Part 16

But though in origin Winchester, Ely, and St. Alban’s hail from Caen, they all far surpass their model in vastness of scale. Winchester and Ely even committed the magnificent extravagance of having a western as well as an eastern aisle to the transepts; and though St. Alban’s, like Canterbury, had an aisleless transept, yet it surpassed them all in the stupendous length of its nave. Still more does it surpass the Caen church and Lanfranc’s Canterbury. Caen and Canterbury had nine bays in the nave, two in the choir, and two in each transept; St. Alban’s had thirteen in the nave, five in the choir, and three in each transept. Caen and Canterbury had naves 193 feet long, transepts 145 feet long, and a total length of 290 feet; St. Alban’s a nave 292 feet long (as long as the whole churches at Caen and Canterbury), transepts 177 feet long, and a total length of about 430 feet. In the fourteenth century, owing to the eastward extensions, St. Alban’s became longer still—520 feet inside, 550 feet outside. Winchester is even longer still: 2 yards longer than St. Alban’s; 3 yards longer than Ely; 4 yards longer than Canterbury.

I. EARLY NORMAN.—Of the work of Paul of Caen (1077-1093), we have the central tower and transept practically complete, and large portions of the nave and sanctuary. The design is strictly conditioned by the material, Roman brick; in consequence of the employment of which the architect was driven to rely for his effects, not on ornament or detail, but on what is nobler far, vastness of scale. It is worth while to compare the brick transept of St. Alban’s with the contemporary transept of Winchester, where the design is conditioned by the use of stone.

II. LATE NORMAN.—A fine arcade and doorway have been removed from the slype to the south wall of the south transept by Lord Grimthorpe. He is good enough to tell us that the new work which he has interpolated is so artful that no archæologist in future shall be able to ascertain which portions are old and which are new.

III. LANCET.—The vast and imposing church completed in the twelfth century found little favour in Gothic eyes. It was apparently regarded not only as ugly, but as incapable of being improved into something better. The only thing was to pull it down. In the end the monks did pull down and rebuild a large part of it, occupying at least a hundred years in the work. Fortunately for us, they wearied of their self-imposed task.

They began at the west end of the nave, so as not to interfere with the monastic services. First, John de Celia (1195-1214) proceeded to rebuild the west front on a lovely design, and with a wealth of costly marble and carving. Of this work there remain portions in the north-west and central porches of the façade.

His successor, William de Trumpington, was more economical, and produced more work. He completed the porches, and built in the western end of the nave four bays on the north and five bays on the south side. It can still be seen how he economised, not only on John de Celia’s design, in the arch that was to have led into a south-western tower, but on the design of his own bases, piers and vaulting-shafts; he even renounced the idea of vaulting his work at all—a very ungothic procedure.

IV. GEOMETRICAL.—His successors went further still; they abandoned all hope of rebuilding the rest of the nave, and turned their attention to the eastern limb. Retaining their stalls, as at present, in the crossing and the three western bays of the nave, they were able to go on with works to the east without any interruption to the services. The works dragged on very slowly. For some reason or other this wealthy abbey—with estates in almost every county of England—never had money to spend on its church.

First, the Norman sanctuary and apses were demolished, except that a length of Norman wall was retained on either side of the sanctuary. If this had been removed the tower might have collapsed. The chief feature of the new sanctuary is the clerestory windows, with tracery of early type, _c._ 1250.

Soon after, the lower parts of the Saint’s chapel and the ante-chapel were built; these were not completely finished till _c._ 1315. There, too, “vaulting ambition had o’erleapt itself”: the monks made preparations for vaults both within and without the ante-chapel, but in the end put up an unworthy flat ceiling.

About 1295 they commenced a new Lady chapel. This was not finished till about 1326. Thus it was built in the last years of the Geometrical and the early years of the Curvilinear style, and, like Selby choir, illustrates charmingly the transition from the one to the other.

V. CURVILINEAR.—The monks had now done all they meant to do. The eastern limb was rebuilt. They meant to do no more in the nave. But their hands were forced. Five Norman bays collapsed on the south side of the nave in 1323, and the monks had perforce to rebuild them. The design is closely assimilated to that of the Lancet bays to the west; the clerestory actually having lancet windows without tracery. To this period belong the shrines of St. Alban and St. Amphibalus, the Rood-screen, the doorway and arcade of the cloister.

VI. PERPENDICULAR.—The great gateway was built _c._ 1380. The nave and choir were better lighted on the north side by the insertion of windows in the triforium. The watching-loft was constructed _c._ 1420; the choir was ceiled, and the great reredos was erected _c._ 1480. To the same period belong the brass of Abbot de la Mare, _c._ 1390; and the grand chantries of Abbot Wheathampstead and Duke Humphrey of Gloucester, _c._ 1440.

VII. TUDOR.—The grandest chantry of all is that of Abbot Ramryge, _c._ 1520.

VIII. In recent times vast sums have been spent in underpinning and securing the walls and tower: and we have had the amazing west front and transept-ends designed by Lord Grimthorpe.

The “Annals of an Abbey,” in Froude’s “Short Studies,” should be read.

The Cathedral Church of St. Mary, Salisbury.

Two English cathedrals surpass all others in external effect: Lincoln and Salisbury; each of them being at its best as seen from the north-east. But Lincoln lacks “the quiet tranquillity of the close of Salisbury, the half-hidden houses, covered with vines and creepers, that nestle among the trees, the sense of being shut off from the work-a-day world. If Durham seems the petrified interpretation of the Church militant, Salisbury is the very type and picture of the Church of the Prince of Peace. Unworldliness and peace brood over church and close.”

The ancient Norman cathedral stood within the fortifications of Old Sarum. The site was cramped, and extension was impossible; there was far too much wind and far too little water; moreover the soldiers who garrisoned the castle were most objectionable neighbours. So, by permission of the Pope, Bishop Richard Poore in 1220 commenced a new cathedral on the present site.

It was a virgin site; and from this fact resulted a cathedral different from any other that we possess, and as a study in design more important than any other we possess. In other cathedrals we study the mediæval architect designing under difficulties; what we see in such a composite cathedral as Hereford or Chichester or Rochester is not one design, but a dozen designs trying to blend into one design; sometimes, as at Canterbury and Rochester, rather ineffectually, sometimes, as at Hereford, with remarkable success. At Salisbury it is not so: the design is one design, all sprang from a single brain, except the west front and possibly the upper part of tower and spire. We have no such homogeneous design in our mediæval cathedrals. The French were less conservative, perhaps less penurious; their Gothic architects were iconoclasts; no French architect could have allowed such frightful solecisms to remain as disfigure our cathedrals to the purist eye, and endear them to the artist.

Bishop Poore’s architect, Elias de Dereham, one of the canons of the cathedral, who a little later designed the eastern transept of Durham, had—what no other of our Gothic architects had—a free hand. Salisbury, then, tells us what no other cathedral does—what an English architect thought a cathedral ought to be like, when not hampered by having to preserve or assimilate pre-existing work or to build on pre-existing foundations. Canterbury nave is not the best the architect could do: it is the best he could do subject to the restriction that his nave must be neither longer nor broader than Lanfranc’s ancient nave. Elias de Dereham had no such restrictions; the cathedral was to be built in the green pastures, he could take as much space as he liked.

Begun in 1220, the whole cathedral, except west front, tower and spire, was complete in 1266, having cost, it is said, what is equivalent to half a million of our money. It has, therefore, the distinction among our cathedrals of having a design which is practically uniform. It is said, indeed, and with perfect truth, that the homogeneity and uniformity of the design of Salisbury makes it less interesting than the usual composite cathedral of England, with its design changing, as in Selby nave, almost in every bay. When we have seen one bay of Salisbury nave, we have seen the other nine, and we know that we shall find practically the same design in the choir. Yet we can afford very well, in England, to have one great design, the product not of a dozen minds, but of one mind; a work completed in less than half a century, and not spread out, like Canterbury, over four centuries, or like Hereford, over seven.

Again, building on a virgin site, Elias de Dereham did what all great builders have always wished to do—he made his building symmetrical. It is the fashion to contrast the symmetry of Greek with the picturesqueness of Gothic architecture. The comparison may be carried too far. The Erechtheium is as unsymmetrical and picturesque as any Gothic building. The great cathedrals of northern France, and Salisbury and Exeter, are quite classical in their symmetry. But they are not pedantic in their symmetry. Because nave balances choir, the north transept the south transept, and the north-east transept the south-east transept, the builder was not so foolish as to construct on the south a big porch to balance that on the north side of the nave. Instead of that, he built to the south an octagonal chapter-house, and this he placed unsymmetrically—_i.e._, picturesquely—because, so placed, it was more to the convenience of those who would have to use it. In other words, he did not purposefully aim at the picturesque and the irregular. Gothic cathedrals are picturesque, either of accident, as at Canterbury, owing to the casual collocation of work of different design in the course of several centuries, or because the different parts of the cathedral, being intended for different functions, have been designed different in plan, in dimensions, and in details. The latter is the case with Salisbury.

Two chief types of cathedral plan were at the architect’s disposal: what we may call the York type, and the Wells type. In the latter, the cathedral continues at its fullest height from the western doors to the far east of the sanctuary, then the retro-choir or ante-chapel (sometimes divided into Saint’s chapel and processional path or ambulatory) is roofed at a far lower height, and east of them is a Lady chapel, similarly on a low level. And, of course, as the upper wall of the east end of the sanctuary requires to be supported, arches have to be built beneath it opening from the sanctuary into the retro-choir. The plan is a beautiful one—it is our English equivalent for the French ‘chevet.’ Externally, the different portions of the eastern limb tell distinctly the purposes for which they are built: standing to the east of Winchester, or Wells, or Hereford, you say at once, “This building is the Lady chapel; there is the processional path and probably the shrine of the local saint, there is the east end of the sanctuary, and the light from that east window at the early morning services streams down on the high altar below.” Internally, too, the effect is delightful; the upper story of the cathedral is of course really greatly curtailed in length, but the cathedral is not shortened to the eye: the mysterious vistas through the east arches of the cathedral more than restore the height lost above; the glimpses of Lady chapel behind ambulatory, and ambulatory behind Saints’ chapel, as seen from far west in the nave, make the termination of the Lady chapel, invisible from many points of view, seem infinitely distant. Of these mysterious distances, shadowy recesses and changing vistas, there is nothing in Lincoln and York; the whole eastern limb is seen at a single glance, and, unfortunately, is foreshortened by the eye. Wells and Salisbury internally look longer than they are, Lincoln and York much shorter. In mediæval days, however, the ritualistic divisions of the church were marked off by a series of screens, each adding apparent length to the interior. So many of these screens, however, have been swept away by Wyatt, Scott and the like, or replaced by paltry open-work, as at Lichfield and Durham, that many of the cathedrals of this type are now mere open barns; all sense of mystery, all sense of magnitude gone. Externally, York and Lincoln and Beverley have the best of it. The sweep of a sky-line, five hundred feet in length, at a height so vast, is sublimely impressive. The cost of an exterior kept at an unbroken height for such a distance is very great, but it is worth the cost.

Cathedrals of the latter type, such as Ely, York, and Southwell, are quite presentable with a central tower of moderate height, provided that it is reinforced by more important towers to the west. An exterior of the former type, that of Wells and Salisbury, demands a very important central tower. Accordingly, every cathedral of this type has a big central tower—viz., St. Alban’s, Hereford, Wells; or a spire as well as a tower—viz., Chichester. (Winchester alone is an exception, an exception which proves how very much a central tower is needed in such elevations.) In such exteriors, looked at from the east, hills rise beyond hills, alps beyond alps, and the eye instinctively looks up to see the highest ranges aspire into the pyramidal outline of a Matterhorn. And this is what is given us at Salisbury.

It is not, however, what Elias de Dereham, the original architect, meant. No spire, or even upper tower, says Sir Christopher Wren, was originally contemplated. The unique slenderness of the piers of the crossing is certain evidence of that. The diameter of the piers on which the central tower of Canterbury rests is twelve feet; those at York and Winchester have a diameter of ten feet; those at Worcester nine feet; those of Peterborough tower (which collapsed) and of Salisbury, seven feet only. It is almost terrible to stand between these four thin piers of Salisbury and think how many hundred tons of stone in tower and spire above they have been made to bear. They were never meant to bear any such weight, especially planted as they are, like the whole cathedral, on the insecure foundation of a spongy bog. Indeed, not merely the tower and spire, but the whole cathedral, ought never to have been built where they are. Recklessness is by no means a strong enough word to use of the mediæval builders’ wanton carelessness about foundations. Peterborough cathedral was built practically without foundations on waterlogged peat; beneath the central tower of Carlisle cathedral were two running springs; Wells is reared on the boggy shores of a ring of pools. At Salisbury continuous bases had to be built from pier to pier.

Elias de Dereham’s design, no doubt, was to give us an exterior something like that of Beverley Minster or Westminster Abbey. The whole elevation of the cathedral, however, clamoured for a tall tower and spire. The original design was abandoned, and the foolhardy enterprise was taken in hand (_c._ 1330) of adding to the existing tower, which only just rose above the roofs, two more stages, and on these a spire, and that not of wood, but of stone. To abut this perilous steeple as much as possible, great flying-buttresses were added, both externally and running through the clerestory and triforium of the interior of the church. Moreover, to lessen the weight, the tower-walls were built in thin shells, while the spire is but nine inches thick. That the builders left their timber scaffolding in the spire, where it still is, to give its sides a little additional support, shows that they were alive themselves to the fragility and insecurity of their work. Later on, in the fifteenth century, stone girders were put across the piers of the central and eastern transepts, as at Canterbury and Wells, by way of struts, to keep the piers from bulging inwardly, though, as a matter of fact, what was wanted was rather ties to keep them from bulging outwardly.

Externally, however, the madness of this engineering feat does not trouble one. The addition of the tower and spire gives to the whole composition that pyramidal outline which always presents such a satisfactory appearance of stability to the eye—and in architectural design it is the eye that has to be reckoned with. In Wren’s masterpiece, St. Paul’s, one has the same central pyramidal outline of all the masses, but in a still higher degree.

In both cases, in St. Paul’s and in Salisbury, unity is secured. Salisbury spire is tall enough, St. Paul’s dome is tall and broad enough, to impose unity on all the diverging masses of the building. The low towers of Worcester and Hereford have no such supreme dominance.

Next to the abiding presence of the spire, unity is secured by scoring strong, horizontal lines round the building, welding its masses into one composition. Most cathedrals are contented, like Wells, with one strong horizontal line—a broad parapet. But at Salisbury there is not only a horizontal parapet, but a horizontal corbel-table as well; and there are no less than three horizontal strings—one running round the base of the walls, a second running along beneath the windows, and a third running round the buttresses; moreover, each of these horizontal lines is scored far more heavily than anywhere else. Especially remarkable is it to find the upper flow of the buttresses stopped by a heavy string. Usually, in Gothic, if a horizontal meets a vertical line, the former gives way; here, as in Greek architecture, the vertical gives way to the horizontal line. This architect saw, what few others of his day saw, that you may make too much of the “aspiring principle” of Gothic; that if you suppress the horizontal lines, you weaken the unity of the building, by failing to tie all its parts together.

As we have seen, the appearance of stability was enhanced by the pyramidal outline which the whole building ultimately assumed. But the eye instinctively looks downwards also to see that the pyramidal outline is continued there, it instinctively demands an emphatic spreading base. It likes to see a rock-like foundation. At Salisbury all below is but greensward. All the more carefully, therefore, has the builder spread out and broadened and emphasised his base-courses, till art gives the appearance of stability which nature has denied.

Stable, therefore, below and above, the exterior of Salisbury has much of that “monumental appearance”—of what Dr. Johnson called “stability and indeterminate duration”—which is the noble attribute of a great architecture, which one desiderates in a building “built not for time, but for eternity, whose walls will long be washed by the passing waves of humanity.” Such a building gravity and simplicity befit; its design should be solid and monumental, sober and restrained; it is not a field for the frippery of ornament, its best decoration is the stain of time. Built for eternity, it should suggest infinity—an infinite length which the eye cannot measure, an infinite height which the eye cannot estimate, a vastness of area that overpowers the imagination. Bigness counts for much in the painter’s work and the sculptor’s; it counts for yet more in architecture. A building in the grand style has to be big, and if it is a Gothic building it has to look bigger still. Salisbury is really big: it is 473 feet long, its spire is 404 feet high, its vaulted roof, 81 feet high, is the highest of any English cathedral. But the eye does not measure in feet. Salisbury spire and Louth spire, acute and slender, look hundreds of feet above their real height; Oxford and Chichester spires look lower than they really are. As for length, follow the wall of Salisbury from east to west. As you pass round cape and headland and promontory, you forget the point from which you started, or the goal for which you are bound. How foreshortened in comparison is the long façade of a Pitti Palace! As for area, in all honesty Mr. Whittington gravely declared that Salisbury was “a much larger church altogether than Amiens.” The error was natural. Amiens has fewer parts than Salisbury, and necessarily seems less. Going round Salisbury we pass no less than seven façades, at Amiens only three. There are thirty-nine bays in Amiens; in Salisbury there are sixty. As usual, multiplicity of parts has produced apparent increase of magnitude.

Much of the impressiveness of this exterior is due, not only to its grandeur of scale, both real and apparent, but to the lovely hue of the stone of which it is built, and the astonishing perfection of the masonry. At first sight, indeed, the appearance of the masonry of Salisbury is almost uncanny. Salisbury is not mouldered or corroded with age. “Time prints no wrinkles on its brow.” Its antiquity is that of a goddess ever young. The masonry, too, is that of a Greek temple; the precision that of the builders of the Parthenon; the joints fine, almost invisible; the blocks squared with mathematical precision; pass round the building from south transept eastward to north transept, and you will find that the stones in each course preserve their height with utmost exactness all these hundreds of feet. And so the building has the feeling of Greece in it. Gothic is a “small-stone” style, with joints openly displayed. Here, as in a Parthenon, the joints are invisible; the whole building seems one solid block—a monumental effect indeed. The crumbling masonry of Ely might belong to the ancient days of Saturn; Salisbury seems the work of a younger race of gods. Only when we scan its colour—the lovely colour of the Salisbury stone, that is seen nowhere else—“a pale, ashy grey, stained below with broad patches of red and yellow lichens,”—do we realise that this is no temple of yesterday, but one that has faced the stress of storm for more than six centuries. Her perpetual juvenility is at once the charm and the disappointment of Salisbury.

Very noteworthy, also, are the sobriety and restraint and repose of the whole design: very ungothic, too. At first it passes unnoticed; after a time it is noticed—and noticed with astonishment—that the beauty of a design of consummate loveliness is gained in some mysterious way without the use of ornament. One realises at Salisbury—perhaps for the first time—that ornament is non-essential even in Gothic design. What ornament there is is of the slightest—a floriated finial to a buttress, a trefoiled corbel-table; the design would be little the worse if even this trifling amount of decoration were omitted.