Peterborough has only eastern, and Norwich none at all. (Click here for illustration-JV) But Winchester, in addition, has also at each end of the transept an aisle, which rises only to the pier-arch level, and consists of two arches only, which rest in the middle on a triple bearing-shaft instead of the compound pier, which is employed throughout the rest of the work.

This kind of gallery is not unusual in the churches of Normandy, as I have already had occasion to remark in my History of Canterbury,footnote n where I have shewn that it was reserved for chapels, or for the preservation of relics of peculiar sanctity. At each end of the platform above the gallery at the level of the triforium, an engaged shaft, with pier edges behind it, springs from the face of the wall, but is carried up only a little above the capitals of the triforium arches, and there terminates abruptly, as at D in fig. 1, and at this part of the wall the masonry is disturbed as if something had once projected from the wall and has been removed. Whether there was once an arcade across the front of the gallery, or a beam of the nature of a roodbeam, or what other contrivance, it is impossible now to say. But it is clear that the shaft in question could not have been carried up to the roof like its neighbor at A, because the clerestory immediately over it has a pair of small open arches so placed as to make such a disposition impossible, as the figure clearly shews.

2. Two Building Campaigns

But the structure of these transepts shews that they were erected at two different periods, and that when the second erection took place changes were also introduced into the previous work. Each transept exactly resembles the other in these respects.

The vaults of the compartments C D E F G in the plan are plain groined vaults. But those of the eastern compartments, H,I, K, are ribbed; one of these compartments is shewn in fig. 2. The Decorated window is a manifest insertion of later times.

Now the general and original plan of the transept piers is that shewn in the darker tint of fig. 3, and the piers d and k of the general plan remain in this state. But the pier[s m and n].. received a subsequent addition to their southwest and south east angles respectively.  [This] changes its form to that represented by the [brown area] in fig. 3. Its ...    face now presents two engaged shafts of equal magnitude with a plain wall between, instead of one only as in the original plan. The evidences of this change are shewn by the joints of the masonry, and by the
capital and its abacus, in a way that when once pointed out cannot be mistaken. 
Also the half-pier on the opposite ... wall has been similarly augmented,[o and p] and the arch which rests upon these piers has been more than doubled in breadth. Fig. 7. [above] is a view of this arch and altered pier as seen by a spectator stand, [whose line of sight is shown by the brown arrow in the plan to its right].

In this view the original engaged shaft has a ring round it, the added shaft (nearest to the spectator) is plain; the piecing of the abacus is shewn, and the original capital of the small vaulting-shaft is seen peeping out between the two. That the vault of this compartment was erected before this change is plain, because it passes through and behind the new arch, as the figure shews, and the new pier also blocks up part of the arcade which decorates the wall below the window, and is crowded awkwardly against the window.

Now, in the compartment H, the same kind of addition has been made to the transept pier n and to the half pier of the wall p, and this half pier obstructs the window-arch of the eastern wall. But the vault of this compartment is ribbed, as before stated, and these ribs, instead of passing behind the additional part of the pier, are made to abut upon it in a way that clearly shews this ribbed vault to have been erected in connexion with, or at least subsequently to, the change. From this I infer, that the plain groined vaults are of the original structure, and that the additions to the piers, together with the ribbed vaults, belong to the second structure.

The motive for this additional strength given to the arches and piers at o and p will appear if we examine the outside of the building. There, from many symptoms, it is shewn that square towers were to have been erected on the compartments E and H, to flank the gable of the transept. Fig. 8. is a sketch of part of the north-east corner of the north transept. This shews an unfinished turret at the corner, the springing of an arcade above, with other marks that shew that the north wall of the side aisle was to have been carried up vertically as for the side of a tower. Moreover, the northernmost clerestory window is inserted under an arch which was meant to open into the tower.

Whether the tower ever existed and has been removed, or whether it was only projected and prepared for, I cannot tell. But traces of similar towers may be found more or less distinct at each angle of the transepts, namely, two towers flanking the north gable, and two the south. In the interior elevation of the transept (fig. 1.) the clerestory window at C is placed under a wider arch than the other. This is one of the arches which were to have opened into the tower. I should have stated that the nature and colour of the masonry outside shews two periods, the first of which terminates just below the billet-molded tablet in fig. 8, and therefore includes the pier-arches alone.

I conclude from all these appearances that the transepts were begun without towers, and that after they had risen to the height of the pier-arches the work was interrupted: When it was resumed, the towers formed a part of the new plan, and it became necessary to fortify the arches of the previous work to enable them to carry them. It must be observed that the eastern arch of the compartment E (in the general plan) is in its original form a strong arch, because it has to carry the entire height of the transept wall up to the clerestory. Therefore it needed no additional strength. But the southern arch of this compartment in the original plan was merely a transverse rib separating the vaults, and having no wall above it. This made it necessary to strengthen it when the tower was planned.

3.Crossing Piers

I must now explain another singular addition or change in the original plan. The portions of the transepts which I have just described, namely, the extreme northern and southern parts of the cross, are of rough masonry and workmanship, but not greater than that of early Norman work in general. But the four piers of the central tower, as well as the two piers contiguous to them in each transept, although of Norman work, are of singularly good masonry, and their form is more square and stronger than those of the other transept piers. These eight piers are distinguished from the others in the plan by a [brown]tint.

The arches also which are in connexion with these piers are of the same close-jointed masonry; and there can be no doubt that all this portion of the work has been rebuilt; thus confirming the account which history has given us, that the tower fell after the burial of William Rufus, and was rebuilt. Fig. 1. shews more clearly the line of demarcation between the two kinds of masonry, and therefore the exact portions that were rebuilt. The pier AB is of the older work,  EF of the newer work; GH is the tower-pier.

The arches of the clerestory, triforium, and lower range, between EF and AB,are all of the newer masonry, while the shafts that project from AB, and upon which these arches rest, are of the older style. This would necessarily be the case, for if the pier EF were rebuilt, the arches that rested upon it must also have been rebuilt. But the plan of the new pier is different from the others, and as we may suppose that the pier which occupied its place before the fall of the tower was like its neighboring pier in the transept, we are led to the conclusion that the plan of the new pier differs from that of the old. The elevation EF shews that this new pier is broader and simpler than the old one AB. But the change of form is shewn more distinctly in fig. 3. The dark shade is the old transept pier; the light shade on the right hand is the additional portion, or a plan of half the newer transept pier in question. It will be observed, that the pier in its new form is considerably stronger than before, although the same number of shafts are disposed upon the respective faces; that, in fact, the general arrangement of the parts of the old pier form a circular group, or rather a square, with the angles turned to the aisle; but that in the new form the parts group into a square, with its side turned to the aisle. The transverse measurement of the new pier is the same as before, but the longitudinal measure (that parallel to the aisles) is somewhat' increased, and as the whole mass of the pier is moved nearer to the tower, the arches between ABand EF, fig. 1, retain nearly the same span as before, but the arches betweenEFand GH are very considerably narrowed.

The tower-piers themselves must have undergone a considerable increase of dimensions. They are at present most unwieldy and intrusive, from their excessive size and awkward squareness of form, and are the largest tower-piers in England in proportion to the spans of the arches that rest on them.

Fig. 9. is a plan of the south-western pier, upon the same scale as the plans of the other piers. I have drawn upon this, in a darker tint, the plan that, in all probability, was that of the former pier which fell.

It will be seen that the newer pier is of a square and simple form, very unusual in Norman piers.

Figure 10. represents, upon the same scale, the south eastern pier, which, with some difference, is like the last. It may be worth while to remark the excessive change in the dimensions of piers that took place during the successive ages in which the present church was completed.

In figs. 3, 4, and 5, the transept pier is shewn in comparison with the piers of the choir and its eastern aisles; and it must be remembered, that the walls carried by them all are of the same altitude, but that increased mechanical experience had taught the builders to reduce both the thickness of their walls, and the size and number of their piers. In fig. 10. R shews the half pier or respond of the choir-arch, (namely, half of fig. 4.) T and S are the remaining fragments which may be seen of the original Norman respond, which occupies more than twice the space. The piers however present a greater contrast of magnitude than the walls they support, for the Norman wall is of the same thickness as its pier; but the Decorated wall of the choir is considerably thicker than the pier which it overhangs on each side. The pier is 3 ft. 4 In. diameter, and its wall 5 ft. 5 in., whereas the Norman wall is 6 ft. 2 in., the same as D 11, fig. 3, namely, that part of the pier upon which it stands.

This sketch, fig. 11, (taken however from the north-eastern pier,) shews the Decorated respond, and the remains of the Norman respond by the side of it.footnote q

The distance from the centre of one pier to that of the next is about the same in the Norman and in the subsequent work.

But to return to the tower piers. It is common in churches with a central tower to give less span to the arches that open north and south than to those which open towards the east and west. By so doing the piers are made greater in the north and south dimensions than in the east and west. The object is of course to keep the span of the east and west arches as great as possible, in order to leave the view from one end to the other of the church unobstructed. The transverse view from one transept to the other is of less consequence, especially in the early churches, in which the choir of the monks always occupied the central tower. The necessary strength is thus given to the piers by increasing their longitudinal dimensions at the expense of the transverse.

This artifice is carried to a much greater excess in some examples than in others. In Winchester to the greatest. But Gloucester and Hereford have very oblong piers. Peterborough too may be quoted, but is not completely in excess, although its squareness of form brings it into this class. In Norwich, Durham, St. Alban's, and St. Stephen at Caen, the shafts and pier-edges are managed so as to throw the tower pier into the mass of a square, or rather parallelogram, with its angles turned towards the aisles".footnote r

This diagonal position is, with very few exceptions, the general system of the piers in the succeeding periods, and it gives a free passage for light and for access between the piers. When the pier is in the square position, it resembles a portion of wall, and the pier-arch is reduced to a mere arch in that wall. But when the diagonal position is employed, the pier, however compound its arrangement, always resolves itself into one or more columns, upon which the arches rest, with greater lightness of effect, and with greater apparent span. In the tower of Winchester, from the excess to which the opposite system is carried out, the arches that open from the tower to the transepts are reduced to narrow arches in a very thick wall, and the interior of these transepts is quite hidden from the choir and presbytery, at a very small distance on each side of the centre of the tower.

There can be no doubt that these piers were erected under the influence of the panic caused by the fall of the tower, and that having no certain principles to guide them in determining the necessary dimensions for strength, the builders contented themselves by making the piers as large as the place would admit, sacrificing beauty and fitness to necessity. And this is really the history of all constructions. It is a great mistake to suppose that the architects of old were governed by scientific principles; practice and experience taught them the necessary proportions. They began by making their structures as strong as they could; if from bad workmanship and unequal settlements the building fell, they made it so Much the bigger next time. Finding it now too large they reduced the next building of the kind, and so on, by gradual experiments, were brought to proportions at once safe and beautiful. But all their works shew that they had no just conceptions of statical principles, and that they were guided by natural ingenuity alone, assisted by the numerous opportunities which the middle ages afforded for the erection of churches.

C. The Building of the Crypt

There is some difficulty in determining, from the historical documents that have reached us, the exact history of this early portion of the church. In the first chapter it has been shewn that Bishop Athelwold, about the year 980, effected a thorough repair, if not an entire rebuilding, of the church and monastery; that his successor added crypts, or rather, as I understand it, that he made more extensive crypts than had previously existed. And from the excessive praise and florid descriptions which are given to these works by contemporary writers, so exalted an idea of their magnitude and beauty has been conveyed, that many persons have thought it impossible that in less than a century Walkelin should have thought it necessary to demolish them entirely. Thus Milner imagines that the crypts of the present church are the work of Athelwold, " the walls, pillars, and groining of which remain in much the same state as he left them in ;"footnote s and also that the eastern arm of the cross of the Saxon church was allowed by Walkelin to remain, and was only pulled down when the present presbytery was erected.footnote t But, as Mr. Britton has well observedfootnote u, " it is a favourite maxim with some antiquaries to carry back the date of every church as far as possible, and like the late Mr. King and Mr. Carter, they do not hesitate to assert peremptorily that the oldest part must be of the age of the first foundation."

The argument from rudeness of workmanship, is best answered by comparing the transepts of Winchester with Norman buildings erected in places where no Saxon cathedral stood before, and where, therefore, there can be no supposition of the kind above stated. Thus the masonry of these transepts is not more rude in its tooling and the width of its joints, than that of Norwich cathedral, which see was removed from Thetford after the Conquest. Again, the architecture, in design as well as roughness of workmanship, of Winchester transept, is nearly identical even in dimensions with that of Ely transept, as they would naturally be, as the work of the brothers Walkelin and Simeon.

That many of the Saxon churches were erected of stone, and on plans of great complexity, with crypts, triforia, clerestories, central towers, and other parts resembling in arrangement the Norman churches, can hardly be doubted, from the descriptions that have been preserved to us. But that in dimensions and decoration they at all equaled the churches of their successors is wholly improbable. That contemporary writers should praise them as immeasurably lofty and spacious is natural, and in perfect accordance with the practice of all writers, who necessarily imagine the great works of their own age to be the greatest works possible, because they have never seen anything better or half so good. Perhaps the best testimony to their comparative merit is given by Bishop Wolstan, of Worcester.

He, after the Conquest, had erected a new church there, to replace the Saxon church which his, predecessor Oswald had built (c. 980). It was on a different site, and when the new church was sufficiently advanced to be occupied by the monks, he ordered the old church, the work of the blessed Oswald, to be unroofed and pulled down. But Wolstan, standing in the open air and looking on, could not restrain his tears at the sight, saying, " We wretched people destroy the works of the saints, that we may get praise for ourselves. That age of happy men knew not how to construct pompous edifices, but they knew well how, under such roofs as they had, to sacrifice themselves to God, and to set a good example. We, alas! strive that we may pile up stones, neglecting the while the cure of souls." footnote xWhatever allowance may be made for the Saxon feelings of Wolstan, himself a Saxon bishop amongst Normans, and therefore in all probability unfriendly to the new modes, yet the phraseology necessarily implies a strong and undeniable contrast between the Saxon and Norman practice and manner of building, and an apology for the inferiority of the former.

If Rudborne's history (or John of Exeter's) alone had been preserved to us, we should have been in greater difficulties, for there we find no other work of Walkelin expressly recorded but the tower and its piers. But the account given in the Annals is so clear that we cannot doubt that Walkelin rebuilt the church, or began it at least, and continued it so far that the monks were able to occupy it. It even appears that it was on a different site from the Saxon one, for the choir was ready for service before the old church was begun to be demolished; and although from the phrase that in one year the whole was destroyed save only one apse (or porticus) and the high altar, some have concluded that that apse was retained in the new building, it seems to me that the phrase that follows shews that the work of demolition went on, the old altar was destroyed, and probably with it the rest of the old building. That the position of the two buildings was different appears from the old tomb of Swithun, which the Saxon Wolstan says was on the west side of the church, but in Rudborne's days it was to be seen at the north door. This proves, at least, that the present church is much longer, and perhaps even that the present church stands farther south. This would be true if Wolstan's words be taken exactly to mean that the tomb stood opposite the west en ' d. And if we suppose the Saxon cathedral to have had a court in front of the west end, with a tower gateway of entrance, then Swithun's tomb would be exactly in the line of the passengers, so as to be trampled under foot, according to his desire. But it may be that the tomb stood north-west, which might be loosely described as west. As we know that the old high altar was in a different spot from the new one, and that the choir of the new church was built first, and occupied the same space as the present one, as shewn by the crypt, I incline to place the Saxon cathedral across the present north transept, which would thus require it to be pulled down to complete the latter.

The crypt of the present church cannot have been any part of the Saxon church, for the reasons above stated shew that the high altars were on different sites. Indeed the plan of the present crypt is in perfect accordance with that of Norman churches in general, and is of very great extent. The identity of the work of the crypt with that of the transept may be shewn by a peculiar abacus which is used in the crypt, and also in the column which stands at each end of each transept, bearing the gallery already described. Fig. 13. shews one of the capitals in the crypt, and fig. 14. that in the north transept. They are distinguished by an abacus so thin that it deserves the name better than any other example I have seen; and by the unusual combination (in Norman pillars) of a round capital with a square abacus, as in the Doric order.

That Rufus was buried in the central tower, and not in some other tower, as some have supposed, is plain, because one account says he 'was buried in the middle of the choir. As therefore the choir, like that of all Norman cathedrals in their original state, stands under the central tower, that must be the place of Rufus' burial. But the tower has plainly been rebuilt, and the account that it fell after Rufus' death, even if its fall had not been recorded in the Annals, is too well authenticated to be denied. Therefore we arrive at the conclusion that Walkelin built a tower on this site, together with the transepts, and that his tower fell after the death of Rufus, and was replaced by the existing one, but I am not sure that the curious discussion which Rudborne has quoted has given us the right conclusion, namely, that the present tower being built with Walkelin's funds, does therefore bear his name, for Walkelin may have been recorded as the builder of the tower that fell. Still it shews that doubts had arisen in the middle ages about the history of this tower. I have shewn that Walkelin's transepts went on slowly, and with changes of plan.

It is worth observing, in comparing Winchester and Ely, the contemporary works of the brothers Walkelin and Simeon, that they were both erected on different sites from their previous Saxon churches, and moreover that the central towers of both of them fell in after ages, Walkelin's in 1107, and Simeon's in 1321.

Click here to go to Chapter III