§ XX. We have hitherto supposed that all grouped or clustered shafts have been the result or the expression of an actual gathering and binding together of detached shafts. This is not, however, always so: for some clustered shafts are little more than solid piers channelled on the surface, and their form appears to be merely the development of some longitudinal furrowing or striation on the original single shaft. That clustering or striation, whichever we choose to call it, is in this case a decorative feature, and to be considered under the head of decoration.
§ XXI. It must be evident to the reader at a glance, that the real serviceableness of any of these grouped arrangements must depend upon the relative shortness of the shafts, and that, when the whole pier is so lofty that its minor members become mere reeds or rods of stone, those minor members can no longer be charged with any considerable weight. And the fact is, that in the most complicated Gothic arrangements, when the pier is tall and its satellites stand clear of it, no real work is given them to do, and they might all be removed without endangering the building. They are merely the expression of a great consistent system, and are in architecture what is often found in animal anatomy,—a bone, or process of a bone, useless, under the ordained circumstances of its life, to the particular animal in which it is found, and slightly developed, but yet distinctly existent, and representing, for the sake of absolute consistency, the same bone in its appointed, and generally useful, place, either in skeletons of all animals, or in the genus to which the animal itself belongs.
§ XXII. Farther: as it is not easy to obtain pieces of stone long enough for these supplementary shafts (especially as it is always unsafe to lay a stratified stone with its beds upright) they have been frequently composed of two or more short shafts set upon each other, and to conceal the unsightly junction, a flat stone has been interposed, carved into certain mouldings, which have the appearance of a ring on the shaft. Now observe: the whole pier was the gathering of the whole wall, the base gathers into base, the veil into the shaft, and the string courses of the veil gather into these rings; and when this is clearly expressed, and the rings do indeed correspond with the string courses of the wall veil, they are perfectly admissible and even beautiful; but otherwise, and occurring, as they do in the shafts of Westminster, in the middle of continuous lines, they are but sorry make-shifts, and of late since gas has been invented, have become especially offensive from their unlucky resemblance to the joints of gas-pipes, or common water-pipes. There are two leaden ones, for instance, on the left hand as one enters the abbey at Poet’s Corner, with their solderings and funnels looking exactly like rings and capitals, and most disrespectfully mimicking the shafts of the abbey, inside.
Thus far we have traced the probable conditions of shaft structure in pure theory; I shall now lay before the reader a brief statement of the facts of the thing in time past and present.
§ XXIII. In the earliest and grandest shaft architecture which we know, that of Egypt, we have no grouped arrangements, properly so called, but either single and smooth shafts, or richly reeded and furrowed shafts, which represent the extreme conditions of a complicated group bound together to sustain a single mass; and are indeed, without doubt, nothing else than imitations of bundles of reeds, or of clusters of lotus:42 but in these shafts there is merely the idea of a group, not the actual function or structure of a group; they are just as much solid and simple shafts as those which are smooth, and merely by the method of their decoration present to the eye the image of a richly complex arrangement.
§ XXIV. After these we have the Greek shaft, less in scale, and losing all suggestion or purpose of suggestion of complexity, its so-called flutings being, visibly as actually, an external decoration.
§ XXV. The idea of the shaft remains absolutely single in the Roman and Byzantine mind; but true grouping begins in Christian architecture by the placing of two or more separate shafts side by side, each having its own work to do; then three or four, still with separate work; then, by such steps as those above theoretically pursued, the number of the members increases, while they coagulate into a single mass; and we have finally a shaft apparently composed of thirty, forty, fifty, or more distinct members; a shaft which, in the reality of its service, is as much a single shaft as the old Egyptian one; but which differs from the Egyptian in that all its members, how many soever, have each individual work to do, and a separate rib of arch or roof to carry: and thus the great Christian truth of distinct services of the individual soul is typified in the Christian shaft; and the old Egyptian servitude of the multitudes, the servitude inseparable from the children of Ham, is typified also in that ancient shaft of the Egyptians, which in its gathered strength of the river reeds, seems, as the sands of the desert drift over its ruin, to be intended to remind us for ever of the end of the association of the wicked. “Can the rush grow up without mire, or the flag grow without water?—So are the paths of all that forget God; and the hypocrite’s hope shall perish.”
§ XXVI. Let the reader then keep this distinction of the three systems clearly in his mind: Egyptian system, an apparent cluster supporting a simple capital and single weight; Greek and Roman system, single shaft, single weight; Gothic system, divided shafts, divided weight: at first actually and simply divided, at last apparently and infinitely divided; so that the fully formed Gothic shaft is a return to the Egyptian, but the weight is divided in the one and undivided in the other.
§ XXVII. The transition from the actual to the apparent cluster, in the Gothic, is a question of the most curious interest; I have thrown together the shaft sections in Plate II. to illustrate it, and exemplify what has been generally stated above.43
II.
PLANS OF PIERS.
1. The earliest, the most frequent, perhaps the most beautiful of all the groups, is also the simplest; the two shafts arranged as at b or c, (Fig. XIV.) above, bearing an oblong mass, and substituted for the still earlier structure a, Fig. XIV. In Plate XVII. (Chap. XXVII.) are three examples of the transition: the one on the left, at the top, is the earliest single-shafted arrangement, constant in the rough Romanesque windows; a huge hammer-shaped capital being employed to sustain the thickness of the wall. It was rapidly superseded by the double shaft, as on the right of it; a very early example from the cloisters of the Duomo, Verona. Beneath, is a most elaborate and perfect one from St. Zeno of Verona, where the group is twice complicated, two shafts being used, both with quatrefoil sections. The plain double shaft, however, is by far the most frequent, both in the Northern and Southern Gothic, but for the most part early; it is very frequent in cloisters, and in the singular one of St. Michael’s Mount, Normandy, a small pseudo-arcade runs along between the pairs of shafts, a miniature aisle. The group is employed on a magnificent scale, but ill proportioned, for the main piers of the apse of the cathedral of Coutances, its purpose being to conceal one shaft behind the other, and make it appear to the spectator from the nave as if the apse were sustained by single shafts, of inordinate slenderness. The attempt is ill-judged, and the result unsatisfactory.
Fig. XVII.
§ XXVIII. 2. When these pairs of shafts come near each other, as frequently at the turnings of angles (Fig. XVII.), the quadruple group results, b 2, Fig. XIV., of which the Lombardic sculptors were excessively fond, usually tying the shafts together in their centre, in a lover’s knot. They thus occur in Plate V., from the Broletto of Como; at the angle of St. Michele of Lucca, Plate XXI.; and in the balustrade of St. Mark’s. This is a group, however, which I have never seen used on a large scale.44
§ XXIX. 3. Such groups, consolidated by a small square in their centre, form the shafts of St. Zeno, just spoken of, and figured in Plate XVII., which are among the most interesting pieces of work I know in Italy. I give their entire arrangement in Fig. XVIII.: both shafts have the same section, but one receives a half turn as it ascends, giving it an exquisite spiral contour: the plan of their bases, with their plinth, is given at 2, Plate II.; and note it carefully, for it is an epitome of all that we observed above, respecting the oblique and even square. It was asserted that the oblique belonged to the north, the even to the south: we have here the northern Lombardic nation naturalised in Italy, and, behold, the oblique and even quatrefoil linked together; not confused, but actually linked by a bar of stone, as seen in Plate XVII., under the capitals.
Fig. XVIII.
4. Next to these, observe the two groups of five shafts each, 5 and 6, Plate II., one oblique, the other even. Both are from upper stories; the oblique one from the triforium of Salisbury; the even one from the upper range of shafts in the façade of St. Mark’s at Venice.45
§ XXX. Around these central types are grouped, in Plate II., four simple examples of the satellitic cluster, all of the Northern Gothic: 4, from the Cathedral of Amiens; 7, from that of Lyons (nave pier); 8, the same from Salisbury; 10, from the porch of Notre Dame, Dijon, having satellites of three magnitudes: 9 is one of the piers between the doors of the same church, with shafts of four magnitudes, and is an instance of the confusion of mind of the Northern architects between piers proper and jamb mouldings (noticed farther in the next chapter, § XXXI.): for this fig. 9, which is an angle at the meeting of two jambs, is treated like a rich independent shaft, and the figure below, 12, which is half of a true shaft, is treated like a meeting of jambs.
All these four examples belonging to the oblique or Northern system, the curious trefoil plan, 3, lies between the two, as the double quatrefoil next it unites the two. The trefoil is from the Frari, Venice, and has a richly worked capital in the Byzantine manner,—an imitation, I think, of the Byzantine work by the Gothic builders: 1 is to be compared with it, being one of the earliest conditions of the cross shaft, from the atrium of St. Ambrogio at Milan. 13 is the nave pier of St. Michele at Pavia, showing the same condition more fully developed: and 11 another nave pier from Vienne, on the Rhone, of far more distinct Roman derivation, for the flat pilaster is set to the nave, and is fluted like an antique one. 12 is the grandest development I have ever seen of the cross shaft, with satellite shafts in the nooks of it: it is half of one of the great western piers of the cathedral of Bourges, measuring eight feet each side, thirty-two round.46 Then the one below (15) is half of a nave pier of Rouen Cathedral, showing the mode in which such conditions as that of Dijon (9) and that of Bourges (12) were fused together into forms of inextricable complexity (inextricable I mean in the irregularity of proportion and projection, for all of them are easily resolvable into simple systems in connexion with the roof ribs). This pier of Rouen is a type of the last condition of the good Gothic; from this point the small shafts begin to lose shape, and run into narrow fillets and ridges, projecting at the same time farther and farther in weak tongue-like sections, as described in the “Seven Lamps.” I have only here given one example of this family, an unimportant but sufficiently characteristic one (16) from St. Gervais of Falaise. One side of the nave of that church is Norman, the other Flamboyant, and the two piers 14 and 16 stand opposite each other. It would be useless to endeavor to trace farther the fantasticism of the later Gothic shafts; they become mere aggregations of mouldings very sharply and finely cut, their bases at the same time running together in strange complexity and their capitals diminishing and disappearing. Some of their conditions, which, in their rich striation, resemble crystals of beryl, are very massy and grand; others, meagre, harsh, or effeminate in themselves, are redeemed by richness and boldness of decoration; and I have long had it in my mind to reason out the entire harmony of this French Flamboyant system, and fix its types and possible power. But this inquiry is foreign altogether to our present purpose, and we shall therefore turn back from the Flamboyant to the Norman side of the Falaise aisle, resolute for the future that all shafts of which we may have the ordering, shall be permitted, as with wisdom we may also permit men or cities, to gather themselves into companies, or constellate themselves into clusters, but not to fuse themselves into mere masses of nebulous aggregation.
CHAPTER IX.
THE CAPITAL
§ I. The reader will remember that in Chap. VII. § V. it was said that the cornice of the wall, being cut to pieces and gathered together, formed the capital of the column. We have now to follow it in its transformation.
We must, of course, take our simplest form or root of cornices (a, in Fig. V., above). We will take X and Y there, and we must necessarily gather them together as we did Xb and Yb in Chap. VII. Look back to the tenth paragraph of Chap. VII., read or glance it over again, substitute X and Y for Xb and Yb, read capital for base, and, as we said that the capital was the hand of the pillar, while the base was its foot, read also fingers for toes; and as you look to the plate, Fig. XII., turn it upside down. Then h, in Fig. XII., becomes now your best general form of block capital, as before of block base.
§ II. You will thus have a perfect idea of the analogies between base and capital; our farther inquiry is into their differences. You cannot but have noticed that when Fig. XII. is turned upside down, the square stone (Y) looks too heavy for the supporting stone (X); and that in the profile of cornice (a of Fig. V.) the proportions are altogether different. You will feel the fitness of this in an instant when you consider that the principal function of the sloping part in Fig. XII. is as a prop to the pillar to keep it from slipping aside; but the function of the sloping stone in the cornice and capital is to carry weight above. The thrust of the slope in the one case should therefore be lateral, in the other upwards.
§ III. We will, therefore, take the two figures, e and h of Fig. XII., and make this change in them as we reverse them, using now the exact profile of the cornice a,—the father of cornices; and we shall thus have a and b, Fig. XIX.
Fig. XIX.
Both of these are sufficiently ugly, the reader thinks; so do I; but we will mend them before we have done with them: that at a is assuredly the ugliest,—like a tile on a flower-pot. It is, nevertheless, the father of capitals; being the simplest condition of the gathered father of cornices. But it is to be observed that the diameter of the shaft here is arbitrarily assumed to be small, in order more clearly to show the general relations of the sloping stone to the shaft and upper stone; and this smallness of the shaft diameter is inconsistent with the serviceableness and beauty of the arrangement at a, if it were to be realised (as we shall see presently); but it is not inconsistent with its central character, as the representative of every species of possible capital; nor is its tile and flower-pot look to be regretted, as it may remind the reader of the reported origin of the Corinthian capital. The stones of the cornice, hitherto called X and Y, receive, now that they form the capital, each a separate name; the sloping stone is called the Bell of the capital, and that laid above it, the Abacus. Abacus means a board or tile: I wish there were an English word for it, but I fear there is no substitution possible, the term having been long fixed, and the reader will find it convenient to familiarise himself with the Latin one.
§ IV. The form of base, e of Fig. XII., which corresponds to this first form of capital, a, was said to be objectionable only because it looked insecure; and the spurs were added as a kind of pledge of stability to the eye. But evidently the projecting corners of the abacus at a, Fig. XIX., are actually insecure; they may break off, if great weight be laid upon them. This is the chief reason of the ugliness of the form; and the spurs in b are now no mere pledges of apparent stability, but have very serious practical use in supporting the angle of the abacus. If, even with the added spur, the support seems insufficient, we may fill up the crannies between the spurs and the bell, and we have the form c.
Thus a, though the germ and type of capitals, is itself (except under some peculiar conditions) both ugly and insecure; b is the first type of capitals which carry light weight; c, of capitals which carry excessive weight.
§ V. I fear, however, the reader may think he is going slightly too fast, and may not like having the capital forced upon him out of the cornice; but would prefer inventing a capital for the shaft itself, without reference to the cornice at all. We will do so then; though we shall come to the same result.
The shaft, it will be remembered, has to sustain the same weight as the long piece of wall which was concentrated into the shaft; it is enabled to do this both by its better form and better knit materials; and it can carry a greater weight than the space at the top of it is adapted to receive. The first point, therefore, is to expand this space as far as possible, and that in a form more convenient than the circle for the adjustment of the stones above. In general the square is a more convenient form than any other; but the hexagon or octagon is sometimes better fitted for masses of work which divide in six or eight directions. Then our first impulse would be to put a square or hexagonal stone on the top of the shaft, projecting as far beyond it as might be safely ventured; as at a, Fig. XX. This is the abacus. Our next idea would be to put a conical shaped stone beneath this abacus, to support its outer edge, as at b. This is the bell.
Fig. XX.
§ VI. Now the entire treatment of the capital depends simply on the manner in which this bell-stone is prepared for fitting the shaft below and the abacus above. Placed as at a, in Fig. XIX., it gives us the simplest of possible forms; with the spurs added, as at b, it gives the germ of the richest and most elaborate forms: but there are two modes of treatment more dexterous than the one, and less elaborate than the other, which are of the highest possible importance,—modes in which the bell is brought to its proper form by truncation.
§ VII. Let d and f, Fig. XIX., be two bell-stones; d is part of a cone (a sugar-loaf upside down, with its point cut off); f part of a four-sided pyramid. Then, assuming the abacus to be square, d will already fit the shaft, but has to be chiselled to fit the abacus; f will already fit the abacus, but has to be chiselled to fit the shaft.
From the broad end of d chop or chisel off, in four vertical planes, as much as will leave its head an exact square. The vertical cuttings will form curves on the sides of the cone (curves of a curious kind, which the reader need not be troubled to examine), and we shall have the form at e, which is the root of the greater number of Norman capitals.
From f cut off the angles, beginning at the corners of the square and widening the truncation downwards, so as to give the form at g, where the base of the bell is an octagon, and its top remains a square. A very slight rounding away of the angles of the octagon at the base of g will enable it to fit the circular shaft closely enough for all practical purposes, and this form, at g, is the root of nearly all Lombardic capitals.
If, instead of a square, the head of the bell were hexagonal or octagonal, the operation of cutting would be the same on each angle; but there would be produced, of course, six or eight curves on the sides of e, and twelve or sixteen sides to the base of g.
Fig. XXI.
§ VIII. The truncations in e and g may of course be executed on concave or convex forms of d and f; but e is usually worked on a straight-sided bell, and the truncation of g often becomes concave while the bell remains straight; for this simple reason,—that the sharp points at the angles of g, being somewhat difficult to cut, and easily broken off, are usually avoided by beginning the truncation a little way down the side of the bell, and then recovering the lost ground by a deeper cut inwards, as here, Fig. XXI. This is the actual form of the capitals of the balustrades of St. Mark’s: it is the root of all the Byzantine Arab capitals, and of all the most beautiful capitals in the world, whose function is to express lightness.
§ IX. We have hitherto proceeded entirely on the assumption that the form of cornice which was gathered together to produce the capital was the root of cornices, a of Fig. V. But this, it will be remembered, was said in § VI. of Chap. VI. to be especially characteristic of southern work, and that in northern and wet climates it took the form of a dripstone.
Accordingly, in the northern climates, the dripstone gathered together forms a peculiar northern capital, commonly called the Early English,47 owing to its especial use in that style.
There would have been no absurdity in this if shafts were always to be exposed to the weather; but in Gothic constructions the most important shafts are in the inside of the building. The dripstone sections of their capitals are therefore unnecessary and ridiculous.
§ X. They are, however, much worse than unnecessary.
Fig. XXII.
The edge of the dripstone, being undercut, has no bearing power, and the capital fails, therefore, in its own principal function; and besides this, the undercut contour admits of no distinctly visible decoration; it is, therefore, left utterly barren, and the capital looks as if it had been turned in a lathe. The Early English capital has, therefore, the three greatest faults that any design can have: (1) it fails in its own proper purpose, that of support; (2) it is adapted to a purpose to which it can never be put, that of keeping off rain; (3) it cannot be decorated.
The Early English capital is, therefore, a barbarism of triple grossness, and degrades the style in which it is found, otherwise very noble, to one of second-rate order.
§ XI. Dismissing, therefore, the Early English capital, as deserving no place in our system, let us reassemble in one view the forms which have been legitimately developed, and which are to become hereafter subjects of decoration. To the forms a, b, and c, Fig. XIX., we must add the two simplest truncated forms e and g, Fig. XIX., putting their abaci on them (as we considered their contours in the bells only), and we shall have the five forms now given in parallel perspective in Fig. XXII., which are the roots of all good capitals existing, or capable of existence, and whose variations, infinite and a thousand times infinite, are all produced by introduction of various curvatures into their contours, and the endless methods of decoration superinduced on such curvatures.
§ XII. There is, however, a kind of variation, also infinite, which takes place in these radical forms, before they receive either curvature or decoration. This is the variety of proportion borne by the different lines of the capital to each other, and to the shafts. This is a structural question, at present to be considered as far as is possible.
Fig. XXIII.
§ XIII. All the five capitals (which are indeed five orders with legitimate distinction; very different, however, from the five orders as commonly understood) may be represented by the same profile, a section through the sides of a, b, d, and e, or through the angles of c, Fig. XXII. This profile we will put on the top of a shaft, as at A, Fig. XXIII., which shaft we will suppose of equal diameter above and below for the sake of greater simplicity: in this simplest condition, however, relations of proportion exist between five quantities, any one or any two, or any three, or any four of which may change, irrespective of the others. These five quantities are:
1. The height of the shaft, a b;
2. Its diameter, b c;
3. The length of slope of bell, b d;
4. The inclination of this slope, or angle c b d;
5. The depth of abacus, d e.
For every change in any one of these quantities we have a new proportion of capital: five infinities, supposing change only in one quantity at a time: infinity of infinities in the sum of possible changes.
It is, therefore, only possible to note the general laws of change; every scale of pillar, and every weight laid upon it admitting, within certain limits, a variety out of which the architect has his choice; but yet fixing limits which the proportion becomes ugly when it approaches, and dangerous when it exceeds. But the inquiry into this subject is too difficult for the general reader, and I shall content myself with proving four laws, easily understood and generally applicable; for proof of which if the said reader care not, he may miss the next four paragraphs without harm.
§ XIV. 1. The more slender the shaft, the greater, proportionally, may be the projection of the abacus. For, looking back to Fig. XXIII., let the height a b be fixed, the length d b, the angle d b c, and the depth d e. Let the single quantity b c be variable, let B be a capital and shaft which are found to be perfectly safe in proportion to the weight they bear, and let the weight be equally distributed over the whole of the abacus. Then this weight may be represented by any number of equal divisions, suppose four, as l, m, n, r, of brickwork above, of which each division is one fourth of the whole weight; and let this weight be placed in the most trying way on the abacus, that is to say, let the masses l and r be detached from m and n, and bear with their full weight on the outside of the capital. We assume, in B, that the width of abacus e f is twice as great as that of the shaft, b c, and on these conditions we assume the capital to be safe.
But b c is allowed to be variable. Let it become b2 c2 at C, which is a length representing about the diameter of a shaft containing half the substance of the shaft B, and, therefore, able to sustain not more than half the weight sustained by B. But the slope b d and depth d e remaining unchanged, we have the capital of C, which we are to load with only half the weight of l, m, n, r, i. e., with l and r alone. Therefore the weight of l and r, now represented by the masses l2, r2, is distributed over the whole of the capital. But the weight r was adequately supported by the projecting piece of the first capital h f c: much more is it now adequately supported by i h, f2 c2. Therefore, if the capital of B was safe, that of C is more than safe. Now in B the length e f was only twice b c; but in C, e2 f2 will be found more than twice that of b2 c2. Therefore, the more slender the shaft, the greater may be the proportional excess of the abacus over its diameter.
Fig. XXIV.
§ XV. 2. The smaller the scale of the building, the greater may be the excess of the abacus over the diameter of the shaft. This principle requires, I think, no very lengthy proof: the reader can understand at once that the cohesion and strength of stone which can sustain a small projecting mass, will not sustain a vast one overhanging in the same proportion. A bank even of loose earth, six feet high, will sometimes overhang its base a foot or two, as you may see any day in the gravelly banks of the lanes of Hampstead: but make the bank of gravel, equally loose, six hundred feet high, and see if you can get it to overhang a hundred or two! much more if there be weight above it increased in the same proportion. Hence, let any capital be given, whose projection is just safe, and no more, on its existing scale; increase its proportions every way equally, though ever so little, and it is unsafe; diminish them equally, and it becomes safe in the exact degree of the diminution.
Let, then, the quantity e d, and angle d b c, at A of Fig. XXIII., be invariable, and let the length d b vary: then we shall have such a series of forms as may be represented by a, b, c, Fig. XXIV., of which a is a proportion for a colossal building, b for a moderately sized building, while c could only be admitted on a very small scale indeed.
§ XVI. 3. The greater the excess of abacus, the steeper must be the slope of the bell, the shaft diameter being constant.
This will evidently follow from the considerations in the last paragraph; supposing only that, instead of the scale of shaft and capital varying together, the scale of the capital varies alone. For it will then still be true, that, if the projection of the capital be just safe on a given scale, as its excess over the shaft diameter increases, the projection will be unsafe, if the slope of the bell remain constant. But it may be rendered safe by making this slope steeper, and so increasing its supporting power.
Fig. XXV.
Thus let the capital a, Fig. XXV., be just safe. Then the capital b, in which the slope is the same but the excess greater, is unsafe. But the capital c, in which, though the excess equals that of b, the steepness of the supporting slope is increased, will be as safe as b, and probably as strong as a.48
§ XVII. 4. The steeper the slope of the bell, the thinner may be the abacus.
The use of the abacus is eminently to equalise the pressure over the surface of the bell, so that the weight may not by any accident be directed exclusively upon its edges. In proportion to the strength of these edges, this function of the abacus is superseded, and these edges are strong in proportion to the steepness of the slope. Thus in Fig. XXVI., the bell at a would carry weight safely enough without any abacus, but that at c would not: it would probably have its edges broken off. The abacus superimposed might be on a very thin, little more than formal, as at b; but on c must be thick, as at d.