In 1849 the young sculptor Jean-Baptiste Carpeaux, who was later to execute the famous group La Danse for the façade of the Paris Opera, regretfully abandoned his studies with François Rude, the greatest of French romantic sculptors (and a major figure in Hugh Honour’s Romanticism). Carpeaux went to work with Francisque Duret, a less inspired teacher, but one much more representative of the doctrines of the Ecole des Beaux-Arts. He gave the reasons for his break with Rude in a letter to a friend:1

Mr. Rude, as I must have told you in my discussions about art, has a method which consists of a way of doing sculpture quite different from that of the Beaux-Arts. For example, he claims that sculpture—and I think he is right—can be done well only by mathematical aids: that is, using compasses, rulers, plumblines, etc…. Finally, he turns his students into practitioners, while at the School, everything is done only by eye.

The Institute prefers to say that the arts in general exist more in emotions than in measurements, which make us cold copyists with only the material spirit of what we are doing, while they…. I think what they say is very pretentious, but still I have to listen to them, and to talk and do as they do. And since there is a great difference between these two methods and one cannot serve two masters at once, I said to myself: let’s make a choice, the School is a question of my future and my existence….

At first sight, this may appear to stand romanticism on its head. For Carpeaux, it is the official academic doctrine which asserts that art is essentially a matter of emotion; by contrast, Rude, the romantic artist, creator of the violent, gesticulating figure of the Marseillaise on the Arc de Triomphe (which for many of Rude’s contemporaries far overstepped the bounds of the permissible in emotional expression), emphasizes the importance of measurement and exact proportion. The academic establishment teaches the young artist to work subjectively, “by eye,” while the revolutionary romantic artist teaches by means of objective mathematical tools. For the academy, the sculptor was an artist; for Rude, the contemporary of Géricault and Delacroix, the sculptor is first a skilled workman, a “practitioner.”

The oddity is not a matter of overcompensation on both sides, but of date. By 1850 many of the romantic attitudes, vitiated and trivialized, had become respectable: the superiority of inspiration, emotion, and subjective judgment over tradition, rules, and skill was now official. The primacy of emotion was, of course, passionately believed in by Rude, but he took it more seriously than the Ecole des Beaux-Arts, and he did not think it could be taught. Rude’s insistence on objective classical training (never abandoned by his generation) became a new form of romantic rebellion.

The history of romanticism is—to a far greater extent than the history of any other artistic or philosophical movement—a history of redefinitions. Hugh Honour devotes his entire first chapter to the problem of defining what was meant by the word. He rapidly sketches the innumerable meanings it has taken on since Friedrich Schlegel first tried to define it in 1798. He remarks on the bewildering variety of original contemporary views of romanticism (of Schlegel, Stendhal, and others), many of them contradicting the other. What Honour fails to remark is that these definitions are intentionally self-contradictory as well, deliberately inconsistent and unstable, fluid and expansive. The rest of the chapter, which attempts to describe romanticism by its contradictions, would have gained in cogency if he emphasized that the oppositions among romantic artists are internal and methodical.

Like every other movement, romanticism started as an opposition to what had existed just before it. In most of its original manifestations around 1800—the Jena circle of the brothers Schlegel, Novalis, Schleiermacher, and others; the “Lake poets” in England; the young French painters who turned away from the neoclassical tradition of David—romanticism presented itself as a break with a previous “classical” style, however variously conceived (either Greco-Roman or French)—a classical style that was understood as normative. This, too, is not without precedent; there have been other anticlassical styles before—early Christian art and rococo decoration are two examples. What set off many of the early romantics from all previous innovators is that they proposed not a new set of norms, but an abolition of norms, an artistic freedom inspired by the political ideals of the recent French revolution.

The true originality of romanticism, however, lies in a still greater ambition: a claim not only to destroy the classical tradition and replace it with something better, but eventually—in the near, or far, or infinitely distant future—to arrive at a higher form of classicism. Previous artistic revolutions had naturally implied that bad art was now being replaced by good, or even (as the Renaissance architects felt) that a barbarous, false, or non-art would at last make way for a revival of true art.


No revolution before romanticism had promised a universal synthesis—even a critical synthesis—that would incorporate all the achievements of the past. The final romantic pretension was the appropriation not only of all art in general, but of everything else: the poeticizing, or “romanticizing,” of life itself. The tearing down of barriers between art and life began systematically with early romantic theory of the late 1790s, although there had been sporadic hints of it before, above all in Diderot.

“Romantic poetry is a progressive universal poetry,” wrote Friedrich Schlegel in 1798 (and by poetry, he meant art in general). “Universal” because in the end it includes every other form (“In a certain sense,” Schlegel added, “all poetry is, or ought to be romantic”). “Progressive,” not only because it is not yet ended, but because it can never be finished (“It is its essence to be eternally becoming, and never to be achieved”). The main forms of romantic poetry for the Jena circle, the first and most influential of the avant-garde romantic groups, were the fragment and the novel. Both forms attacked the integrity of the classical work and the classical genres, the fragment because it escaped any ordering in a hierarchy of genres, and the novel because it blended all the genres together—the early German romantic novel combined autobiography, lyric poetry, drama, history, and fairy tale. (The best example is Franz Sternbald’s Travels by Ludwig Tieck, the story of a young pupil of Dürer.)

The revolutionary character of early romantic art was generally a matter of defiant pride. The young Astolphe de Custine, that shadowy figure on the margins of romanticism in France who was later capable of a penetrating book on Russia, wrote in 1814 in a letter from Vienna:2

These denominations of romantic and classic that the Germans created some years ago were used to designate two parties who will soon divide the human race like the Guelphs and the Ghibellines in the past. The romantic spirits understand each other at the first word: their opinions in literature, in politics, and I would almost say in religion, in spite of the different sects, are the same…. They form the new German school. The classics are equally consistent, but they do not understand the others, while the others understand them quite well…. This is what prepares us for a great revolution in literature, politics and religion…. I, for example, am essentially modern, and therefore romantic.

Sixteen years later, moved partly by personal dislike of Victor Hugo, Custine was to write that he hated the distinction of romantic and classic:

There is classic in Shakespeare and romantic in Racine…. I love romantic poetry, but I detest, for the present at least, the romantics. Their intrigues, their coterie spirit, their ill-founded pretensions to genius, their innovations which are nothing but impudent imitations, all this disgusts me with their circle, which does not prevent me at the same time from being bored with the classical party.

This increasing ambivalence—having a distaste for romanticism while holding tight to much of the early romantic heritage and doctrine, the new broader tolerance even for the once-hated French classical style and the way it is annexed to romanticism—is paralleled in many other examples. Delacroix, too, partly turned away from his earlier work and set out deliberately to reappropriate the grand tradition of mural painting: the label “romantic” made him nervous in later life. Wordsworth’s growing conservatism, Keats’s attempt at the end of his short life to recapture the high Miltonic style, Schumann’s abandonment of his early forms after the age of thirty and his concentration on the classical genres of symphony and quartet—all these are analogous developments.

They have a double aspect: they are partly betrayals of the early romantic faith and partly attempts to carry out its explicit program, present from the beginning, for a move to a new classicism, or, more precisely, to a synthesis of romantic and classical ideals. The early romantic style was hard to sustain: it needed either young nerves or a deliberate assumption of madness (that was Blake’s strategy). Many artists and writers of the first romantic decades either died young or wrote themselves out at an early age. The synthesis of romantic and classic, however, presented even greater problems: few examples of it escape a sense of inflated and unmotivated grandeur.

An account of romanticism always demands a sense of this irregular and constantly sustained movement, which can be traced both in the lives of the individual artists and in the general artistic and social development of Europe in the nineteenth century: it cannot be the history of the success or failure of a program, but of a series of programs and ideals continuously redefined. The consistency and integrity of romanticism lie not in the resemblance of these programs to each other but in the specific ways they were redefined and consistently expanded. In a brilliant epilogue to his book, Hugh Honour has a beautiful sentence that sums up both the constant change and the peculiar integrity of the romantic movement, and the way it extends into our own time:


The Romantic revolution which began in the 1790s was like the battle which “men fight and lose” in William Morris’s A Dream of John Ball; “and the thing they fought for comes about in spite of their defeat, and when it comes turns out not to be what they meant, and other men have to fight for what they meant under another name.”

An understanding of the instability and variability of meaning—of words, styles, actions—is central to the romantic movement, which both exploited this instability and suffered from it.


Hugh Honour has written the best book yet available (and perhaps ever written) on romanticism in the visual arts. In reality, his work is somewhat more restrictive than its general title: more precisely it should be called “Themes in the visual arts in the first half of the nineteenth century.” These themes embrace almost the full range of preoccupations, of art and life, which give the romantic period its extreme complexity and dynamism; and whether he is dealing with landscape or “the cause of liberty,” Honour writes of them with informed feeling and precision. Only Leon Rosenthal, at the beginning of this century, had the knowledge and the powers of synthesis to give a penetrating view of this art comparable to Hugh Honour’s; and Rosenthal limited himself to France while Honour deals with the whole of Europe.

In the second chapter, on the transformation of landscape painting in the early nineteenth century, Honour describes sympathetically and with great brilliance the way Constable, Turner, Runge, Friedrich, and Théodore Rousseau made landscape the vehicle for the expression of the most profound sentiments and even of religious ideas. He gives the familiar comparison between Constable and Wordsworth a new force, observing of Constable’s “six foot canvases,” for example, that

Each of them includes some incident which seems to link it with a particular moment of vision, when Constable experienced the “sense of harmony.” Often it is something unexpected if not, at first sight, untoward—the haywain is not on dry land but splashing through water; the lumbering cart-horse prepares to leap over a barrier on the tow-path; the white horse is not drawing a barge but standing in one. It seems likely that these and many of Constable’s other paintings are pictorial equivalents to the “spots of time” of The Prelude—moments of childhood to which poet (and painter) returned with changing views in a world which was also changing.

“Frozen Music,” the following chapter, starts with the treatment of music in romantic painting, and the way music became the ideal art for the period, a model of sensuous immediacy which all the other arts tried to imitate in various ways: the chapter continues with remarks on the development of romantic sculpture and architecture, perhaps because they are the most obviously frozen of the arts. This leads Honour into a long consideration of the Gothic revival and the treatment of medieval themes in romantic art, which is called “The Last Enchantments of the Middle Age,” and which deals not only with Pugin and the Pre-Raphaelites but with such obscure painters as Victor Orsel.

Other historical themes (classical antiquity, the Renaissance, the French Revolution) are considered in “The Sense of the Past,” and the political aspects of the arts in “The Cause of Liberty,” where the relation of political and artistic liberty is discussed skeptically (Honour’s most cogent remarks here are hidden in his notes in the back of the book). The romantic image of the alienated artist, the self-portraits and the symbolic treatments—the artist as lonely seer, as Don Quixote, as madman—are treated in “Artist’s Life.” Chapter 8, “The Mysterious Way,” is a mosaic, oddly mixing various related themes, death, sex, childhood, dreams, and religion.

In his epilogue, Honour convincingly demonstrates the way the realistic reaction to romantic art around 1850 was itself explicitly derived from romantic ideas and may even be said to have carried out much of the original romantic programs. Honour comprehends that romanticism must be conceived as continuously changing and continuously varied, as a movement through time rather than a settled body of doctrine.

Unfortunately the way he presents the subject and the format he has chosen prevent him from conveying his understanding fully. He tries to define romanticism almost as if it were an object, however fuzzy, and then throws up his hands in despair when it refuses to sit still for its portrait. We can see an example of this failure when he juxtaposes details from Géricault’s Wounded Cuirassier of 1814 with Friedrich Overbeck’s portrait of Franz Pforr of 1810. He contrasts the “sweeping brush-strokes and splashes of pigment dashed on with apparently exuberant spontaneity” by Géricault with Overbeck’s forms, “precisely, coolly articulated with firm outlines”—the dynamic, open structure of the one with the other “static and claustrally closed” (although Overbeck’s picture seems much more open than Honour maintains). Honour goes on to claim that in view of such contrasts “there is no Romantic ‘style’ in the visual arts, if by that is meant a common language of visual forms and means of expression, comparable with the Baroque or Rococo,” and he adds that it “is therefore impossible to write about Romanticism as about Neo-classicism or any earlier international style.”

This distinction between romanticism and earlier “styles” is not tenable except with a certain amount of juggling and no little confusion. It only works by a narrow definition of, say, the baroque and a wide one of romanticism: it would be difficult to claim that a picture by Vermeer and one by Pietro da Cortona provide less of a contrast than the one Honour finds in the work of Géricault and Overbeck. A “common language” for baroque art could only be arrived at by resolutely excluding anything that does not fit, and casting it out as non-baroque. The problem here is a naïve methodology, which defines a style by listing the characteristics that a given number of roughly contemporary works have in common.

Underlying this method is often an even more naïve belief that works that belong to the same style ought to look alike. Sir Anthony Blunt, for example, defined baroque architecture by the Roman style of Bernini and Borromini; he admitted as baroque anything that resembled their works or was obviously derived from them, and disqualified everything else. This kind of definition, which Honour seems to view as an unattainable ideal for romantic style, is distinguished by its consistency, rigidity, and poverty. It does not stimulate understanding but leads to pigeon-holing, to a dead end of classification, of no use as a tool of analysis. Works are disposed of as baroque or non-baroque, romantic or non-romantic, as if these were categories that had some genuine historical reality. But terms like “baroque” and “romantic” designate not well-defined entities or even systems. They are primitive shorthand signs for long-range historical developments which one feels nevertheless to have a certain integrity. These terms proclaim a confidence that all the various phenomena of the period, including the most eccentric and exceptional, are interrelated and interdependent in ways that can be indicated and represented by the historian.

Honour deeply appreciates the eccentric, but has difficulty fitting it into his scheme. He takes refuge in the easy formula that what the romantic artists had in common was that each was different, but that is not much help in achieving a coherent picture. This may partly explain why, in spite of many references in Romanticism, Ingres is not given his proper place. Hugh Honour closed his Neo-classicism with Ingres’s Oedipus—we are told there that Ingres’s art is no longer neoclassical, in spite of superficial similarities, but already the art of the nineteenth century.

In his new volume, however, Honour cannot quite rid himself of the traditional view of the “classic” Ingres opposed to the “romantic” Delacroix, even if he is aware of the ineptness of this myth, invented and enforced by Monsieur Ingres himself, in his later years. Although Honour refers to Ingres’s critical disaster at the Salon of 1819, where he fared much worse than Géricault, and to the defiant pose of his early self-portrait, the reader would never guess that Ingres was in life almost the prototype of the romantic artist, consistently rejected and misunderstood until he was well into his forties. The self-pity that pervades his letters during his long years of isolation in Rome is almost boundless.

Honour has no trouble accepting the linear primitivism of the German painters called the “Nazarenes” as a strictly romantic alternative to a painterly style, but is not able to take the same step with Ingres, perhaps because Ingres’s art does not seem so obviously neo-Gothic to us. But his contemporaries thought that it was and the Gothic label stuck to the painter of Jupiter and Thetis until the 1820s, when Ingres reintroduced a greater degree of modeling and three-dimensional depth to his painting.

Hugh Honour sensitively observes that Ingres’s work, in its rendering of space and surface quality, is radically opposed to the neoclassical principles advanced by David. But he misses the chance of pointing out that Ingres’s enamel-like finish is another way of attracting attention to the surface of the picture, a striking alternative to the bold brushwork of Delacroix and other romantic “colorists.” Furthermore, Ingres’s anti-Davidian attitude also affected his choice of subject matter and his manipulation of genres. For many years, the artist survived by drawing the pencil portraits we all love but which he himself considered as demeaning work. To one who sought him out and inquired whether this was the home of the portraitist, he proudly answered: “He who lives here is a history painter.” In spite of his declaration of principle, Ingres’s art undermined neoclassical standards. He painted very few canvases with subjects that would have been judged entirely appropriate by followers of David. The subject of the vast Jupiter and Thetis, one of the most idiosyncratic works ever painted, was criticized by the Academy as too trivial for such a large composition; it lacked the kind of moral message that would have justified such an expanse of paint. The Dream of Ossian (which Honour curiously leaves out of his list of dream paintings) would probably have fared no better if submitted to Davidian criticism.

Most of the time Ingres painted what is usually known as “historical genre,” smaller paintings of historical episodes treated as genre scenes—historical reveries rather than historical examples. Such costume pieces, whose subject was often taken from literary works, were popular with collectors, and were despised as a bastard kind of painting by critics who held to the grander neo-classical standards of history painting. Ingres’s (and for that matter Delacroix’s) efforts to elevate such costume pieces by increasing their expressiveness and investing them with a projection of the artist’s own situation and feelings (Ingres’s Aretino in Tintoretto’s Studio; Delacroix’s Michelangelo in his Studio) are characteristic of the tendency to blur categories. Ingres, of course, did paint a few real historical paintings, although not always with equal success, but even in later life his fierce classical position is largely a mystification and a romantic paradox or irony. His final masterpiece, The Turkish Bath, is a self-indulgent extravagance.

Honour clearly understands the long-range development inherent in romanticism, but his mode of presentation is static. He often approaches the artists of the time simply by inquiring what their treatment of the great romantic themes had in common, or how they differed. From a certain explicit shyness about reading future developments into the period, he evades asking what all that artistic activity was leading to. The result is that the single, telling detail which would make the subject historically more intelligible is often withheld.

In a sympathetic and moving discussion of the work of Auguste Préault, the most radical of all the romantic sculptors, Honour concisely and with extraordinarily persuasive language conveys both the qualities of Préault’s sculptures and the shocked reception of them by his fellow artists: he makes us see why contemporary poets such as Charles Baudelaire and Théodore de Banville admired Préault. Yet the most recent dictionary of proper names in France (the Petit Robert II) in its few lines about Préault still tells us something about him that we cannot find in Honour and that illuminates romantic sculpture in general. Préault’s great ambition—no one ever let him carry it out—was reportedly to sculpt a mountain, to turn one of the peaks near Le Puy into a work of art. Honour remarks elsewhere on the attempt of romantic sculptors to integrate their work into a landscape, but he never tells us how far they were prepared to go. Was he afraid that the apparent prefiguring of some twentieth-century earth works would seem too lurid? In any case, a chance was missed to point out that romantic style is often best characterized by what the artists and poets hoped to achieve: an absurd magnitude of ambition is an integral part of the time.3

Only a sense of history as movement rather than as a series of solid states can restore significance to the words and images of the early nineteenth century. Honour quotes E.T.A. Hoffmann on Beethoven:

For E.T.A. Hoffmann, writing in 1810, Beethoven was “ein rein romantischer…Komponist,” because his music “sets in motion the lever of fear, of awe, of horror, of suffering, and awakens that infinite longing which is the essence of Romanticism.”

The date alone will not place this: one must also know that for E.T.A. Hoffmann the first “romantic” composers were Haydn and Mozart. Only by this can one see that his characterization of Beethoven is part of a double program that Hoffmann holds with many of his contemporaries: first, to appropriate an already acknowledged classicism and assimilate it into romantic art, and, second, to take pure, instrumental music, the most abstract of the arts, as a model of romantic poetry, and to claim for poetry the same ability as music to create an independent world of its own—a visionary world, whose relation to the real world is always ironic as the absolute purity of instrumental music is finally unattainable for the poet. It is unattainable for the painter as well, unless he is willing to move toward non-representational art: there were proposals to do so in the late eighteenth and early nineteenth centuries, but Honour does not tell us about them, and oddly discusses music as a model for the other arts only in its relation to sculpture and architecture.


Honour fails, except intermittently, to convey a feeling of the historical development of romanticism in the visual arts because his decision to attack the subject exclusively by themes leaves him no place for the two aspects of the arts of the time which would allow us to understand the nature and direction of most of the changes: the division of genres in the romantic period, and the marginal arts—book illustration, caricature, decoration, prints, works by amateurs, and so forth. The romantic movement might with some justice be defined as a progressive abolition of the hierarchy of genres and of the distinction between central and marginal forms.

Honour comes close to discussing the question of genre when he treats landscape, but he somehow misses the opportunity. In the end, landscape for him is a matter of subject matter, rather than of genre. Yet there is no other way except through genre to understand the radical transformation of artistic expression during the romantic era. The division of genres, which is fundamental to Western art since the Renaissance, is not simply a matter of specialization or division of labor, but implies that different subjects are to be treated in a particular way. It is a specific application of the concept of decorum, the idea of appropriateness. (Perhaps an even better definition of romanticism in the arts would be a progressive destruction of decorum.) Thus a very large format and a great elevation of style were considered appropriate only to what was called history painting, which meant compositions with several human figures dealing with a serious subject, such as allegorical, religious, or historical topics of importance, conveying a great lesson. Subjects of daily life, known as “genre painting,” implied a smaller format but also a less severe handling and less tightly constructed composition.

This basic and mostly tacit convention was flexible and underwent considerable adaptations at different times and different places, but it became most explicit and rigid with the neoclassical painting that just preceded romanticism. In their assault on this tradition romantic artists followed two principal courses: on the one hand they attempted to blur the distinctions, although capitalizing on the potentials of the tradition; on the other, they developed the supposedly minor or “inferior” kinds of art such as landscape or, even more drastically, book illustration, and gave new significance to “sketches” and “studies” which began to compete with finished official works.

This attack on the distinction and hierarchy of genres was particularly clear in France where its authority was most strongly established. In Germany and in England landscape developed with less constraint, and also very much earlier, even if Thomas Lawrence still told Constable, when he was finally elected at the Royal Academy in 1829, that he was lucky to get in when there were several history painters among the candidates.

The attack on the genres by the romantic artists betrays one of their deepest ambitions: the achievement of “immediacy,” of forms of expression directly understandable without convention and without previous knowledge of a tradition. It was no doubt an unrealizable ambition, but it goes back to Rousseau, to his deep distrust of language, and the way it betrayed and deformed one’s inmost thoughts and feelings. The romantics wanted an art that would speak at once and to all. The attack on the system of genres is an attack on a tradition that made intelligibility dependent on connoisseurship. Above all, the immediacy of statement was best achieved, not in history painting, but in the inferior genres of landscape and portrait. For this reason, too, the immediacy of the sketch and the study took on an importance it had never yet had in the history of art, as much as they may have been appreciated before.

With few exceptions Honour mostly discusses and reproduces large finished works, “important” statements. For most periods, this would be the preferable course, the significance of art being defined by the major projects and in the major cities. The art of the seventeenth century, for example, is shaped in Rome and at Versailles. Wonderful as they are, the etchings of Benedetto Castiglione, a charming artist from Genoa, could safely be ignored in a study of seventeenth-century style. The caricature drawings of Bernini, too, are fascinating, but it is St. Peter’s Square and the great sculptures that matter. But with the early nineteenth century minor or marginal phenomena become essential to an understanding of the period, because an inescapable aspect of the art of that time is the undermining of the main tradition and of the center of powers. One might say of romanticism that the destruction of centrality is at its center: “Any line is an axis of the world,” as Novalis put it.

The most extreme statements, where romantic doctrine finds its purest expression in the visual arts, were often made by amateurs, craftsmen, eccentrics outside the accepted centers of professional activity. It took a dilettante like Victor Hugo to produce large totally abstract drawings and to give material expression to the theory of abstraction more than a half-century after Alexander Cozens’s attempts to do so. Cozens’s “blots” or semi-abstract land-scape drawings were largely disregarded; in fact his advocacy of abstraction in his extraordinary “New Method” of 1785 was practically suppressed and disappeared altogether in the nineteenth century, but it had a significant impact on Constable.

In prints and book illustrations, romantic artists were freer and in fact more consistently successful than in painting. Not only was printmaking considered a “minor art” and consequently less governed by conventions, but the new media of lithography and wood engraving encouraged innovation. Thomas Bewick, who perfected the new technique of wood engraving and effected a revolution in book illustration, was a thoroughly peripheral figure; a craftsman and self-taught popular zoologist, he spent his entire life in Newcastle. Yet he became a sort of romantic hero. When Wordsworth in one of his poems sets him above Sir Joshua Reynolds, what is most extraordinary is not so much the outrageous judgment as the very fact of comparing the incomparable. Bewick ought to be mentioned in any account of romantic art because he not only made technical discoveries but also invented the romantic vignette, a completely new type of illustration which swept Europe. The vignette is the most characteristic form of romantic art, a sort of visual metaphor for the romantic fragment.

Lithography was no less important, if only for the extraordinary flowering around 1830 of French caricature (left unmentioned by Hugh Honour, like book illustration). It was on the basis of his prints that Daumier was considered one of the greatest artists of his time by Baudelaire. Daumier’s paintings are great indeed, but there is an unfortunate tendency today to neglect his lithographs (not one is reproduced in Romanticism).

In fact, even before Daumier began, the great romantic school of comic art was already established. Grandville, discussed by Honour in his introduction as anticipating some twentieth-century trends, is a major figure who confined himself to caricature and book illustration. He and Henri Monnier, draughtsman, actor, writer, and altogether a highly picturesque figure, perfected a new style of comic lithography in the late 1820s. But it was Philipon, less an artist than an organizer, who fully realized the potential of this kind of art for political opposition. He made images the major attraction of his short-lived periodical La Caricature (1831-1835), and then of the long-lasting daily Le Charivari. It is within the pages of these popular publications, far away from the official institutions of art, that the genius of Daumier matured.

Caricature was essential to romantic art. The romantic ideal of immediate expression pushed in two directions. If expression was not conventional, it had to be natural, but it was not exactly clear what that implied. One possibility was to follow the model of music, which seemed to affect people so directly, in order to make art as independent and self-sufficient as possible. This turned art toward abstraction, and explains how the theory of art for art’s sake developed within romanticism. (Honour briefly but cogently indicates Ingres’s importance for this trend.) But another tendency led to realism. Natural forms taken by themselves were believed to be inherently expressive, and taken directly into art they would convey their meanings without any appeal to convention: this was felt to be particularly true of physiognomy, which provided the basis for the new type of caricature, Grandville’s as well as Daumier’s. Caricature had the further advantage of being a genuinely popular medium which gave some substance to the left-wing libertarian aspirations of the “romantic revolution.”

The difficulties that the neglect of the genres and the marginal makes for Honour can be seen even in a passage as fine as the following one on Géricault’s Raft of the ‘Medusa’ and its relation to the work of Baron Gros:

By stressing the realistic, at the expense of the idealistic or classic, elements in David’s work, Gros evolved a new style eminently suitable for pictures of modern subjects. And although Géricault made a more decisive break with tradition in his choice of themes, he was in other respects more faithful than Gros to Neo-classical principles. Working in the approved academic manner, he proceeded systematically from numerous swiftly drawn sketches, in which he gradually determined the general composition, to wonderful nude studies of individual figures drawn with a precision of outline and volumetric clarity equalled by few Neo-classical draughtsmen. In the pursuit of truth he frequented dissecting rooms and made the extraordinary paintings of heads and limbs of corpses which Delacroix was to call “the best argument for beauty as it ought to be understood.” Yet, in the final work, he followed the conventions of the grand style by endowing the survivors on the raft with the healthy physique of Greek athletes, rather than depicting them as they appeared when rescued—bearded, emaciated, covered with sores and wounds.

If Honour appears here to minimize the revolutionary break with the classical tradition in favor of the more conventional aspects of the finished work, it is partly because of what he leaves out. He glosses over the relation between the studies and final work and his account leaves unexplained Delacroix’s enigmatic characterization of the wonderful oil studies of severed limbs: “the best argument for beauty as it ought to be understood.” Why so describe these magnificent but repellent works? We must turn to Delacroix himself for an explanation. In 1857 he made a note for an article on “Subject” intended for the Dictionary of Fine Arts: “Painting does not always need a subject. The painting of arms and legs by Géricault.” It is clear that Delacroix understood how radical was Géricault’s accomplishment: the abolition of the traditional subject in favor of a new symbolism, immediately comprehensible and independent of culture and convention. Perhaps more than any other artist Géricault realized this romantic ambition, and with the abolition of subject painting took a first step toward abstraction.

“Paintings without subjects”—“like music”—these were the reproaches thrown at Constable by the conservative French critics alarmed by the sensation his work created among the young artists in Paris—and Constable welcomed these reproaches as if they were the highest compliment. His landscapes, indeed, have no “subject” in the early nineteenth-century sense: all narrative content has been removed and most of the sites chosen have no intrinsic interest. It is a pity that Honour never discusses music as a model for the visual arts specifically in relation to landscape, because it was in this connection that it generally came up at the time, along with the abolition of subject and the hesitant proposals for abstraction.4 It was in landscape that the two extreme tendencies of romantic style—realism and art for art’s sake—could be reconciled. This synthesis gave landscape its supremacy among the genres of the time, and eventually allowed it to push history painting out of the center of art.

Géricault’s severed limbs do not release their full significance considered only as a study for The Raft of the ‘Medusa.’ Although they may have been painted in connection with that work—it is not certain—they are not preparations for particular motifs in the painting but a study in mood, an exercise in horror with no anecdotal justification. Thousands of studies of arms, hands, and legs made as preparation for larger works or simply as exercises have come down to us over the ages, some of these elaborately worked out in oils. But the Severed Legs and an Arm of the Musée Fabre in Montpellier is not simply a study of limbs; we see hacked-off limbs, composed and balanced into a full picture, a still life, a gruesome pun on the idea of a nature morte.5 It is the “best argument for beauty” because of the abolition of subject: what is beautiful comes entirely from the paint, the composition, the light, the handling. The objects portrayed repel, and in so doing demonstrate the supreme power of the act of portraying.

A marginal work, the study, here attained a form of the sublime to which the central finished work could not pretend. The process by which the “finished” work was finally to be pushed aside until, with the impressionists, only the “study” remained began with the romantics, and it was clear from the outraged protests of critics that the direction of things was already understood at the time.

It is interesting that so much of the significance of romantic art for the artists themselves can be discovered by looking ahead at what was to come out of what they had only started. No doubt the artists would have been, at the very least, disconcerted at the future of their innovations, but they were the first to comprehend that artistic work partially escapes the intention, control, and understanding of the artist. The greatest English architect of the early nineteenth century, Sir John Soane, when he designed the Bank of England, sent three oil sketches to the governors of the Bank. The first showed the Bank brand new, shiny and gleaming; the second portrayed it when it had weathered a bit, acquired some ivy and a mellow patina; the third imagined the Bank in a thousand years, as a noble ruin.

This Issue

November 22, 1979