In selecting a subject, no matter whether from mythology, Scripture, history, fiction, or every-day life, care should be taken to choose one which has unity of action. There ought to be a story in your subject, but not more than one story. In your secondary groups you may have separate action and by-play, but they ought somehow to be connected with the main story of the picture, and instead of distracting the attention from the subject, they ought rather to assist in concentrating it. Where there is more than one centre of interest in a picture, the effect, dramatically speaking, is weakened.

The old masters often disregarded the tolerably self-evident rule.

The famous Transfiguration picture of Raffaelle is a well-known instance in point. The interest is divided between the Transfiguration proper and the demoniac boy. Although some of the figures are pointing upward, yet the faces are all turned toward the demoniac, and he is certainly the principal focus of interest.

This blemish in Raffaelle’s picture is all the more unaccountable, as no mention is anywhere made of a demoniac having been present at the time; but the old masters (especially those of the German schools) abound in incongruities of this kind. I remember seeing somewhere a picture of the “Martyrdom of St. Lorenzo.”

The saint is about to be roasted alive, but the largest and most prominent figure in the picture is one of the executioners, who is making a horrible face, having got some of the smoke in his eye. The introduction of these irrelevant and grotesque episodes cannot be justified, however well they may be painted; and if it be granted that it is undesirable to select a subject in which there is more than one centre of interest, how much more objectionable is it to invent disturbing incidents which are not recorded in the text of the subject.

As an extreme instance of a bad selection of subject, I have always thought that nothing could beat Shakespeare’s “Seven Ages of Man.” The lines suggest seven distinct subjects having no connection whatever with each other. Each is very good of its kind; to attempt to amalgamate them all into one picture is quite absurd. The result is extremely unpleasant; suggesting a company of strolling players, each rehearsing his part, or perhaps the court-yard of a mediæval lunatic asylum.

In justice to Mulready I ought to mention that he did not select “the seven ages of man” as a subject for his picture. He had the impossible task imposed upon him by a liberal but injudicious patron.

For decorative work (for a frieze, for instance) such subjects as the “seven ages of man” are well suited, because each “age” can be treated separately, forming as it were a picture of itself, the only bond of union between the seven being that the figures should be of the same proportion, and should be similar in style and execution.

Another good rule to observe in selecting a subject is to choose one which has illustrative action in it. What I mean by this is that the action of the figures should be sufficient to explain the subject. You cannot put words issuing from their mouths as is done in caricature, you must therefore explain your story by action and expression. We will take as examples two not dissimilar subjects. One shall be a meeting of conspirators, and the other a conference of philosophers. Of course, I don’t mean to insinuate that there is any analogy between philosophers and conspirators, but that in both cases we have five or six figures seated round a table. In the first we should represent our conspirators, in close conclave, leaning over the table with their heads near together, one or two perhaps grasping their daggers, another looking round anxiously—in short, it would be very evident from the expression and attitude of the figures that they were about some villainous work.

If we now turn to the other subject, the conference of philosophers, how are we to express the purport of their conversation? What facial muscles are called into play when men are talking metaphysics or expounding their theories of evolution? It is clear that, however exquisite the execution of the picture may be, the subject of it will be unintelligible, without explanation, and even with the necessary elucidation it will be inferior to the conspirators in dramatic interest.

The subject I gave you in the life-school some time ago (I mean Peter’s denial of Christ) is an eminently good one, because if properly treated it is impossible to mistake the meaning of the figures. The menacing interrogatory of the woman, Peter’s alarm for his personal safety, and the jeers of the soldiers who are sitting round the fire, are all well adapted for pictorial expression. Any one who had never read the New Testament, an unconverted Chinaman for instance, would say at once: “This young woman is taxing a middle-aged man with something he denies, but there is such downright assertion in her action and such fear mixed up with his denial, that the accusation, whatever it is, must be true.”

No subject can be called a really good one which requires a long explanation to make it intelligible. Thus subjects in which the figures are assuming characters which do not properly belong to them are unfit for painting. For example, in the “conspirators” just mentioned, it might very well have happened that to conceal their sinister designs they assumed the mask of joviality, but you should not select this particular phase of the story.

On the stage, this kind of make-believe is managed by an “aside.” The actor takes the audience into his confidence when he says, “Here comes the king, let us dissemble,” and accordingly for the next ten minutes or so you are to understand that he is not the obsequious sycophant he pretends to be, and lest by chance you should forget that he is dissembling, he will come forward and frown, clench his fist or point contemptuously over his shoulder at his fellow-actor, who, strangely enough, never seems to see these ominous gestures.

All this is understood and accepted on the stage, but it does not do in a picture. I would, therefore, advise you as much as possible to choose subjects which can be understood at a glance. Let your personages appear in their natural characters, and not assuming parts which do not belong to them.

Acts of mercy, such as clothing the naked, feeding the hungry, visiting the sick, etc., are all good subjects, because the meaning is explained directly by the action of the figures.

Speaking for myself, I have but little sympathy with subjects taken from works of fiction.

The artist who selects them for pictorial treatment seems to me to abnegate whatever creative power he may possess, and to become an illustrator or translator of other men’s thoughts. Homer, Dante, and Milton are of course exceptional poets. Their creations are heroic, and the personifications of their heroes would be either nude or sternly classical. Besides, they never descend to minute particulars, and the artist is left very much to his own invention. The more detail an author gives, and the more picturesque the detail, the less fitted are his works for figure-pictures. Scott and Dickens are eminently unpaintable; that is, it is a hopeless task to illustrate them. Pictures taken from their works are always disappointing. The Ivanhoes, the Mrs. Gamps, and the Pecksniffs of our imagination are always superior to their effigies on canvas, and this is more or less the case with the personages of Shakespeare, Cervantes, and Molière.

Costume has a great deal to do with the choice of a subject, and this, no doubt, is the reason why the works of Shakespeare, Cervantes, and Molière are such favorite hunting-grounds for artists. If the Prince of Denmark had been a modern heir-apparent, attired in a frock coat, tweed trousers, and a chimney-pot hat, or if Malvolio had worn the dress of an ordinary British butler, we should not often see them painted.

For one picture taken from Thackeray’s modern novels, we find dozens illustrating Tennyson’s “Idyls of the King,” or his “Holy Grail.”

Now, although the question of costume must always be an important factor in the selection of a subject, it ought not to be the only one. A picture should not be painted merely for the sake of the costumes. This seemed to me the principal fault in the large Austrian pictures of the International Exhibition; and I may add that it is a fault which is not altogether unknown in England.

There is one more class of subjects which I have not yet noticed, and that is the domestic or “genre” class; the pictures, in short, of every-day life. Pictures of this kind are much less dependent on a good choice of subject than those which illustrate some historical incident. They are generally prized for the brilliancy and harmony of their color, or for the delicacy of their execution; and if these qualities exist in a high degree, the subject is a minor consideration.

Still it ought to be a consideration, and in choosing subjects of this class you should prefer those which are typical of the personages you have to represent. If you attempt rustic pictures, not only should your figures look like peasants, but the subject should be thoroughly bucolic.

A dirty ploughman plodding wearily homeward along a muddy lane on a dull November evening seems commonplace and prosaic enough, and yet the subject would not be deficient in pictorial interest. It would be typical of the man’s hard and comfortless life. It would be in perfect harmony with his furrowed face, his bony limbs, and his stooping gait.

It would not only represent that particular ploughman in that particular lane, but it would give a true though mournful impression of farm-laborers generally.

I should much prefer for the subject of a picture, a common episode from the life of a laborer to an uncommon one.

Again, if I wished to represent the same man at home, I should endeavor, without exaggeration, to give the squalor of his surroundings, and should not, out of my inner consciousness, evolve an ideal peasant surrounded by a comely family, and looking (as Dickens has somewhere said) as if he had “spent his little All in soap.”

Artists understand pretty well nowadays that in painting rustic subjects, honesty is the best policy. The great success of the French painter, Millet, was due entirely to his uncompromising honesty of purpose, and to the unerring judgment with which he selected his subjects.

There are pretty girls (even in France) amongst the peasant class, although they are certainly rare. There are plenty of fête days when every woman makes herself as smart as she can; but Millet knew better than to paint pretty girls and smart dresses. Instead of this, he depicted the true types of French peasantry, gaunt, hard-featured women, dressed in the coarsest garments, and shod with wooden sabots. The novelty of truth was unwelcome at first to the Parisian public. They had so long been accustomed to Opera Comique peasants that they had lost relish for the genuine article; but by degrees they began to perceive that these uncouth figures were very like the Jeannes and the Victoires they knew à la campagne. Moreover, they did not fail to observe that the subjects chosen by the artist were of that homely, agricultural kind peculiar to the French peasantry. They smelt of the village dunghill, and this was the great secret of their success.

I am often told by people who don’t know much about art, that they have thought of “such a capital subject for a picture,” and it generally turns out to be something odd or incongruous, and not at all fitted for painting. For several years we have had pictures sent in for exhibition, representing children playing at judge and jury, police-courts, auctions, etc. In these pictures the children are all dressed up to represent policemen, barristers, plaintiffs, and defendants. Moreover, they have so thoroughly learned their parts that their action is no longer childlike. Some of these pictures are very well painted, but the principle is so wrong and false that we now invariably refuse them admission.

Children should, in a picture, be engaged on something childlike. Thus it would be perfectly natural for children to play at being wild beasts, making use of any bear or wolf skin which happened to be handy. Coach-and-horses, hen-and-chickens, are again legitimate games for children, and therefore proper for painting; but in the arts we don’t want elaborately got-up burlesque.

A group of young children on the sea-sands, at work with their wooden shovels, would by some be thought a stupid kind of subject, hardly worthy of being painted at all; but make the same children overtaken by the tide, with a steep cliff behind them, and probably you will have a great success, especially if you make your little figures expressing their fear or courage in a theatrical and unchildlike manner.

The first group would be a typical one—typical of the seaside and childhood; the second would not be absolutely impossible (like the bewigged and behelmeted youngsters above mentioned), but it would be somewhat exceptional, and therefore, in my opinion, not so suitable for painting as the first group.

In the same way with landscape, the spot you select for pitching your umbrella should not be mean and ugly, neither should it be overpoweringly grand and beautiful. Pictures representing the Falls of Niagara or the gorges of the Rocky Mountains are generally failures. I have in a former lecture praised the Belgian landscape-painters, and I think that a good deal of their merit lies in the happy choice of subjects. They are certainly not classical, like the old school of French landscape-painters, nor do they affect the dreariest commonplace, like some of the moderns. They neither paint precipices and snowy mountains, nor dull stretches of poplar-skirted high-road. Their pictures are to me most interesting, not only on account of their technical excellence, but from the good taste shown in the selection of the subjects.

Incidents which are out of harmony with the character of the persons engaged, form capital materials for caricature. The late John Leech showed the nicest discrimination in his selection of subjects. When he gave us pictures of character, nothing could be better than his sporting scenes, or his bits from the mining districts. When he wanted to raise a laugh at something paradoxical, he would give us a lot of mutes making merry after a respectable funeral, or a used-up swell eating periwinkles with a pin on the top of a ’bus. In both these cases it was the sharp contrast between the usual habits of the persons and their exceptional occupation at the time which made the fun, and very good fun it was too; but in an oil picture which takes some months to paint, the humor ought to be of a more delicate kind. I know of no better example of the kind of humor I mean than Wilkie’s “Blind Fiddler.”

Before closing my lecture I should wish to notice a certain kind of pictures which do not fit in well with any of the classes I have mentioned. The pictures I mean are those which are painted expressly to teach some lesson, or to inculcate some moral precept. The great originator of this kind of art was Hogarth. Before him nothing of the sort had ever been done, and since his death no artist has equalled him in this particular line. Much, however, as I admire Hogarth as a painter, I cannot coincide with all the praise that has been showered on him as a great moral teacher. He has often been compared to Molière, but the great Frenchman attacked the vices and follies of his day with a sharp rapier, whereas Hogarth wielded a heavy bludgeon. Indeed, I think it very doubtful whether our art can be converted into an active agent in the cause of morality. The touches of ridicule which a clever writer uses with so much effect are very apt to become ponderous when embedded in oil paint. Hogarth’s reputation may well be allowed to rest on his numerous technical excellences without hoisting him upon a pedestal as a great apostle of morality. In like manner the name of Cruickshank will be preserved as the clever draughtsman and caricaturist, and not as the champion of teetotalism.

In mitigation, however, of Hogarth’s sledge-hammer style of belaboring vice, we must bear in mind that the age in which he lived was a very gross and brutal one, and that his “Rake’s Progress,” his “Marriage à la Mode,” and similar works, which to us appear exaggerated or caricatured, were considered by his contemporaries to be very true to nature.

To return to the proper business this evening, which is not to criticise painters and their work, but to discuss subjects for painting, I cannot say I particularly delight in the class under notice.

Whoever takes up these subjects becomes (involuntarily perhaps) a kind of missionary agent for the cause he takes up, whether it be teetotalism, humanitarianism, or the redressing of the wrongs of our old friend, the “poor governess”; and as with some other agents, his zeal often outruns his discretion, and he is apt to thrust forward his moral too obtrusively. When this kind of picture is painted in pairs, after the fashion of Hogarth’s “Industrious and Idle Apprentice,” there is a sort of poster or advertisement flavor about the work, reminding one a little of “what I was, and what I am” in connection with Mrs. Allen’s hairwash, or of “before and after using anti-fat.”

No one can, of course, object to such antithetic pictures as “Summer and Winter,” “Peace and War,” “Youth and Age,” etc.; but where the practice of showing both sides of the medal becomes objectionable, is when the work is evidently intended to be didactic. I don’t know what effect these didactic pictures may have on others, but I always feel a kind of impatience at having the contrast between virtue and vice thrust before me in this infant-school fashion.

I do not wish in these lectures to enter upon the domain of high-art ethics; I have a very decided aversion to the union of painting with abstruse theories of all kinds, but a few words on morality in art may not be out of place.

It must be generally allowed that certain pictures have an immoral tendency: we may, therefore conclude by analogy that others have a moral tendency, but beyond this general truism it is difficult to get.

The art-loving portion of the public needs no Lord Chamberlain to ostracise immoral subjects, but on the other hand, it is rather intolerant of what are called “goody” pictures. Let us rather, instead of preaching homilies with our brush, endeavor to set an example of pictorial morality by adherence to truth, by abstaining from clap-trap tricks and meretricious execution; by ceasing to pilfer ideas and modes of painting from other artists, and by general honesty of purpose.

If we do this, we may rest assured that our work will have a healthier influence than it would have if more directly enlisted in the cause of morality.

LECTURE XI.

ON THE COMPOSITION OF DECORATIVE AND HISTORICAL PICTURES.

I ought perhaps to avoid using the word “laws”; art is not an exact science, and no strict law can be laid down about a matter of taste. Still there are certain principles which seem to be accepted by all masters of composition, and certain others which, although not generally accepted, occur to me as likely to be of use to you.

The golden rule for the arrangement of figures in a picture, is that the nature of the subject ought to dictate the lines of the composition. If you have to paint a subject of a quiet, majestic, and dignified class, a subject for all ages, where you wish to express perfect repose and stability, you cannot do better than go back to the pyramid. This pyramidical theory of composition has been much quizzed and laughed at, but that is because the old-fashioned dilettanti who advocated it wanted to apply it universally. Now it is clearly unsuitable for subjects of action, or for filling with figures low long panels; but for altar pieces, or for pictures which are destined for central places, it is at once the most natural and the most effective method. The quiet and serene dignity of many of the ancient Holy Families and other subjects of sacred art is due mainly to the pyramidical form of grouping.

Sometimes the form is that of a truncated pyramid, as in the Hemicycle at the Ecole des Beaux Arts, where the object of the painter was to represent an ideal Areopagus of Art.

In the compositions of Masaccio and Filippino Lippi we have good examples of a horizontal style of arrangement. The structure of these groups is suggestive of solid simple architectural forms, and has a kind of dignity of its own; but though suitable enough for frescoes of the fifteenth century, it is hardly picturesque or varied enough for modern oil-pictures.

Mural paintings, particularly when they represent grave or sacred subjects, should more or less partake of this horizontal and rectilinear form of composition.

A certain amount of deviation is necessary, and it is in fixing the limit of this deviation that the skill of the artist is shown.

Too little would make his composition formal and lifeless.

Too much would take away from the symmetry which befits such subjects.

The Stanze of Raffaelle are noble examples of skill and taste in composition.

Nothing can be finer than his “School of Athens,” his “Parnassus,” and his “Theology.”

Here we find variety of line combined with a dignified simplicity. It is the arrangement and composition of these grand frescoes which in my opinion justifies the position Raffaelle holds in the history of art, rather than the beauty of his Madonnas or the bold drawing of his somewhat over-rated cartoons.

Later artists of the Roman school overdid the picturesque element, and lost the stately simplicity which characterizes the second manner of Raffaelle.

In modern times, Ingres’ “Apotheosis of Homer” is a good example of what a mural painting should be. Severe in drawing and dignified in composition, it is yet by no means deficient in those more attractive qualities which are commonly expressed by the word “picturesque.”

Flandrin’s frieze at St. Vincent de Paul is another magnificent specimen of an exquisite sense of fitness. There is hardly a figure in the whole procession of apostles, saints, and martyrs, which could be improved. I know of no modern work which is so perfect of its kind.

It is, of course, preposterous to suppose that good composition can be taught in a couple of evenings; but if I succeed in impressing on you the importance of this rather neglected branch of study, I shall not have lectured in vain.

We will begin with the simplest problem, namely, how to fill up with figures an elongated rectangular space or frieze.

The most obvious method is to set up a row of figures of the same size and all in profile, as was done by the ancient Egyptians. Now this mode of treatment, though suitable enough for the Nile temples, would obviously be unfit for buildings of the nineteenth century.

The figures should preserve a certain regularity, a certain frieze-like arrangement, and yet the attitudes should be varied, and the work should not look as if it had been done by machinery.

The first improvement on the Egyptian method would be to break the monotony by here and there grouping two or even three figures together. As in these groups of two or three one figure must be behind the others, and therefore farther off from the spectator, it would be smaller, the head would be lower, and you would at once get a little variety in the line of heads. To vary your line of heads simply by arguing that some people are six feet high whilst others are only five, does not answer in decorative painting. You may assume that the male figures are bigger than the female, but you must proceed as if your men were all of the same (or very nearly) of the same height, and your women ought also to vary very little in stature. Children you may introduce of any size, from the infant in arms to the youth or maiden of fifteen; but let them be unmistakably children, and not little men and women.

The individual action of the figures would of course depend very much on the destination of the work. If it were intended for the decoration of a church, the figures would of course represent patriarchs, apostles, or martyrs, and a severe and simple arrangement would be necessary. If, on the other hand, your frieze were to decorate a theatre or ball-room, the figures should have more action, and naturally the lines would be more broken. Whatever the subject, however,—whether maskers, musicians, or morris-dancers,—there should be a certain frieze-like symmetry in the composition. You should never forget that you are engaged on a decorative work, and not on an easel picture.

A rule which it is well to observe in all decorative work is to avoid cutting off any portion of the figures. This is quite unavoidable in many easel pictures. If you have a crowd of people to represent, you cannot isolate some of them so completely that no portion of the others should be visible.

An easel or gallery picture is bounded on every side by the frame, and the eye is not shocked at all by seeing portions of the figures cut off. Although every one knows that the figures do not extend behind the frame, yet it is easy to suppose that they do, and the eye allows itself to be cheated into this belief; but in decorative or mural painting there is no solid framework round the picture isolating it from the surrounding wall. There may be an ornamental border or possibly a light moulding, but this is not enough to permit the practice. Michael Angelo, in his decorations of the Sistine Chapel, often carried his figures and draperies right out of the panels allotted to them, and this boldness adds to the grand, free character of the work.

The problem of how to fill up the irregular-shaped wall spaces which continually occur both in Gothic and Palladian architecture is of course more complex.

These spaces have generally curved sides, and in many of them—as, for instance, the spandrels of arches—the curve is concave. Straight horizontal lines of heads which are generally so desirable for long rectilinear spaces here become very objectionable.

Bold convex outlines for the groups, and an arrangement for the heads which does not suggest either horizontal or vertical lines, ought to be the rule in these cases.

Nothing can be finer than M. Angelo’s treatment of the sybils and prophets in the Sistine Chapel. There is a majestic dignity about them which is due rather to their full convex outlines than to their colossal proportions.

On the other hand, in many of the compositions by the early Florentines we have long horizontal rows of heads which seem out of harmony with the arched space they fill. The circular nimbi take off somewhat from the meagre character of these lines, and there is considerable beauty about the individual figures, but viewed as decorative works they are very unsatisfactory.

It is, of course, impossible to devise rules for all conditions of decorative and historical painting, but a few general hints may be useful to you.

1st. Beware of concave lines for the outlines of your groups.

2d. Avoid sharp angles, and particulary right angles, unless you wish to draw special attention to them.

3d. Be very careful about the relative position of the heads, so that viewed as points of interest they do not form any regular geometrical pattern.

These three rules seem to me the most important ones to be observed in the composition of decorative figure-pictures, and we will examine them seriatim.

The first rule I have given is to avoid concave forms for the general outline of the groups. There is no rule without exceptions, and to this one there are a good many; still it will be found that, speaking generally, convex outlines give grandeur wherever they are introduced. This convexity in the form of the groups need not be dependent on the outlines of the figures themselves; it may be got by introducing drapery or other accessories. I know of no example showing the value of full convex outlines more strikingly than the Madonna di S. Sisto. In the pictures of the Umbrian school, on the contrary, we find extreme poverty of line. The figures themselves are not particularly attenuated, but they are not sufficiently connected together nor enveloped in those useful pieces of flowing drapery which give such grandeur and fulness to the works of Fra Bartholomeo, Sebastian del Piombo, and other painters of the Roman school.

I have in former lectures entered fully into the subject of convex lines being almost always associated with forward movement, and concave with retreat, and need not go over the same ground again.

I would, however, observe that the terms boldness and convexity are almost synonymous when applied to outline; thus when we speak of a mountain having a bold outline, we mean that though steep and precipitous it is bluff or convex in form. A mountain with a depression on the top, or surmounted by a sharp-pointed cone, would hardly ever be noted for its bold outline.

The second rule to which I wish to call your attention is the avoidance of right angles in the composition of your figures.

All angles, unless they be obtuse ones, are to be deprecated, but the most objectionable of all are the right angles. In a single figure, rectangular outlines are not so unpleasant, but I cannot agree with those who think that the big seated Egyptian figures with which we are all familiar, owe their grandeur to their rectangular contour. I have no doubt but that these gigantic figures in their native swamp, and illumined by an Egyptian moon, would look very imposing, but the solemn grandeur of their aspect would be due to their size and to their surroundings, and not to their harsh rectangular outline. If the “Moses” of Michael Angelo could be magnified to the size of these figures and transported to the banks of the Nile, I fully believe he would be far more impressive.

Simplicity and grandeur are often bracketed together as though the terms were almost synonymous; but they certainly are not so. The street and chapel architecture of the Georgian era was simple even to baldness, but no one can call it grand.

It is not, however, in single figures that right angles are so much to be avoided, as in complicated groups of several figures. Here they arrest and distract the eye, giving harshness to the composition, and destroying the look of spontaneous action and easy-flowing movement which figures always should have.

Rubens in his “Descent from the Cross” has avoided these disagreeable angles, but in many of his more careless compositions, where there is violent action, they are painfully conspicuous, in spite of his liberal use of flying draperies. Hence his cavalry skirmishes, though full of violence and contortion, are quite wanting in spontaneous “go.”

Right angles in a group of figures convey the idea of immovability. Hence, although generally undesirable, it is well sometimes to introduce them.

Thus a kneeling warrior firmly planted to resist onslaught might with propriety have both knees right angles.

In this case we wish to give the idea of fixture, and therefore rectangular forms are not only allowable but very useful. Again, in the case of a wounded man endeavoring to raise himself, the angle formed by his right arm might with propriety be a right angle, because we want to show that the man is wounded and cannot raise himself without difficulty.

If he were uninjured and in full possession of his strength, we ought to represent his springing up in some other way.

In the very frequent case of two arms crossing each other, they should not cross at right angles.

There is no reason here for expressing immovability at the point where the arms cross, and therefore the formality of right angles should be avoided.

We will now pass on to the third rule, namely, that relating to the heads of the figures.

Whatever the subject of the picture, the eye is always attracted to the heads. It is therefore of the highest importance that their relative positions should be carefully considered.

In the annexed diagram, it is of no use arguing that one of the heads is a full face, another three-quarters, a third a profile, and the fourth a back view of the head. The four heads are all points of interest. They are equidistant, and placed on a segment of the same circle, and, turn them whichever way you will, you cannot get rid of the unpleasantness of the arrangement, so long as you keep them in their present relative positions.

In the next figure we have four heads suggesting a quadrilateral of lozenge shape.[4] This is also very objectionable, and it is a case of frequent occurrence. In both these diagrams, by shifting the position of one of the heads, we should break up the symmetrical arrangement which so much offends the cultivated eye.

There is no objection, in a composition of many figures, to placing two or more heads on the same horizontal line. Indeed, in many cases it is most advantageous to do so; but what ought to be avoided is having heads on the same vertical line. If you have a kneeling or sitting figure in front of an erect one, arrange your kneeling figure so that the one head shall not be perpendicularly below the other.

If you have two erect figures, arrange your kneeling figure so that the head shall not come on the same vertical line as either of the other heads, or half-way between the two.[5] I might urge a good deal more about the extreme importance of a picturesque and irregular arrangement of the heads; but I have probably said enough to call your attention to this very prominent feature in good designing, and will now give a few hints about other kindred matters.

Converging lines are to be avoided, unless there is something of interest to which you wish to direct attention at the point of convergence. This is by no means an exaggerated specimen of the evil; but the effect of these four arms all converging toward one point is unpleasant. If the personages were disputing over a manuscript, or trying to clutch a bag of gold lying on the table, then the manuscript or the gold would be the centre of interest in the picture; and converging lines would not only be excusable, but absolutely necessary. Where there is nothing of particular interest at the point where the lines meet, the eye feels disappointed at being misled.

Although converging lines are generally to be avoided, it often happens that a repetition of the same kind of curve gives force and unity of purpose to a group. Observe the convex curves formed by the backs of these suppliants. Their repetition gives unity of purpose. A perpendicular kneeling figure might individually be just as expressive, but as one of a group he would take away somewhat from the general character of unity in supplication.

One of the most difficult problems the designer of large mural pictures has to solve, is to introduce with good effect raised arms and hands, especially when they belong to the background figures. When possible, it is better to keep them out of sight altogether; but in some subjects you would by so doing inevitably lose expression and animation, and it becomes necessary to introduce here and there an upraised arm with extended hand. This is easy enough to do if you are reckless about the lines of your composition, but if you are fastidious, it is a very difficult problem.

In the first place, they distract the eye, destroying the full bold outline of your groups, and, secondly, there is a comic element about them which it is rather difficult to avoid. When, as in many of Raffaelle’s Loggie, the whole of the figures which are raising their arms are seen, the effect is bad and trivial; but there is nothing particularly comical about it. When, however, an arm crops up here and there from the unseen figures of the background, it is difficult to avoid the ludicrous. Cases may occur when a whole forest of hands will have to be raised, as in an oath of allegiance; but here the action of raising the arms is inseparable from the subject.

My remarks apply only to upraised arms as indicative of wonder, joy, or grief.

All these hints about designing may appear to some of you rather far-fetched, but if ever you get experience in decorative painting, I think you will find they are not far from the truth.

The art of good grouping is not of spontaneous growth. You may have a general idea of how you are to fill your canvas or wall-space, and that idea may be a good one, but all the details of the groups have to be worked out bit by bit. A change in the attitude of one figure will be almost sure to entail a change in a good many others, and it often happens that, after giving yourself a good deal of trouble, you will have to go back to your first idea.

A conscientious and fastidious designer may be compared to an Arctic explorer picking his way in an ice-pack. He will have to saw through one ice-barrier, to blow up another with gunpowder, to circumvent a third, and when, after surmounting all these difficulties, he thinks his course clear and open water at hand, he may have to retrace his steps and seek some other channel.

I am perfectly aware that in painting small easel pictures all this groping after fine lines may be unnecessary, nay, even detrimental to the life-like spirit of the composition.

“Our own correspondent’s” sketches at the seat of war (if done on the spot, which I am afraid they not always are) will be not only more interesting but better composed than if he had sat at home and trusted to his imagination; but in this lecture I am not dealing with easel pictures and realistic subjects, and I repeat that in decorative figure-painting excellence can only be obtained by a continuous process of altering, modifying, adding, and omitting.

In the same way that the lines and general grouping of a picture should be arranged with a view to expressing the subject with dignity and grandeur, so the management of light and shade should tend toward the same end, and it is as impossible to lay down strict rules for light and shade as for outline designing. Didactic writers on art will tell you that the principal light ought to fall on the principal figure—

Sir Joshua Reynolds remarks very justly on this piece of doggerel, that there is no necessity for the principal figure to be placed in the middle of the picture, or receive the principal light. He goes on to say that “this conduct, if always observed, would reduce the art of composition to too great a uniformity,” and that “it is sufficient if the place he holds, or the attention of the other figures to him, denote him the hero of the piece.”

In works which partake strongly of a decorative character this axiom about “Fair in the front in all the blaze of light” for the principal figure may be tolerably true, but in historical pictures something more unforeseen is wanted.

In the often-painted subject of the “Death of Cæsar,” I should be very much inclined to put the Cæsar in the shade, and the tyrannicides with their flashing daggers in the light. It appears to me that to throw a shade over the face of the prostrate emperor would somehow or other convey the idea of the shadow of death, which is overspreading him, and the reproachful “Et tu Bruté” would come with greater pathos from a figure half-veiled in shadow than from one in broad daylight. We will suppose now that instead of having the death of Cæsar to paint we have the “Stoning of St. Stephen.” The subject is analogous. The young man named Saul and the Jewish executioners of Stephen were not common assassins any more than the murderers of Cæsar. Shall we, therefore, adopt the same plan with the figure of Stephen as we did with that of Cæsar and put him in the shade? I say, Certainly not. Stephen was the first Christian martyr. We read that his face was as that of an angel, and he ought to be surrounded by an angelic halo of light, and this treatment need not be dictated by the text. We should come to the same conclusion simply on the grounds of pictorial fitness. Stephen was a voluntary martyr, and gloried in his own death. Cæsar was assassinated much against his will; and although we are told that he covered his face with his toga and died with dignity, yet he certainly cannot be called a martyr.

I have introduced these two subjects to show you how hopeless it is to attempt to lay down general rules such as old Du Fresnoy gives us in his poem on the art of painting. Every new theme you undertake to illustrate ought to have a treatment special to itself if you wish to produce a fresh and original picture. When the master of a vessel is starting on a voyage, he would not steer S. W. by W. ½ W. because that happened to be the course he steered the last time he was at sea, nor would he run up his skyscrapers and set his studding-sails because he carried all his light canvas the last voyage out.

He would consult his chart, the state of the tide, the direction of the wind, and act accordingly. In short, for this new voyage, the condition of the wind, tide, and barometer being new, he would give new orders to his mate and crew.

Substituting the brain for the master, the hand for the mate, and the brushes for the crew, we ought to set about our pictures much in the same way.

After giving the subject of light and shade a good deal of thought, it appears to me that there is only one rule which invariably applies to all pictures, and that is, that there should be a uniform scale of tone throughout the work. The gradient, from light to shade, may be very steep as in Rembrandt, or very gentle as in P. Veronese, but this gradient or transition should not be abrupt in one part of the picture, and gentle in another. The whole work (whatever scale you adopt) should be homogeneous.

Sir. J. Reynolds and others have endeavored to ascertain the proportion of light to shade in the works of the old masters. I believe these experimental blots have been made rather with a view to black and white than legitimate light and shade; but whatever their object, I don’t think that any theory can be built up on them. I am convinced that what the old masters called the chiaro-oscuro of their pictures was a matter of feeling, and sometimes of accident, but never of calculation.

Theorists often talk learnedly about secondary and tertiary lights, but the artist never dreamt of them. They are nothing more than the efforts he has made, and the means to which he has resorted, in order to connect the highest light of his principal group with the gloom of his background.

Rembrandt’s vigorous light and shade and Correggio’s luminous breadth ought to be ascribed to the natural idiosyncrasies of the painters, intensified probably by the conditions under which their works were executed. They were assuredly not the results of calculation or learning.

Modern artists are often credited by their critics with subtleties of which they are perfectly innocent. They introduce into their pictures certain harmonies of tone or color by a kind of pictorial instinct, but certainly not in obedience to theoretical laws.

In designing a composition of many figures, it is natural to begin with the principal group or centre of interest. When you have got this satisfactorily arranged, you proceed with the less important figures, and it is here that beginners (and some who are by no means beginners) often come to grief.

They get a fine action or a noble attitude for some accessory figure, and they are so much in love with it that they must introduce it, whether it is in keeping with the principal group or not. It may (viewed as a single figure) be very good, and yet be injurious to the general harmony of the composition.

Recollect that accessory figures, however good in themselves, if they mar the general effect, ought to be sacrificed.

By so doing you will doubtless raise a cry of lamentation from your friends. They will say, “What could have induced you to have scraped out that figure? Why, it was the best thing in the picture,” and so on. To which you might reply that you did not want it to be the best thing in the picture, and therefore you erased it.

It was this tendency to introduce some favorite figure where it was not wanted, which rather mars Raffaelle’s latest manner, as exemplified in the “Transfiguration,” and in the “Incendio del Borgo”; and what in Raffaelle was only an incipient tendency became a confirmed habit in the work of his imitators.

Sir J. Reynolds, in his discourses, is continually urging the student of composition to think how the old masters would have treated the subject he is engaged upon, and advises him to imitate their style and manner. Indeed, the sixth discourse is devoted entirely to this principle of imitation. Now, if we were vastly superior to M. Angelo, Raffaelle, and Titian, and held the same relative position to them that they did to their predecessors, I could understand our occasionally adopting their figures, after greatly improving them; but as we should not be likely to improve any figure we had appropriated, we had much better leave the old masters alone. Plagiarism, or, to use a plainer word, “stealing,” can only be excused when the plagiarist makes a better use of the property he has appropriated than the original possessor did.

Sir Joshua certainly says that you should “imitate,” and not copy servilely. Imitation is the sincerest form of flattery, and if Philippino Lippi could have seen Raffaelle transferring his St. Paul into the famous cartoon of the saint preaching at Athens, no doubt he would have felt flattered. But how about Raffaelle? Is it not true that this plagiarism on Raffaelle’s part detracts somewhat from his fame? Does not every one, on seeing the Carmine Chapel at Florence, and recognizing the familiar figure of St. Paul, think somewhat more of Philippino Lippi and somewhat less of Raffaelle? I believe that nothing can be more fatal to the career of an artist than intentional imitation of another man’s work. I say “intentional,” because we are all more or less imitators quite unconsciously. We often confound a reminiscence of something we have seen in a picture with a reminiscence of nature, and so become unconscious imitators; but this is a very different thing from deliberately setting aside our own ideas and endeavoring to fancy what would be the ideas of some one else.

It may be argued that Sir J. Reynolds addressed his advice to students and not to mature artists, but the habit of imitating others, when once acquired, is not easily got rid of. A certain degree of excellence may doubtless be attained by following this method, provided the masters imitated are excellent, but, after all, it is only a kind of reflected light, and not to be compared to the electric light of original genius. Besides, the student who follows Sir Joshua’s advice may begin by honestly attempting to paint his pictures in the style of Raffaelle without downright imitation of the figures, but he soon learns to adopt Raffaelle’s attitudes, Raffaelle’s expression, and even Raffaelle’s mannerisms. He becomes, in short, a mere copyist. If this be deplorable in the case of the imitator of Raffaelle, how much more deplorable is it to adopt the modes of thought and expression of an inferior master! It may be thought by some that in these lectures I often speak disrespectfully of the old masters, but it is certainly not my intention so to do. I have the greatest respect for many of them, though not for all; but I respect nature and truth still more, and it appears to me that the true artist should go to the fountain-head for his ideas and inspiration, and not to second-hand sources.

It may be answered that it is all very well saying that an artist should go to nature and rely on his own powers of creation and invention, but supposing he is relying on a broken reed; suppose he cudgels his brain in vain for ideas, what is he to do? In this case I should advise him, instead of borrowing from the old masters, that he should turn his attention to portrait-painting, landscape, or some branch of the profession where the creative and imaginative faculties are not much required. He may have great imitative power with a dexterous execution; he may be a charming colorist, or, again, he may be a refined and accomplished draughtsman, and yet be totally unable to give dramatic vitality to a scene he has not himself witnessed.

It has always been the fashion to apply the term “high art” to heroic or Scriptural figure-subjects, but I think there is almost as much high art in a noble portrait of Titian or a fine landscape by Claude as in any historical painting whatever. I object to the term altogether; but if it means any thing, it ought to mean a dignified and poetic view of nature, in contradistinction to a trivial or prosaic view. It ought certainly never to be applied to a pasticcio of the old masters, however plausible such an imitation may be.

In my opinion there is high art in Turner’s early pictures, because in them we get the man’s own poetic interpretation of nature, but in those works where he attempts to rival Claude I can see nothing but the labor of a skilful imitator.

I have wandered away from the proper subject of this lecture, and have but little time left; but before concluding I should wish to explain that although I am continually urging the extreme importance of originality in painting, I do not mean forced singularity or oddity. I mean by the word, the expression of the painter’s own sober ideas. A sane man should produce sane work. It may not be very powerful, it may in no way recall Michael Angelo, but it will have qualities of its own. How charming, simple, and unaffected are Flaxman’s designs until he got inoculated with the Sistine Chapel lymph! After this inoculation we notice (at least I do) a great change for the worse in his compositions. To graft successfully, the parent stem ought to be of the same nature as the scion or graft. Now Flaxman’s nature was gentle, and very appreciative of beauty and grace. With such a nature he ought to have abstained from attempting the grand and the terrible.

If Flaxman erred in grafting Michael Angelo’s manner on his own, what shall we say about Blake? Flaxman was at any rate a good draughtsman, but Blake’s ignorance of the first principles of drawing makes his Michael Angelesque imitations simply ludicrous. The successful attempts which have been made of late years to rehabilitate Blake, and to elevate him into a kind of British Michael Angelo, make me almost despair of high art in this country. I do not wish to speak contemptuously of Blake as a poet, but in his pictures (even supposing he had grand ideas) I cannot accept the will for the deed. The frog in the fable had grand ideas when he wished to rival the ox in size, and yet he only made himself ridiculous. Were I to express all I think about the Blake revival, I could hardly confine myself to parliamentary language. I will, therefore, in closing my lecture, simply protest to the best of my power against this strange infatuation.

LECTURE XII.

COMPOSITION OF INCIDENT PICTURES.

We are far more particular now about the arrangement, or what the French call the mise en scène, of a picture, than the old masters ever were. We may not be able to paint like Titian or Correggio, but we attempt an approximation to truth which they never did; and not only is a modern historical painter more truthful about the costumes of his personages and the architecture of his backgrounds, but in the disposition and action of his figures he honestly endeavors to represent the scene as it actually may have occurred. When I say that the modern painter does this, I mean that in my opinion he ought to do it. I am quite aware that many artists prefer to look at nature through the spectacles of the old masters, but it appears to me that all art should be in some measure representative of the age in which it exists. When we come upon a Romanesque, Umbrian, Venetian, Flemish, or eighteenth-century work of art, we can tell at a glance to what period it belongs, and I think that our own time, being one of original thought and research, should in some measure be similarly reflected in our painting.

I have no objection to Gothic architects repeating in modern buildings the narrow staircases, the dim lighting, and other inconvenient peculiarities of the style.

Were they to give us large plate-glass windows and noble flights of steps, they would cease to be Gothic architects; but I don’t think that, however much we painters may admire the old masters, we ought to adopt their modes of composition when we know them to be the result of ignorance, error, or carelessness.

The present graphic method of treating figure-pictures is of quite modern growth, and the innovation extends to all kinds of subjects. Compare any of Giulio Romano’s, Rubens’, or Lebrun’s battle-pieces with those of Raffet, Horace Vernet, or, better still, De Neuville. How unreal the old masters appear!

Recall to mind the Romans of David and his school, and compare them with the best modern representations of Roman manners and customs. In the one case we may admire the noble drawing and even the classical lines of the composition, but we are never transported back to the scene; whereas in certain modern pictures we feel on much more intimate terms with the personages. We fancy we are actually a spectator at the Colosseum or a participator in a fête intime.

The realism of modern art is due partly to a greater knowledge of, and a greater attention to, costume, architecture, furniture, and all the properties of the stage on which we place our personages, but it is also due to our making truth a primary object.

An incident may be treated truthfully in fifty different ways, but some of these versions of it will be dull, some obscure, and some vulgar, and it is for the artist to select a rendering which, though perfectly truthful, shall be neither dull, obscure, nor vulgar. As soon as he loses sight of truth he ceases to be a realistic painter. He may produce a beautiful picture, but it will partake more or less of what I call semi-decorative work. It is sometimes very difficult to fix a boundary-line between realistic and decorative painting. To which class, for instance, belong the cartoons of Raffaelle? Although designed for tapestry, and therefore for decorative purposes, there is too much truth and reality about them to allow of their being classed among purely decorative works; whilst, on the other hand, we can hardly admit that they are like the scenes they are meant to represent.

The heads are Italian rather than Jewish or Oriental, and sometimes (as, for instance, in the “Miraculous Draught of Fishes”) pictorial liberties are taken which are quite inadmissible in realistic work.

I may here observe that in this lecture I shall not use the word realistic in the bad sense in which it has generally come to be used.

The term is now generally employed to designate some ugly or offensive piece of reality which is prominently thrust upon our notice by the artist; as when Quintin Matseys gives us wrinkled and abnormally ugly old men, or when a modern French painter throws all his talent into depicting the thick viscosity of a pool of arterial blood. Reality is only in rare instances repellent, and I can see no good reason for confining the word to these exceptional cases.

In historical, or what may be called incident pictures, the main object of the artist ought to be to tell his story forcibly, clearly, and pathetically.

We have seen that in work partaking of a decorative character the principal object of the designer should be to group his figures in a noble and picturesque manner, to attend to his drawing, and if possible to add the charm of agreeable color to his work.

In realistic historical painting he has something else to occupy his thoughts. He must by no means neglect the lines of his groups, he must avoid disagreeable angles, equidistant heads, convergent lines where they are not wanted, and all the other rocks and shoals on which many a composition has been wrecked, but in addition to this he must tell his story truthfully and clearly.

Much more latitude in the matter of arrangement may be allowed him than would be conceded to the painter of decorative subjects.

He may (if he thinks fit) huddle up all his figures into a corner of the canvas, or he may place them all in the centre, leaving the sides unoccupied.

In short, he may take great liberties with the laws of composition, provided always these liberties tend to assist in giving reality to the scene.

The more picturesque or melodramatic the subject, the more he may depart from the usual rules of composition.

Paul Delaroche was, I think, the first of the numerous cohort of modern painters who have striven to combine truthful sentiment with pictorial fitness, and of all his works the “Assassination of the Duc de Guise” is perhaps the most striking.

The arrangement of this picture is as dramatic as it is truthful. On one side of the picture we have the murdered duke lying on his back, stone dead. The group of assassins are quite separated from their victim, and are giving themselves no further trouble about him; and yet the greatest ignoramus, who knew nothing whatever about the story, would have no hesitation in divining it, so graphically is the incident told.

Again, if we recall to mind another and a better known picture by the same master, I mean that known as “Les Enfants d’Edouard,” we find the same subtle taste displayed.

I may here note that the color of neither of these pictures is in any way remarkable. Indeed, that of the “Princes” is positively bad, being very purple and inky; but their enduring popularity rests on a more solid foundation than mere color. It rests entirely on their truthful and poetic treatment. I call the treatment “poetic,” because a dull prose reading of both these subjects would have represented the murders as actually being committed, whereas by choosing the moment in the one case immediately after the murder, and in the other just before, the artist avoids all the stabbing, hacking, and smothering business, and increases rather than diminishes our interest in the victims. Gerome’s “Death of Cæsar” is another example of novel treatment of a hackneyed subject. He also represents the deed as done. The conspirators have sneaked off. The benches of the senate-house are all but deserted, the only occupant being a very fat senator, who is fast asleep on one of the benches, somewhere near the centre of the amphitheatre. How much more empty the senate-house looks, with this portly old Roman snoring on his bench, than it would do if entirely deserted!

I do not wish to lecture on modern pictures, but I mention this “Death of Caesar” by Gerome as an instance of a happy departure from the usual treatment of the subject. Indeed, it appears to me that all assassinations, martyrdoms, executions, and such-like subjects, if painted at all, should be approached in some roundabout way.

The action of stabbing, cutting a head off, or sending a bullet through a man’s body, is instantaneous; and although an executioner, with his drawn sword and uplifted arm about to decapitate his victim, may be startling and sensational at first sight, yet after a time the feeling of horror or of pity gives place to a sort of impatience that he is so long before striking the blow.

One of the Orleans princes had a picture of a military execution, which he admired very much at first. By and by, however, he got tired of it, and ultimately sold it or gave it away, not because it was too much for his feelings, but because he was heartily sick of seeing the squad taking aim day after day and month after month, and never firing.

Although the best modern masters of dramatic composition have probably been guided by sentiment rather than by rule, still a few observations on the treatment of certain subjects may not be out of place in this lecture. Thus, if the subject be a departure of pilgrims or emigrants, the figures should be placed on that side of the canvas which is opposed to the direction in which they are going. If it be an arrival, they should be placed on the side opposed to the direction whence they came. In both these cases, the large portion of canvas without figures is not wasted; it assists materially in telling the story.

In the first case, it represents the journey to be undertaken, and in the second the journey just performed. If we had to paint a shipwrecked sailor who has just reached the shore, we ought to let very little of the shore be seen, but plenty of raging sea. Here the interest of the subject lies in the formidable dangers he has escaped, so we ought to devote the greater portion of our canvas to the breakers, and relegate our mariner and the bit of slippery rock to which he is clinging to a corner.

If, on the other hand, we wished to represent our shipwrecked man clinging to a spar in the open sea, with no land visible, we ought to place him right in the middle of the canvas, so as to give the impression of hopeless isolation; and if we wished to convey the idea that he might possibly be rescued, we would paint a sail on the horizon, and near the edge of the picture. I should place it near the edge, in order that it might appear to have just come in sight, and that hope of rescue was dawning. If we were to put the same vessel in the middle of the picture, and bearing down upon the drowning man, we might feel equally certain that he would be saved, but the effect would hardly be as dramatic.

Again, let us suppose that we have an elongated space to fill, and that the subject is a “fugitive escaping.” Where ought we to place him on the canvas? If we place him in the middle, he will look too much like a professional runner doing his ten miles within the hour, and we should feel inclined to pull out our watches and time him. Supposing him to be running from right to left, if we place him near the right side of the picture we shall not know whether his pursuers are not close at hand, and as our sympathies are always with the fugitive, whether he be a prisoner of war, a convict, or a fox, we should be glad to see him safe over to the other side of the picture.

If we place him near the left edge our wish is gratified. There is now the whole width of the picture intervening between him and any sign of pursuit, and we feel naturally, though perhaps illogically, that he has a better chance of escape.

The word “artful” has come to signify cunning, and is always taken in a bad sense, but I suppose that originally it meant literally “full of art,” full of that curious compound of observation, good-sense, and poetic feeling which is so noticeable in Raffaelle, Poussin, and all the great masters of composition.

In the examples I have given you there has always been some good reason for placing the figures on one side of the picture, but where no good reason exists, it ought not to be done.

It may not be out of place here to say something about the size of the figures in proportion to the canvas. This is a very important element in the composition of a picture, and many a good and careful work has been spoiled by the figures being either too large or too small for the canvas.

In these days, when the general destination of pictures is to decorate dining-rooms or to fill small galleries, space ought to be economized. We should avoid, as a rule, large areas of background; but, on the other hand, when the figures are too large for the canvas the effect is very unpleasant. An erect figure with the head bent down should have space enough above it to allow of the head being raised, otherwise the figure has an uncomfortable look, as if she could not lift up her head without rapping it against the frame.