CHAPTER XV.

MIDDLE GOTHIC GLASS.

Towards the fourteenth century, it seems, a wave of realism swept over Gothic art. So much is this so that a relatively speaking naturalistic form of ornamental detail is the most marked feature of the Decorated period, giving it its name, and, indeed, its claim to be a style.

No great stress has been laid in the foregoing chapters upon this new departure in naturalism, because it did not so very vitally affect design. When it is said that glass followed always the fashion of architecture, that is as much as to say that, as the sculptors took to natural instead of conventional foliage, so did the glass painters; and there is not much more to tell. To trace the development of naturalistic design would lead us far astray. Enough to say that, by the naturalistic turn of its ornamental foliage you may recognise the period called "Decorated." How far that naturalism of Decorated detail may be to the good is a question there is no need here to dispute. It was a new departure. The new work lacked something of the simple dignity and self-restraint which marked the earlier, and it had not yet the style and character which came in the next century of more consistently workmanlike treatment. In so far it was a kind of prelude to Perpendicular work. This is not to deny that excellent work was done in the Decorated period, especially perhaps in glass, where naturalism, at its crudest, is less offensive than in wood or stone. But there is no getting over the fact that the period was intermediate; and Decorated glass is in a state of transition (1) between the archaism of the early and the accomplishment of the later Gothic; (2) between the conventional ornament which merely suggests nature and natural foliage conventionally treated; (3) between strong rich colour and delicate silvery glass. The transition of style is nowhere more plainly to be traced than in the grisaille of the period. At first the character of fourteenth century grisaille did not greatly differ from earlier work, except in the form of the painted detail. That from S. Urbain, Troyes, on page 333, is a typical instance of Early French Transition foliage, in which the scroll is only less strong and vigorous than before. Precisely the same kind of detail is shown again in the lower of the two instances, likewise from Troyes, opposite; but already natural leaves begin to mingle with it; whilst in the illustration above it, though the mosaic border is characteristically early, the foliage in grisaille is deliberately naturalistic. The grisaille at Troyes, by the way, often reminds one of that at York Minster. It is mainly by the naturalistic character of the ivy scroll, or perhaps it would be nearer the mark to say of the leaves upon it, that the design from Norbury, Derbyshire (page 162), betrays its later date, by that and the absence of cross-hatching on the background. The glazing of the window is still thoroughly mosaic.

There is a different indication of transition in the little panel from S. Pierre at Chartres, almost entirely in white glass, on page 163. The foliated ornament is here still early in character; but, it will be seen, there is no longer any pretence of leading up the bands of clear glass in separate strips. They are only bounded on one side by a lead line. That is so again in the three designs from Chartres Cathedral above, where, further, the background is clear of paint; and in those from Evreux, on pages 165, 284. There the background is cross-hatched; but in one case the foliage is naturalistic.

The coloured strapwork in the grisaille from the Lady Chapel of Rouen Cathedral on page 165 is frankly mosaic; but the foliated ends of the straps, gathered together into a central quatrefoil in a quite unusual fashion, indicates the new spirit. The white glass is there painted with trailing foliage in outline upon a clear ground, not shown in the sketch, which is merely a diagram of the glazing. The grisaille from Stanton S. John, Oxford, here given, still hesitates rather between two opinions. The foliage is naturalistic, but the background is cross-hatched; the broad diagonal bands, patterned with paint, are glazed in colour; the rings of white are not separately leaded. That sort of thing has occurred, as already pointed out, before; but it was not till the fourteenth century, or thereabouts, that the strapwork of white lines, forming so characteristic a feature in Decorated grisaille, are systematically indicated by painted outlines and not glazed in if it could be helped.

You have only to examine the crossing of the white lines in any of these last-mentioned patterns to see that, now that they are not separately glazed, they do not really interlace as before. It is out of the question that they should.

It is easy enough to glaze up bands so that they shall interlace; but, when some of the drawing lines are lead and some paint, it occurs continually that you want a leaded line to pass behind a line of clear glass--which, of course, is a physical impossibility. It follows that the pretended interlacing comes to grief. The pattern is confused (it is worse when there is no hatched background) by the occurrence of leads, stronger than the painted lines, which, so far from playing any part in the design, occur just at the points where they most interfere with it.

That this did not deter them, that they made a shift with interlacing which does not truly interlace, marks a falling off in what may be called the conscientiousness of the Gothic designers. French and English Decorated grisaille, effective as it often is in the window, is distinctly less satisfactory in design than the common run of earlier work. Its charm is never in its detail.

The patterns may be ingenious and not without grace, but they are never altogether admirable, any more than are the figures.

What you most enjoy in it is the distribution of white and colour; and you enjoy it most when you do not too curiously examine into the detail of the design, when you are satisfied to enjoy the colour, and do not look for form, which after all is of less account in glass.

So far as effect only is concerned, quarry work, the mere glazing in squares, answers in many places (such, for example, as the clerestories of narrow churches, where you could not possibly enjoy any detail of design that might be there) all the purpose of grisaille; and it was commonly resorted to. But the painting upon such quarries counts for very little; it is far too small and fine in detail to have any effect further than to tone the glass a little, which would have been unnecessary if the glass employed had been less clear. In fact, delicate paint on distant clerestory glass is much ado about very little; and one cannot help thinking that plain glazing would there have answered all the purpose of the most delicately painted pattern work. The fourteenth century glaziers seldom complicated their quarry work by the introduction of bands or straps of colour between the quarries, or by the introduction of colour other than such as might occur in rosettes or shields and so on, planted upon them, rather than worked into the design. Occasionally, however, as at Châlons-sur-Marne, you come upon an ornamental window (page 167) in which quarries are separated by bands of clear white, a certain amount of colour being introduced in the form of yellow quarries substituted at regular intervals for the white. On the same page is another coloured diaper window designed on quarry lines, also at Châlons. In that quarries of white and yellow are separated by a trellis of blue. Something of the sort is to be seen also at S. Radegonde, Poitiers.

In these cases the painting, as will be seen, is strong enough to hold its own at a considerable distance from the eye, but the effect is not very happy. When, by the way, it was said that delicate painting on distant quarries was lost, it was not meant to imply that strong painting on quarries would be a happy solution of the difficulty. As a matter of experience, it is seldom satisfactory. On the other hand, the common expedient of leading up the coloured backgrounds to figure work in small squares of ruby, green, and so on, was generally the means of securing good broken colour.

It can hardly be said that geometric pattern windows in strong colour are ever very successful. The Germans, who, it should be remembered, call their second Gothic period the "Geometric," often attempted it, but without conspicuous success.

In Germany it was customary to use geometric diaper work long after it had gone out of use in France. In fact, it is there more likely a sign of the second period. The crosslines in the diaper from Regensburg (opposite) would have been in lead, not paint, if the work had been executed in the thirteenth century; again, the diaper below it would not at that period have been painted in the likeness of oak-leaves. Diaper of this kind was not used merely to fill up between medallions, but as background, for example, to canopy work. Frequently it was very small in scale, as well as elaborate in pattern. It can hardly be said that it was always worth the pains spent upon it--often it was not; but the Germans avoided, as a rule, the dangerous red and blue combination, and preferred, as did also the Italians, less stereotyped arrangements of green and yellow, or of red and green, or of red and green and yellow; if they ventured upon red and blue, it was with a difference very much to their credit. For example, they would enclose diamonds of ruby in bands of purple-brown, with just a point of blue at the interstices; again, they would make a diaper of purple, purple-brown, and grey; and in many another way show that they deliberately aimed at colour in such work--whereas many of the Early diapers suggest that the glazier was thinking more of pattern. An instance of heraldic diaper is given on page 169.

In Italy also you find sometimes, as at Florence and Assisi, medallion windows with mosaic diaper between, or mosaic diaper used as background to figures which certainly cannot be described as Early.

The Germans differed from the rest of us in their frank use of geometric pattern. We habitually disguised it more or less, clothing it most likely with foliation; they used it quite nakedly, and were not ashamed. Instances of this innocent use of geometric form are here given. At Freiburg are quite a number of windows entirely of geometric pattern work. There is a good deal of white glass in them, but they count rather for colour than for grisaille. It would not be quite unfair to say they fall between the two stools. These designs are much more pleasing in the glass than in black and white (where they have rather too much the appearance of floor-cloth), but they are by no means the happiest work of the Germans of that day. Where they were really most successful, more successful than their contemporaries, was in foliated or floral pattern windows, and those of a kind also standing dangerously near midway between colour and grisaille. The method of execution employed in them was to a large extent strictly mosaic; but there is quite a refreshing variety and novelty, as well as very considerable ingenuity, in their design.

The window from Regensburg on page 389 sets out very much as if it were going to be a grisaille window; but it has, in the first place, more colour than is usual in grisaille, and, in the second, it will be seen that the little triangular ground spaces next the border are filled with pot-metal. The contrast of the set pattern and the four coloured leaves crossing each circle with the flowing undergrowth of grisaille is unusual, and so is the cunning alternation of cross-hatching and plain white ground. The designs from Munich Museum on pages 171 to 174 have nothing in common with grisaille. The design consists of natural foliage chiefly in white, growing tree-like upon a coloured ground up the centre of the light. In the one the stem is waved, in the other it takes a spiral form, in the third it is more naturalistic. But nature is not very consistently followed. What appears like a vine on page 171 has husks or flowers which it is not easy to recognise; and the ivy here is endowed with tendrils. The border of convolvulus leaves and the hop scroll, opposite, are unmistakable, though there is some inconsistency between the naturalness of the leaves and the stiffness of their growth. The ivy pattern differs from the others inasmuch as the leaves show light against the yellow ground, whilst the green stem and stalks tell dark upon it, and there is a band of red within the outer border which holds the rather spiky leaves together. The most interesting window of this kind illustrated is that on page 174, in which the stem is ingeniously twisted into quatrefoil medallion shapes, so as to allow a change in the colour of the ground, and the leaves are designed to go beyond the filling and form a pattern upon the border. The rose is a hackneyed theme enough, but this at least is a new way of working it out. Fourteenth century German windows are altogether more varied in design than contemporary French or English work. The glass is not so much all of one pattern. There are more surprises in it. The Germans treated grisaille in a way very much their own. At the risk of a certain coarseness of execution, they would paint out the background to their natural foliage in solid pigment, or in brown just hatched with lines scratched through to the clear glass. That is very effectively done, for example, at the Church of S. Thomas at Strassburg. It is not contended that this is at all a better plan than that practised in France or England: it is on the whole less happy; but there are positions in which it is more to the purpose; and it has at least the merit of being different; it suggests something better than it accomplishes, and it is a timely reminder that the best methods we know of cannot be accepted as final.

Again at Regensburg there is some distant ornamental work, so simple in execution that it is little more than glazing in colours; in fact just what distant work should be--effective in its place without any waste of labour.

A word or two remains to be said about borders. The narrower decorated light implied, as was said, a narrower border. It was, as a rule, only when a wide Early window had to be glazed that there was room for a broad one. In that case it showed of course the new naturalism, with perhaps the added interest of animal life, as here illustrated; but there lingers in German borders such as this and the one on page 338, something of early tradition. It looks as if it would not be difficult to accept glazing lines like these and fill them in with painted detail _à la Romanesque_. In one of the windows in York Minster there is a border of alternate leaves and monkeys, both much of a size, which broadens out at the base, affording space for the representation of a hunt, men, dogs, grass and all complete.

There was another reason for the adoption of a narrower border. Not only were windows narrower now, but their arched heads were cusped, which made it exceedingly difficult to carry any but the narrowest possible border round them satisfactorily. It will be seen how awkwardly the border fits (or does not fit) the window head on page 155. Even the simplest border had to be very much distorted in order to make it follow the line of the masonry; and, in any case, it gave a very ugly shape within the border, and one again difficult to fill. Already, at the beginning of the fourteenth century, the designer found it convenient to run his border straight up into the cusped head of the light and let the stonework cut it abruptly short; that occurs at Carcassonne. Sometimes, as at Tewkesbury, the inconvenient border is allowed to end just above the springing line of the arch, against a pinnacle of the canopy, beyond which point there is only a line or two of white or colour, by way of frame or finish to the background. An unusual but quite satisfactory way of getting over the difficulty of carrying the border round the window head is, to accept the springing line of the arch as the end of the central design, and to make the foliated border spread and fill the entire window head above. Some quarry lights in the triforium at Evreux are effectively treated in that manner.

Types of ordinary Decorated borders, English, French, and German, are shown in this and the preceding chapter. The leafage springs from one side or the other or from a central stem, or from either side of a waving stem, or from two stems intertwined (page 158). Sometimes the ground on one side is of a different colour from that on the other; in any case the glazing is usually simple. One of the leaf borders at Rouen Cathedral includes a series of little green birds; another, an oak pattern, is inhabited at intervals by squirrels and wild men of the woods. Rather interesting variations upon the ordinary type of border are given on this and the preceding pages. The broader one above is of distinctly unusual character, inasmuch as it has no background except the painting out, and the colour of the leafage varies quasi-accidentally.

The use of the rosette borders on pages 171, 172 is sufficiently accounted for by the desire to get contrast to the foliated filling, but it occurs at all periods more or less. So does the "block border"; but for all that it is almost as characteristic of Decorated work as the leaf border. It is seen in its simplest form on page 144. On page 389 it is associated with foliage and rosettes. A typical form of it is where the blocks are charged with heraldic devices, which may serve to indicate the date, or to confuse one. In the design from Evreux on page 160 there occur, for example, the Fleurs-de-Lys of France alternating with the Castle of Castille. These particular charges occur frequently in the windows of the S. Chapelle at Paris, and in the lights from that source now in the South Kensington Museum; and they go perhaps to show that Blanche of Castille (who married Louis VIII.) gave them to the chapel, or that they were in her memory. She died in 1252. It is most improbable that the Evreux glass should belong to so early a date as that. Were it so, the occurrence of this kind of thing in such early work would only go to show that heraldic devices are as old as heraldry, and that when the glazier had a narrow light to fill he treated it as a narrow light, with a border in proportion to its width: he certainly did that at the S. Chapelle. The fact remains that this particular form of "block" border marks, as a rule, the approach of the fourteenth century.

It may be as well to remind the reader that dates and periods are only mentioned in order to save circumlocution. When the thirteenth century is mentioned, it is not meant to convey the year 1201, nor yet 1299, but the century in its prime. And, what is more, it is not meant to say that the work ascribed to that period was quite certainly and indisputably done after the year 1200 or before the year 1300, but only that it bears the mark of the century--which, from the present point of view, is the important thing. The precise and certain year in which this or that device was by exception for the first time employed, or until which by chance a practically obsolete practice survived, is interesting (if it can be ascertained) only as a question of archæology. Anyway, a workman would rather believe the evidence of his eyes, which he can trust, than of documents, which, even if authentic, may not be trustworthy, and which are perhaps open to misinterpretation.

Typically Decorated glass, apart from the ornamental windows just referred to, is the least interesting of Gothic. There is in it a straying from Early tradition without reaching the later freedom and attainment. In colour it has neither the strength of the Early work nor the delicacy of the Late. It marks some progress in technique, but little in design, and none in taste.