chapter xii
, but little is said there of the technique. It should be noticed that the medallion in coloured wax as a form of art has been revived with considerable success in our own time and country by the Misses Casella and others. The artists just named consider that it would be impossible to finish work on the usual small scale in coloured waxes alone, without touches of pigment added with the brush. It would be interesting in this connection to know what were the exact processes of painting in wax used by the ancients. Paintings, which must have been on a small scale because they were on a ground of ivory, were executed in coloured waxes laid on by the ‘cestrum’ (Pliny, _Hist. Nat._, XXXV, 147), which is usually described as a sort of spatula, something like one of the steel tools used by artists for finishing figures in plaster. However the substance was applied, the whole process was apparently carried out in the coloured waxes. There must have been some similarity between this technique and that of the wax medallions of Renaissance and modern times.
PROPORTIONATE ENLARGEMENT.
[§ 48, _Transference of the full-sized Model to the Marble Block_, ante, p. 151.]
‘To enlarge the figure proportionately in the marble.’ Vasari has said, ante, p. 150, that the model is to be the full size of the marble so that there would be no question of enlargement but only of accurately copying the form of the model in the new material. For this mechanical aids are invoked, the latest and most elaborate of which is the ‘pointing machine’ now in common use. The appliances in Vasari’s time were much simpler. Cellini, in his _Trattato sopra la Scultura_, describes the mechanical arrangements he made for enlarging a model to the size of a proposed colossal effigy, and the principle is the same whether there is to be enlargement or exact reproduction.
The model, and a block roughly trimmed by rule of thumb to the size and shape required, but of course somewhat larger than will ultimately be needed, are placed side by side on tables of exactly the same form and dimensions. About the model is set up a sort of framework simple or elaborate, according to the character of the piece, and a framework precisely similar in all respects is disposed about the block. A measurement is then taken from one or more points on the framework to a point on the model, and from a point or points similarly situated on the other framework, and in the same relative direction, a similar measurement is led towards the block. As this is _ex hypothesi_ a little larger than the model, the full measurement cannot be taken until some of the superfluous marble has been removed by suitable tools. When this is done a point can be established on the block exactly corresponding to the point already fixed on the model. This process can be repeated as often as is necessary until all the important or salient points on the model have been successively established on the marble block, which will ultimately have approached so nearly to the exact similitude of the model, that the artist can finish it by the eye.
The nature of what has been termed the framework, from which all the measurements are taken, may vary. Cellini, on the occasion referred to, surrounded his model with a sort of skeleton of a cubical box, from the sides and corners of which he measured. In the _Encyclopédie_ of Diderot and d’Alembert, of the middle of the eighteenth century, similar square frames, like those used as stretchers for canvases, are suspended horizontally over model and block, and plumb lines are hung from the corners, so that skeleton cubes are established, which would answer the same purpose as Cellini’s box. See Plate X, A. The arrangement contemplated by Vasari was somewhat simpler. He does not establish a complete hollow cube about his model and his block, but is apparently satisfied with erecting perpendiculars beside each, from which the measures would be led. The carpenter’s square (squadra) he has in mind consists of two straight legs joined together at right angles. If one leg be laid horizontally along the table the one at right angles to it will be vertical, and from this the measurements are taken. In the treatise on Sculpture by Leon Battista Alberti there is an elaborate description of a device he invented for the purpose in view, and one of his editors has illustrated this by a drawing reproduced here in Plate X, B. The device explains itself, and any number of similar contrivances could be employed.
THE USE OF FULL-SIZED MODELS.
[§ 49, _Danger of Dispensing with the Full-sized Model_, ante, p. 151.]
The question here is of the possibility of dispensing altogether with a full-sized clay model, and proceeding at once to attack the marble with the guidance only of the small original sketch. In modern times this is practically never done, but it was the universal practice of the Greek sculptors at any rate down to the later periods of Hellenic art. These remarks of Vasari come just at the time of the change from the ancient to the modern technique, for we shall see that Donatello in the fifteenth century worked according to the simpler ancient method, while Michelangelo in the sixteenth after beginning in the same fashion finally settled down to the use of the full model, which has ever since remained _de rigeur_.
[Illustration:
PLATE X
_B_
DIAGRAM to illustrate Alberti’s method of measurement
_A_
INTERIOR OF A SCULPTOR’S STUDIO IN THE EIGHTEENTH CENTURY
With illustrations of methods of measurement ]
[Illustration:
FIG. 11.—Two views of unfinished Greek marble statue blocked out on the ancient system. In quarries on Mount Pentelicus, Athens. ]
The technique of the Greeks furnished the subject for an article by Professor Ernest Gardner in the 14th volume of the _Journal of Hellenic Studies_. He shows there by a comparison of unfinished works that the Greek sculptors attacked the marble directly, and proceeded apparently on the following method. Having obtained a block about the size and shape required they set it up before them as if in a front view, and then hewed away at the two sides till they had brought the contour of these to the exact lines required for the finished work. They then passed round through a right angle to the side, and treated in a similar fashion the front and back of the block, bringing these to the shape of the front and back of the desired figure. The block would then, when looked at from the front or back or from the sides, present the required outlines, but the section of it would still be square in every part—there would be no rounding off. The sketches, Fig. 11, show two views of a figure blocked out in this fashion by an ancient Attic sculptor. It was found in old marble workings on Mount Pentelicus, and is preserved at the modern marble quarry at the back of that mountain. We owe the use of the photographs employed to the courtesy of M. Georges Nicole, of Geneva. They were published in the volume entitled _Mélanges Nicole_, Geneva, 1905, in connection with an article on the figure by the archaeologist just named. The next process was to cut away these corners and with the guidance of the already established contours gradually bring the whole into the required shape. A small model may in every case be presupposed and there must have been some system of measurement. Indeed on some antiques, as on a crouching Venus in the gallery leading to the Venus of Milo in the Louvre, there are still to be seen the knobs (puntelli) to which measurements were taken during the progress of the work. Of the use of full-sized clay models there is in Greece no evidence at all, until the late period of the first century B.C., when we are told of Pasiteles, a very painstaking sculptor of a decadent epoch, that he never executed a work without first modelling it (nihil unquam fecit antequam finxit). This no doubt implies a full-sized model in clay, for a small sketch would not be mentioned as it is a matter of course.
The practice of the Italians is described by Cellini in words which are important enough to quote. They are from the fourth chapter of his treatise on Sculpture. ‘Now although many excellent masters of assured technique have boldly attacked the marble with their tools, as soon as they had carved the little model to completion, yet at the end they have found themselves but little satisfied with their work. For, to speak only of the best of the moderns, Donatello adopted this method in his works; and another example is Michelangelo, who had experience of both the methods, that is to say, of carving statues alike from the small model and the big, and at the end, convinced of their respective advantages and disadvantages, adopted the second method (of the full-sized model). And this I saw myself at Florence when he was working in the sacristy of S. Lorenzo (on the Medici tombs).’ As regards Michelangelo’s early practice, Vasari records in his Life that he carved the colossal marble ‘David’ with the sole aid of a small wax model, according to Vasari one of those now preserved in the Casa Buonarroti at Florence. This was in 1504. The Medici tombs date twenty years later.
In connection with the direct practice of Donatello, it is worth while referring to some words of Francesco Bocchi, a rhetorical eulogist of the arts and artists of his native Florence, who wrote in 1571 a literary effusion on the sculptor’s St. George. He notes in his introduction that Donatello was accustomed to compose his marble figures compactly and to avoid projecting hands and arms, while for his effigies in bronze he used much greater freedom in action. The difference is really one of material, and Donatello’s practice of working directly on the marble would necessarily involve this restraint in composition. Anyone accustomed to deal with marble blocks as vehicles of artistic expression, would avoid unnecessary projections as these cause great waste of material and expenditure of time. When plastering clay or wax on a flexible armature this consideration is not present, and modelled figures will naturally be freer in action than carved ones. As will presently be seen, certain marked differences in the treatment of relief sculpture depend on these same considerations of material and technique.
In direct work on the marble there is of course always the danger of the sort of miscalculation that Vasari goes on to notice. Greek figures sometimes show variations from correct proportions, for example, the left thigh of the Venus of Milo is too short, but the errors are not such as to destroy the effect of the works. Greek work in marble shows a marvellously intimate knowledge on the part of the carver of his material as well as a clear conception of what he was aiming at. Even Michelangelo yields in this respect to the ancients, for though no one was ever more thoroughly a master of the carver’s technique, he made serious mistakes in calculating proportions, as in the ‘Slave’ of the Louvre, where he has not left enough marble for the leg of the figure. Moderns generally have not the ease which tradition and practice gave to the Greek sculptors, and the full-sized model is now a necessary precaution.
ITALIAN AND GREEK RELIEFS.
[§ 52, _Pictorial or Perspective Reliefs_, ante, p. 154.]
Vasari ascribes comprehensively to the ‘ancients’ the invention of the pictorial or perspectively treated relief, which was not in use in mediaeval times, but came into vogue in the early years of the fifteenth century. The first conspicuous instance of its employment was in the models by Ghiberti for the second set of gates for the Baptistry at Florence begun in 1425, but as these gates were not finally completed till 1452, other artists had in the meantime produced works in the same style. Donatello’s bronze relief of the beheading of John the Baptist, on the font at Sienna, was completed in 1427 and shows the same treatment in a modified form. It is a treatment often called pictorial as it aims at effects of distance, with receding planes and objects made smaller according to their supposed distance from the foreground. The style has been sufficiently criticized, and it is generally agreed that it represents a defiance of the barriers fixed by the nature of things between painting and sculpture. It depended mainly however not on the influence of painting but on the study of perspective, which Brunelleschi brought into vogue somewhere about the year 1420. Brunelleschi’s perspectives, or those which he inspired, were worked out in inlaid woods, or tarsia work, see postea, p. 303 f., and exhibited elaborate architectural compositions crowded with receding lines. These compositions were adopted by Donatello and others for the backgrounds of their figure reliefs, and Ghiberti filled his nearer planes with a crowd of figures represented as Vasari describes according to the laws of linear, and so far as the material permitted, of aerial perspective.
The question of the amount of warrant for this in antique practice as a whole calls attention to an interesting moment in the general history of relief sculpture. This has been dealt with recently by Franz Wickhoff, in the Essay contributed by him to the publication of the Vienna ‘Genesis,’ and issued in an English translation by Mrs Strong under the title _Roman Art_ (London, 1900), and also in Mrs Strong’s own _Roman Sculpture_ (London, Duckworth, 1907). The tendency to multiply planes in relief, and to introduce the perspective effects and the backgrounds of a picture, shows itself in some of the late Greek or ‘Hellenistic’ reliefs, published by Schreiber under the title _Die Hellenistischen Reliefbilder_ (Wien, 1889 etc.), and more especially in the smaller frieze from the altar base at Pergamon. Etruscan relief sculpture is also affected by it. It is however in Roman work of the early imperial period that we actually find the antique prototypes for the kind of work that Vasari has in his mind. The reliefs on the tomb of the Julii at St Rémy, of the age of Augustus, those on the Arch of Titus, and most conspicuously the decorative sculpture connected with the name of Trajan, are instances in point. They show differences in plane, and occasionally a distinct effort after perspective effects, and it is possible that the study of some of these Roman examples by Brunelleschi and Donatello at the opening of the fifteenth century may have contributed to the formation of the picturesque tradition in Florentine relief sculpture of the period. This style was however by no means universal in Roman work. The carved sarcophagi are not much influenced by it, and these sarcophagi are of special importance in the later history of sculpture, in that they were the models used by Nicola Pisano and the other French and Italian sculptors of the so-called ‘proto-Renaissance’ of the twelfth and thirteenth centuries. In relation to antique sculpture as a whole the pictorial style is quite abnormal. The genius of the classical Greek relief is indeed totally opposed to that of the Italian reliefs represented centrally by those of Ghiberti. The difference is fundamentally one of material and technique. It is the same distinction that was drawn by Michelangelo in his letter to Benedetto Varchi, and noticed already in the Note on ‘The Nature of Sculpture,’ ante, p. 179, the distinction, that is, between sculpture that proceeds by additions, ‘per via di porre,’ and that which advances by taking material away, ‘per forza di levare.’ The normal Italian relief was in cast bronze and was necessarily modelled work. The classical relief was in marble and was essentially carved work, for the diversifying of a flat surface. When the Greek relief was in baked clay it was generally stamped from moulds and not modelled up by hand. The cast bronze relief in classical Greece may be said only to have existed in the form of small plaques for the decoration of vases or other objects in metal. The normal bronze relief in the ancient world was beaten up in sheet metal by the repoussé process.
In the case alike of the relief carved on the marble slab, or stamped in clay from moulds, or beaten up in the sheet of metal, the nature of the technique renders flatness of effect almost obligatory. The Greek decorative relief cut ‘per forza di levare’ in a smooth marble slab, that is most often one of the constructive stones of an edifice, naturally sacrifices as little of the material as is possible, and in all Greek reliefs, whether low or high, as much as possible of the work is kept to the foremost plane, the original surface of the stone. Again, a mould that is undercut, or at all deeply recessed, cannot be used for stamping clay, while the difficulty of relief work in sheet metal is greatly increased in proportion to the amount of salience of the forms. Hence the general flatness of the antique relief, observable even on the late Roman sarcophagi which served as models at the first revival of Italian sculpture. Whether the field of the relief is open or crowded, the objects all come together to the front.
How different are the conditions when the relief is modelled up by hand in clay or wax! Here the starting point is the background, not as in the carved relief the foreground, and the forms, worked ‘per via di porre,’ can be made to stand out against this with an ease and effectiveness which tempt the modeller to try all sorts of varieties of relief in the same composition or the same figure, and to multiply planes of distance till the objects on the foremost plane are starting out clear of the ground. There is direct evidence (see ante, p. 194) that in the first century B.C. the use of clay modelling as a preliminary process in sculpture was greatly extended, and Roman pictorial reliefs may themselves have been influenced from this side. There is no question at any rate that the Italians of the Renaissance surrendered themselves without hesitation to the fascination of this kind of work, and the style of it dominates their later reliefs. The contrast in this respect between Ghiberti’s second, or Old Testament, Gates, and his earlier ones which adhered to the simpler style of Andrea Pisano’s reliefs on the first of the three Baptistry Gates, is most instructive. Andrea’s reliefs are in character mediaeval, and the nearest parallel to them are the storied quatrefoils on the basement of the western portals at Amiens. It is worthy of notice how classical these are in style, and this is due to the fact that like Greek reliefs, such as the frieze of the Parthenon, they were cut in the constructive stones of the edifice _in situ_, and are in true stone technique.
This contrast of Greek and Italian reliefs furnishes a most conspicuous object lesson on the importance of material and technique in conditioning artistic practice. As was pointed out in the Introductory Essay, these considerations have not in the past been sufficiently emphasized in the scheme of education in design recognized in our Schools of Art, though in several quarters now there is a promise of better things.
THE PROCESSES OF THE BRONZE FOUNDER.
[§§ 55–69, ante, p. 158 ff.]
Vasari’s account of the processes attendant on casting in bronze is intelligible and interesting, though he had himself little practical acquaintance with the craft. Benvenuto Cellini on the other hand was an expert bronze founder and the account he gives of the necessary operations in the first three chapters of his treatise on Sculpture is extremely graphic and detailed, and may be usefully employed to amplify and explain Vasari’s notice. An expert knowledge of the founder’s craft was not by any means universal among the Italian sculptors of the Renaissance. Donatello did not possess it, nor did Michelangelo. In the case of the former this is somewhat remarkable, for Donatello was such a vigorous craftsman that we should have expected to find him revelling in all the technical minutiae of the foundry. We are expressly told however by Pomponius Gauricus that Donatello lacked this knowledge, and never cast his own works but always relied on the help of bell founders (Hans Semper, _Donatello_, Wien, 1875, p. 317). Michelozzo on the other hand, who worked with Donatello, could cast, and so could Ghiberti, A. Pollaiuolo, and Verrocchio, while Alessandro Leopardi of Venice, who cast Verrocchio’s Colleoni statue, was famed for his practical skill in this department. It was customary, when expert help in casting was required, to enlist the services of bell founders and makers of cannon, but Cellini warns sculptors against trusting too much to these mere mechanicians who lacked ingenuity and resource.
The following general sketch of the processes of casting will render Vasari’s account more easy to follow. A successful cast in bronze consists in a thin shell of the metal, representing on the exterior the exact form of the model, or the complete design in the artist’s mind. The best way to procure this is to provide first a similar shell, perfect on its exterior surface, of wax, and then to melt away the wax and replace it by molten bronze. For this to be possible the shell of wax must be closely sealed between an outer envelope and an inner packing or core. It can then be got rid of by melting, but care must have been taken lest the core when it loses the support of the wax should shift its position in relation to the envelope. To prevent this, metal rods are run, skewer-fashion, through core and envelope to retain them firmly in their relative positions. Molten bronze may then be introduced into the space formerly occupied by the wax, and this, when it is cold, and the envelope and core are both removed, will be the cast required.
Attention has to be paid to secure that there shall be no moisture and no remnant of wax in the parts where the molten bronze is to come, otherwise steam may be generated and a dangerous explosion follow. Similarly, air holes or vents must be provided, so that the air may escape before the flowing metal. The cast when cold should, theoretically, give a perfect result, but as a matter of fact, unless very accomplished skill or great good fortune have presided over the operations, the metal will be blistered or seamed or flawed in parts, and these imperfections will have to be remedied by processes which come under the head of chasing, and are described by Vasari at the close of the chapter.
A direct and ingenious method of procuring the needful shell of wax is that described by Vasari in § 67 as suitable for small figures and reliefs. Over the model for such a small figure an envelope is formed, in the shape of a hollow mould of fire-resisting material, so constructed that it can be taken to pieces, the model extracted, and the mould closed up again. The mould must now be cooled with cold water and it is then filled with melted wax. Contact with the cold sides of the mould chills the wax, which hardens all over in a sort of crust or skin. The rest of the wax, still liquid, is then poured out and the skin or crust suffered to harden. The interior is then filled with clay of a kind that will stand heat. Rods or skewers are passed through this and the envelope, the wax is melted out and its place taken by the molten bronze.
This process, one of course only suitable for small objects, presupposes the existence of a completely finished model to be exactly reproduced in the metal. The simplest of all processes of bronze casting dispenses with this model. Vasari does not describe this simplest method but Cellini, who employed it both for his ‘Nymph of Fontainebleau’ and his ‘Perseus,’ gives an account of it which is worth summarizing because the process is probably the one adopted in most cases by the old Greek masters.
Cellini tells us that in making his large lunette-shaped relief of the ‘Nymph of Fontainebleau,’ now in the Louvre, he began by modelling up the piece in fire-resisting clay, of course over a proper armature or skeleton. He worked it out to full size and then let it dry till it had shrunk about a finger’s breadth. He then covered it with a coating of wax of rather less than this thickness, which he modelled with the utmost care, finishing it in every detail so that it expressed to the full his own idea for the finished work. This was then carefully covered in successive layers with an envelope of fire-resisting material, which would be properly tied by transverse rods to the core, and braced on the exterior by an armature. The wax was then melted out, and the core and envelope thoroughly dried, when the molten bronze was poured in so as to reproduce the wax in every detail.
It is obvious that this is not only the most direct but the most artistic method of work. The wax forms a complete unbroken surface and receives and retains the exact impression in every detail of the master’s hand. If the cast be thoroughly successful, the bronze will reproduce the surface of the wax so perfectly that no further work upon it will be needed, and an ‘untouched cast’ will be the result. This method would suit the genius of the Greeks, and was no doubt commonly employed by them, but it has the practical drawback that if anything go wrong, and the bronze do not flow properly, the whole work is spoilt, and will have to be built up again _de novo_ from the small study. Cellini tells us that his ‘Perseus,’ which he was casting in this fashion, nearly came to grief from the cause just indicated, and he accordingly recommends what he calls the second method, a longer and less direct process, which has however the advantage that a full-sized completed model is all the time preserved.
This process is the one described by Vasari. A full-sized clay model is prepared and finished, and this is then covered with a plaster envelope made in numerous sections, so that it can be taken to pieces and put together again without the model, which may be preserved for further use. The next step is to line the inside of the empty envelope, or piece-mould, as it is called, with wax, and to fill up all the rest of the interior with a core. The piece-mould is then removed, and the surface of the wax is carefully gone over to secure that it shall be perfect in every part. Over it is then laid in successive coats a fire-resisting envelope between which and the core the wax is hermetically sealed. The wax can then be melted out and replaced with bronze. The piece-mould, which has been detached section by section from the wax, will serve again for other reproductions. The processes in which wax is employed are called _cire perdue_ processes, because the wax is got rid of in order to be replaced by the metal. The usual process in vogue at the present day dispenses with any employment of wax. The figure to be cast is piece-moulded and reproduced in a suitable material, a certain thickness of which is in every part pared away according to the thickness required for the bronze. This core is then replaced with proper adjustments within a fireproof mould, and the bronze is poured into the space prepared for it.
OF PAINTING
##