English Cathedrals

IN this handsome volume 1 Mrs. Van Rensselaer has collected and augmented, from the pages of The Century Magazine, a series of articles in which she describes twelve English cathedrals ; those, namely, of Canterbury, Peterborough, Durham, Salisbury, Lichfield, Lincoln, Ely, Wells, Winchester, Gloucester, York, and Loudon. They are chosen as typical examples illustrating the cathedral architecture of England from the Norman period to that of the Renaissance. The choice is well made ; and the reader who follows the author studiously to the end will hardly fail to gain a substantially true and serviceable understanding of the subject so far as concerns those general characteristics that appeal to the average intelligent unprofessional inquirer.

Mrs. Van Rensselaer evinces a genuine interest in her theme, and hence her book is well calculated to inspire a corresponding interest in the minds of others. In her Introduction she modestly states that she has written for amateurs, yet that she has tried to make a book that architects would willingly put into the hand of ignorance. That she has succeeded, for the most part admirably, in accomplishing her aim will, we believe, be generally conceded by architects as well as others. She has much felicity of expression in descriptive writing, — a felicity that is born of her own enthusiastic appreciation of the intrinsic merits of these ancient monuments, and the charm that is lent them by historic and poetic associations.

A good opening account of the conditions that prevailed in England at the time when the great impulse in church building set in — conditions which largely determined, as she shows, the peculiar character, situation, and general aspect of the English cathedral — is followed by an explanation of the collegiate and monastic chapters which formerly governed the cathedral churches, and how the suppression of the monastic establishments at the time of the Reformation led to the present distinction between cathedrals of the old and new foundations. The reasons why little pre-Norman work remains are stated ; and how all the successive phases of style, from Norman to Renaissance, are represented, often in the same building, is shown. Travelers do not generally take sufficient account of this. The fact that hardly any of the groat mediæval monuments of Europe exhibit the harmonious carrying out of a single original architectural scheme is too often overlooked. Hence these buildings are not seldom unjustly criticised. The uninstructed tourist, if he would rightly understand one of these great buildings, must learn to distinguish the parts of it which belong to the different periods of construction. It is true, as the author points out, that these successively constructed and often widely unlike portions sometimes group not ineffectively one with the other, and combine to produce a whole that, if not entirely coherent, has usually an historic charm which goes far to atone for the lack of artistic unity. Yet at the same time these different phases of design cannot, by a discerning and disciplined eye, be regarded with equal favor. They are the products, respectively, of different conditions, and they manifest the different degrees of constructive and artistic capacity that existed in England at different times. Mrs. Van Rensselaer is not always sufficiently discriminating. With a laudable desire to see good wherever it may exist, she is apt to fancy that she finds architectural merit where there is comparatively little of it; and what she apparently means for catholicity sometimes betrays, we think, a lack of sound judgment. This is shown in the latter part of the opening chapter, where, after admitting the superiority of the French Gothic in terms which logically imply (what we apprehend to be true, but what she is unwilling to acknowledge) that there is no true Gothic architecture in England, she endeavors, with evident embarrassment, to discover grounds for an equal admiration for English pointed design. She concludes thus: “ Most often we may feel that, whether French or English churches are the finer, it is well for us that French churches are tall and English ones are low ; that some were reared on narrow ancient streets, and others on broad verdurous lawns; that we have there the circling apse, with its arching chapels and its coronal of flying buttresses, and here the great flat eastern wall, — at Ely with its lancet groups, at Wells with its vista into lower further spaces, at Gloucester with its vast translucent tapestry of glass. Surely, the more variety the better, for us who have not to teach or to build, but only to enjoy.” We do not suppose that Mrs. Van Rensselaer means by this that the amateur should not endeavor to exercise discrimination ; yet such remarks savor strongly of the idea that in matters of art it is better to avoid a critical spirit. We do not share this idea. An open mind is of course essential; but without a critical (not a captious) spirit it is impossible that the merits of the best art should ever be apprehended. The amateur should not be counseled to avoid a critical temper, but rather to seek always a solid basis for critical judgments.

Errors arising from an imperfect acquaintance with the history of architectural developments and of structural principles are, unhappily, numerous in Mrs. Van Rensselaer’s work. On page 7, for instance, referring to the round arch and column, it is said that the Romans had used them side by side, but had never united them. The great arcade of the court of the palace of Diocletian at Spalato, and the basilica of Maxentius in the Roman Forum, however, bear witness that the Romans did unite these elements, though not at an early time of Roman building. On page 8 is the misstatement that “ by the end of the eleventh century all parts of great churches in Normandy were covered with vaults of stone.” There is, we believe, no evidence that vaults of stone were ever constructed over the naves of Norman churches during the eleventh century. The two great churches of Caen, the Abbaye-aux-Hommes and the Abbaye-aux-Dames, were vaulted for the first time early in the twelfth century. On page 12, the second transept of English churches is referred to as “ a feature never found except in England,” yet the great abbey church of Cluny was furnished with double transepts long before any church was built with them in England. On page 19, the lowness of an English cathedral is given as a reason why “ small service ” was required from the flying buttress. But the altitude of a building has little to do with its need of flying buttresses. The character of the structure alone determines that. In the Gothic system this member is essential to the stability of an edifice; in the English pointed system it is largely unnecessary, because the construction retains so much of the Romanesque character.

On page 60 we are told that in France the “ early Gothic followed immediately upon the perfected Romanesque.” But what is perfected Romanesque ? “ The novel constructional desires ” which, Mrs. Van Rensselaer, in this connection, tells us, “ preceded, predicted, and inspired the broad new ideal which was to realize itself in Gothic architecture,” did not arise as any sudden new departure. It was not as if the Romanesque had been perfected, and a desire for novelty then led the builders to cast about for some new constructional principles. Romanesque architecture, in its manifold varieties, exhibits a series of steps leading from the simple types of ancient design to the highly organized Gothic type. Gothic architecture itself is perfected Romanesque. On page 118, speaking of the moderate slant of the roof of Salisbury, the roof ridge is referred to as lying “ near to the vaulted ceiling. " This is impossible in a building with a clerestory, since the tie beams of the timber roof have to pass over the vaulting. On page 229, “the shafts which encircle the piers ” of Wells are pronounced to be “ more organically grouped than those of thoroughly English work.” But these shafts are arranged on precisely the same principle as are those of the most characteristically English pointed buildings ; that is to say, they are arranged in conformity with the arch orders of the ground story, and with these only, as at Lincoln (nave) and Salisbury.

The poorest part of the book is that, beginning on page 318, in which the writer attempts to explain the rationale of mediæval vaulting, and to describe its early progress. This section contains many erroneous statements. We are told, page 318, that " the earliest form of stone ceiling used by the Romanesque builders in the north of Europe was the barrel vault.” Now, southern Gaul, the only region in which barrel vaults were extensively used over naves by the Romanesque builders, can hardly be called the north of Europe. In northern Gaul, the barrel vault was never, so far as is known, made use of over naves. We are next informed that “ while church naves were still covered in this way [that is, with barrel vaults], the narrower, lower aisles were often covered with groined vaults.” It is true that the great abbey church of Cluny and some Romanesque churches of the ancient diocese of Macon were constructed in this irrational manner ; but the system was not employed in those northern regions where the Romanesque was quick with the germs of organic development. To oppose the concentrated thrusts of groin vaults in aisles to the continuous thrusts of barrel vaults over naves is an illogical proceeding of which the ingenious early architects of the north could hardly have been guilty. The logical form of aisle vaulting in connection with barrel-vaulted naves is the half barrel vault of the southern builders. The illogical association of the barrel vault and groined vault that is sometimes met with in southern Burgundy and its neighborhood seems to have been the result of the opposite influences, from north and south respectively, that were felt by the builders of this region.

On page 319 it is stated that “ these groined vaults had also been built by the Romans, although they preferred to cover square areas with domed ceilings.” Where and when did the Romans ever do such a thing? The building of domes over square areas was the great constructive innovation of the Byzantine architects, and was first perfected in Justinian’s great temple of Sta. Sophia at Constantinople. Roman domes were supported on circular drums, like that of the Pantheon. The most unmeaning misstatements occur in the second paragraph on page 319, and the first, paragraph on page 321. The author says, page 319 : “ As round arches which rise from the same level to the same height cannot vary in span, he [the Romanesque builder] could use groin vaults well only above square compartments ; over an oblong compartment he was obliged either conspicuously to stilt some of his arches, or to use for others a segmental form which meant both ugliness and constructional weakness, or to start different arches from different levels, which was not easily managed with current methods of design.” Here the writer does not see that to start different arches from different levels is neither more nor less than to stilt some of the arches. She then explains that the greater convenience of vaulting square spaces led to the necessity of having two smaller vault compartments in the aisles to one larger vault in the nave, — since the aisles are usually about half the width of the nave, — and says, “ This necessity is revealed by the alternation of form in the piers of the great arcade which we find in many late Norman and early Gothic churches.” The fact is that this alternation of supports occurs in the earliest of all vaulted Romanesque architecture of western Europe, namely the Lombard, and is shown in such buildings as S. Ambrogio of Milan and S. Michele of Pavia. It occurs also in early Norman design, as at Jumièges and the Abbaye-auxHommes at Caen.

Again referring to this alternate system, Mrs. Van Rensselaer says : “ Thus we have a clear instance of the way in which the character of the vault was expressed by the design of the church’s wall, the concentration of part of the thrust of the vaults breaking that uniform series of piers which we see, for instance, in the nave of Peterborough, and which was appropriate when a flat ceiling was used, or a barrel vault whose thrust was more equally distributed along the walls.” The uninformed reader would naturally infer from this (what the author certainly cannot mean to imply) that for a regular system of piers, like those of Peterborough, groined vaulting would be inappropriate. This regular system is, however, not only most appropriate for such vaulting, but it was not seldom used to carry it by the Romanesque builders of the twelfth century, as at Vezelay. The regular system merely requires, when groined in a logical manner, compartments of an oblong plan. As regards a flat ceiling of timber, the regular system is no more appropriate than the alternate system ; for a timber roof needs no piers or shafts whatever ; it rests - upon the walls. The presence of shafts rising from the pavement, in many early timber-roofed buildings, merely shows that they were derived from vaulted types of structure, and that they may at first have been intended to carry vaulting. As a fact, both the alternate and the regular systems were employed in many early structures which were roofed with timber,—the one, for instance, at Jumièges and the Abbaye-aux-Hommes, the other at Peterborough and the Abbaye-aux-Dames.

On page 320 it is said that “ before the middle of the eleventh century it was perceived in France that pointed arch forms would exert a much less powerful thrust, and would give the architect much greater freedom in design.” What is the evidence of this ? The earliest use of the pointed arch in connection with groined vaulting that we know of occurs in the apse of the abbey church of Morienval. The vaults of this apse are supposed by the most competent French authorities to have been constructed early in the twelfth century, though it is possible that they may date from the end of the eleventh. Such use as was made of the pointed arch in the supports of the dome of the church of St. Front of Pèrigueux, and in some of the barrel-vaulted edifices of southern France, supposing some of these last to date as early as the middle of the eleventh century, which is doubtful, has nothing whatever to do with the development of groined vaulting. On page 321 we read : “ Even with the pointed-arch forms the architect was not perfectly free to design as he chose ; he could not build arches of any span and height he might desire, and spring them all from the same level.” Why could he not? Is not this precisely what most of the so-called Gothic architects of England did ? In the genuine Gothic of France, indeed, the pointed arch could not be so used ; not, however, because of any inherent difficulty in thus adjusting one pointed arch to another, but because in this way it is impossible to secure that concentration of vault thrusts upon a narrow pier which, more than any other structural characteristic, distinguishes Gothic architecture from the pointed architecture of England which is not Gothic in principle. The author continues " But he could stilt vaulting-ribs without producing forms as disagreeable as those which result from the stilting of round arches ; and he soon discovered that he could spring them beautifully from different levels by allowing them to interpenetrate. That is, instead of carrying down all the ribs which met above his vaulting-shaft to the capital of this shaft, he could allow one to die into another at some distance above it ; the eye would fancy it continuing down behind its neighbors, and thus unity of design could be preserved with much freedom in constructional processes,” etc. This is incorrect. Ribs that interpenetrate do not, in early Gothic, die away above the shaft. Interpenetration merely diminishes the bulk of the group at the impost, so that a smaller capital, or group of capitals, may carry them. It has nothing whatever to do with the springing of arches from different levels. The clerestory arch, which must spring from a higher level than the other arches, or ribs, of the vault, is necessarily in the plane of the wall ; and it cannot, therefore, interpenetrate with the other ribs of the vaulting which are necessarily, except. at the springing, out of that plane.

But although these errors detract from the merits of the book, they do not outweigh them. The book contains, as we have said, a great deal of good material; and a corrected edition would, we think, be useful as a popular introduction to the study of English cathedrals.

The illustrations vary considerably in merit. The best are those in which open line work prevails, such as the Central Tower of Canterbury, page 46 ; Durham Cathedral from the Southeast, page 91 ; Wells from Tor Hill, page 246; and Gloucester from the Southeast, page 312. In his best work of this kind Mr. Pennell has great facility, felicity, and economy of touch in expressing the richness and mystery of architectural subjects. His broken pen stroke is finely suggestive of the weathered lines and surfaces of ancient walls, as well as of foliage and herbage. Yet there are evidences in all these drawings that the artist has not done all that he is capable of doing. A lack of that rigorous precision which forms the basis of the best delineation is more or less apparent. This is conspicuously shown in the careless perspective of the tower of Gloucester, on page 312, where the stringcourses that should be parallel are running towards separate horizons. In many cases a tendency to introduce over-emphatic spots of shade is manifest, as in Peterborough Cathedral from the Market Place, on page 74, where the solid black under the central archway is in violent contrast with the delicate open delineation of the rest of the drawing, and in false relationship to the shadows under the arcade in the left foreground. The same excessive blackness occurs in the archway of the Exchequer Gate, and in the figures to the right, in the view of the façade of Lincoln, on page 162, and in many of the other drawings. It is most unpleasantly manifest on page 283, where the extreme darkness of the trees is harshly discordant with the colorless background ; and on page 290, The Long Walk in Winter, where the inky blackness of the tree trunks and figures is as false in tone as it is painful in effect. In open delineation, solid blacks are always out of place ; they are proper only where they are connected with the general scheme by intermediate tones. The supreme master thus far in the treatment. of such subjects is Samuel Prout. Pront’s genius was indeed limited. He could not appreciate the beauty of architecture ; but he had a rare feeling for its picturesqueness, and matchless skill in tranquil, suggestive, and harmonious delineation. Few men have understood so well as he how to make his points of vigorous shade telling without rendering them inharmonious. He prepares for them, even in his most open line work, by deftly executed transitional tones, giving richness and color throughout the drawing.

Mr. Pennell’s best work is so good that it seems a pity he should not make it a great deal better. He has an admirable native gift which ought to be more finely cultivated. His popularity with a not over-discriminating public has been too easily won. His ideals appear to have been derived largely from the vigorous line work of modern etchers like Lalanne and Seymour Haden, men who though possessing strong artistic feeling and unusual executive skill, have, like himself, failed to discipline their powers thoroughly by exact and refined practice. The offhand power and brilliancy of such work are attractive, but not permanently satisfying to a cultivated taste.

Mr. Pennell’s poorest work among these illustrations appears to us to be that which he has done in fuller chiaroscuro. The drawings of this kind are all defective in tonic relations, and are often singularly eccentric and inharmonious in method, as well as sloppy and often unmeaning in form. In the Durham Cathedral, page 76, for instance, the distant hillside and the nearer tree have almost the same value and quality. The details of the far-off cathedral are made out with dry and unnatural distinctness, while all other parts of the subject are vaguely, though not masterfully, suggested. The graceful subject on page 193 is spoiled by the false value given to the distant cathedral, which could not tell as darker than the shaded parts of the trees between which it is seen. The illustration of the West Front of Lincoln, page 163, is conspicuously bad. It is both inartistic and untrue in light and shade. It renders none of the beauty of this remarkable monument, and, while reducing its rich details to formless vagueness, the artist has taken the pains to elaborate the ugly costumes of the over-prominent foreground figures.

We make these remarks in no captious spirit. We recognize in Mr. Pennell an artist of superior natural abilities ; but we think that he has here largely failed to do himself justice. For this, however, the present state of public taste is, we apprehend, more to be blamed than the artist himself.

The make-up of the book is appropriate, though the highly calendered paper used is offensive both to the eye and to the touch. It is to be hoped that the necessity for the use of such paper in the printing of process blocks may erelong be obviated.

  1. English Cathedrals. By Mrs. SCHUYLER VAN RENSSELAER. Illustrated with one hundred and fifty-four Drawings by JOSEPH PENNELL ; also with Plans and Diagrams. New York: The Century Company. 1892.