Lectures on the rise and development of medieval architecture; vol. 1

by Sir George Gilbert Scott

The Romanesque windows were simply openings with round heads, the jambs and arches being either perfectly plain, moulded, with or without enrichment, or the jambs shafted. These windows were most usually isolated, but were here and there grouped into couplets, triplets, etc., or made to form portions of continuous arcading.

In the early days of the transition the windows remained unaltered, otherwise than as to the general refinement of their details. Later on the arches were made pointed, and their proportions somewhat elongated; and even in the fully-developed Early Pointed style—properly so called—the window differs little in principle from that of the Romanesque period, though, in fact, it assumes a widely different form, through its carrying towards their ultimate results the principles of grouping begun during the previous style, and those of refinement and elongation incident to the transition.

It is in carrying out these principles to a still greater extent that the Early Pointed of England differs from that of France. It is really the same style, and no important feature can be pointed out in the one country which is not to be found in the other; but just as the Germans, by dwelling longer on the Romanesque style, rendered it more refined and perfect than elsewhere, so the English, by the continued retention of the unmullioned window, systematised its use in a manner not equalled in other countries. I see no difference of principle in the fenestration of the Early French and the Early English Pointed styles: in both the principle was the decoration and combination of single lights. Nor do I see that in England this was done in a manner essentially differing in any respect from what was common in France. The great difference was the far greater width of the French openings, which often rendered their windows inelegant in proportion, while it offered a noble field for stained glass. The characteristic of the English windows, on the other hand, was narrow and tall proportions, and a greater amount of enrichment of the jambs and arches, though none of these are, by any means, constant features. Sometimes we find in English works lightness carried to a vicious extreme, as in the beautiful but frail eastern transept at Worcester; though in a majority of instances it retains a masculine firmness and solidity, as in the east end of Whitby.

Time would fail me to illustrate the magnificent combinations of this early class of window to be found in cathedrals and monastic churches—as the east end of Ely, the west at Llandaff, or the north transept at York; nor would it be possible to enumerate the simple and impressive village churches to which, in their humbler forms, though with equal artistic merit, they lend such a charm. The style is too well known in England to need minute description, and its merits too fully acknowledged to need enforcement from me.

I will rather proceed to consider that great invention which may be considered to complete the series of developments which constituted Pointed architecture: I mean the mullioned and traceried window; not that I consider it as in all points better than its predecessor, nor that in our own revival it should supersede it; but that, as a matter of fact, Gothic architecture would have been imperfect without it. Like almost every other feature of Pointed architecture, the traceried window grew out of the Romanesque.

In all periods of Romanesque we find occasionally two or more arched openings comprised under one enclosing arch. The arrangement is more frequent in belfry windows and triforium openings than elsewhere, but occurs in ordinary windows, especially in secular buildings. The space intervening between the large arch and the two or more placed below it was, even as early as this, occasionally pierced with circles or other forms of opening. Here, then, we have the elements of the mullioned window before even the introduction of the pointed arch. In the same situations it gradually developed itself, step by step, during the Early Pointed period, so that we have in triforium arcades and in other positions a pretty full development of what is called plate tracery before its use became frequent for ordinary windows. The case was pretty much the same both in France and England, though on the whole the love of placing two openings under one arch was greater in France; thus, we see in the aisles at Chartres two plain lights under one arch with a circular opening, and above, in the clerestory, a very large circle in the head with somewhat complex subordinate piercings. The same is the case at Bourges, where three lights are often comprised under one arch, with a single circle in the head.

The next great element which aided in producing tracery windows was the wheel, or other richly-pierced circular window. This, again, originated under the Romanesque style, as we may see at Barfreston and elsewhere. It is, in fact, a very close approach to tracery, and when placed in the space between comprising and comprised arches, it almost completes the change. All that is wanted is the piercing of the intervening spaces in forms whose outlines are parallel to the main piercings, so as to form what Professor Willis calls bar tracery. This was, I fancy, commenced in France—though there are very early traces of it in England—and was done at first in a partial and clumsy manner, as in some windows at Le Mans and Tours, but soon was systematised.

I do not see that in any of the previous steps the French were in advance of the English architects, but in this last step I think they were so, and this led them to a much earlier abandonment of the single window and its combinations; so that for some time the French were using tracery windows, while we were rendering more perfect the unmullioned system—not from want of knowledge of the other, but rather from a preference for a system in which we were producing more beautiful combinations than our neighbours had attained.

It is not a very profitable question to inquire by how many years the French may have been in advance of us in this development, and it is so exceedingly difficult to get at positive dates of the erection of buildings in either country, that it would be impossible if desired. The fact, no doubt, is, that for many years the two kinds of window were contemporaneous. Thus, traceried windows may have been in use at Rheims and Amiens, while the older kind was being used at Bourges and Chartres.

It is said that in England the fully-developed bar tracery was first used in Westminster Abbey, which was commenced in 1245; but this is merely an assumption; and it is clear that it was used in the eastern part of St. Paul’s, a part of which was consecrated in 1240. The east window of Netley Abbey looks very early, but I do not know its real date, but believe it is said to have been finished in 1249; while the eastern windows at Lincoln look too thoroughly developed to be very early specimens, though known to have been erected between 1256 and 1280. In any case the change had fully established itself in England during the third quarter of the century.[40]