Art; drawing by David Levine

A good race horse, one supposes, is one that wins races, a good chess player one who can beat his opponents; we can tell a good watchmaker by his skill in making or repairing watches which accurately show the time, and a good linguist by his testable mastery of foreign languages. But how can we tell what is a good work of art or who is a great artist? The question does not only arise in the auction room where genuine or sometimes spurious masterpieces change hands at astronomical prices. Every year countless art students in all civilized countries are admitted or rejected by art schools and colleges, receive degrees of varying grades or prizes for the best work; later they submit their productions to owners of galleries, to critics, and to the public on whose evaluation of their efforts they will depend for their livelihood. By what criteria are they judged? Is it all a matter of fashion and pretense? If it is not, why is there so much disagreement among critics, and, if it is, how can we explain the fact that hard-headed businessmen “invest” in works of art?

Many specialists have preferred to withdraw behind a smokescreen of esoteric jargon at the approach of such philistine questions, but in the book under review Professor Jakob Rosenberg has had the courage to come out into the open and to attempt an answer in plain language. He has devoted his Mellon Lectures in the Fine Arts of 1964 to what he calls “Quality in Art.” The title, perhaps, bears traces of the author’s background and professional origins. In German we speak of Qualitätsgefühl (a feeling for quality) as a first prerequisite of the Museum man and the connoisseur. Professor Rosenberg has long acquired a reputation in both these fields. He belonged to the great team of specialists who looked after the famous print room of the Berlin Museum, and after emigrating to the United States he continued his distinguished career at the Fogg Museum of Harvard University both as curator of prints and as a teacher. He has never been content merely to practice the intuitive skills of the discriminating collector but has regarded it as his task to pass on his insights to the next generation, and not to specialists only. For many years he gave courses at Harvard in which he tried to open the eyes of his students to the subtle beauties of prints and drawings, and he succeeded, by all accounts, in communicating his love and appreciation to many young people whose lives he has thus permanently enriched.

THE METHOD employed in this book goes far to explain his success. For the main body consists of thirty-six pairs of well-chosen comparisons between individual works of art, mostly drawings, which differ appreciably in artistic quality. Comparisons are routine in courses in art appreciation, but what distinguishes Professor Rosenberg’s juxtapositions from similar ones is the skill of selection based on an unrivaled knowledge of old master drawings. He has succeeded in finding a large number of works genuinely comparable in style, subject, and technique, works, that is, of the same kind whose relative value needs to be judged.

It was an excellent idea to select the material from the workshop of the connoisseur. The type of problem exemplified in this book: whether a given drawing is by the hand of van Gogh or by that of a forger, or which of two very similar drawings in Rembrandt’s style is the copy and which is the original, raises the question of “quality” in a severely practical way. It will be assumed, in other words, that the forger or copyist will be less good an artist than the master he imitates, and that this artistic weakness will give him away. The same is also true, up to a point, of that other question which traditionally concerns the connoisseur, the question whether a given work is by the master himself or merely by his “school.” Here, too, it is generally believed that a drawing in the manner of a certain master which is not quite up to his artistic standards must be attributed to one of his apprentices or imitators.

Of course, where other criteria are lacking there will always be an element of circularity in the procedure. To take the first comparison in the book: two drawings of heads of orientals of which one is attributed to Schongauer, the other to an imitator. It was Professor Rosenberg himself who made this attribution as long ago as 1923 in his pioneering monograph on the artist. He made it on the grounds of quality and he was surely right. But is he equally right now to demonstrate the difference between a master and an imitator by means of the same example? It is a pity, at any rate, that the book lacks any bibliographical references which would enable the reader to check the source of the attributions and the consensus achieved. Among connoisseurs, there are expansionists and restrictionists. The first are popular with collectors who like to see the name of famous masters on the mounts of drawings in their possession, the second are preferred by art historians who want to keep the image of the masters pure and unadulterated. Professor Rosenberg belongs to the latter group. He would no doubt endorse the remark attributed to his compatriot Max Liebermann that it will be the job of future art historians to deny that he ever painted his bad pictures. Just as Homer sometimes nods, so even great artists sometimes may have muffed a drawing, but this caution about the safety of attributions does not invalidate Professor Rosenberg’s method of sharpening our eyes to quality in performance. In reading him we learn how a sensitive and experienced connoisseur argues cases of this kind; no teacher of art history will neglect this lucid demonstration.


There is one limitation to this method to which the author himself has frankly and frequently drawn attention. Rosenberg’s criteria derived from connoisseurship apply ideally only to the rating of works very close to each other whose attribution can conceivably be in doubt. They permit him to show the superiority of a study of drapery attributed to Leonardo over one he assigns to Lorenzo di Credi, or of a Biblical illustration he gives to Rembrandt over one he considers to be by his pupil Eeckhout. But they leave still unanswered the larger question whether Leonardo or Rembrandt was the greater artist or, for that matter, what may have been the relative merits of Lorenzo di Credi and Eeckhout. Yet Professor Rosenberg is confident that in his attempts to show up the difference between the excellent and the mediocre he will gradually arrive at a set of criteria of more general validity. Modestly not staking his claims too high, he has phrased his descriptions in an impressionistic language from which he yet hopes to extract a series of uniform concepts of excellence. Here the result is more vulnerable.

NOT SURPRISINGLY, perhaps, he is most convincing where he is least theoretical. Going through his descriptions one notices how often he draws attention to success or failure in the accurate and persuasive representation of reality. “There is in the Leonardo a much clearer expression of the body’s form and movement under the drapery” (p. 136). Dürer “clarifies the complicated opening of the fur coat and the shirt” (p. 162). In a drawing of Rembrandt’s school “the structure of the bridge leading through the middle distance on the left is hard to understand…” (p. 185). “Compared with Watteau’s figure, Lancret’s appears stiff, in spite of its movement. The light does not model the form with the same intensity, flexibility and ease” (p. 190). This is the language of the good old-fashioned drawing master who urges the pupil to look carefully both at the model and at his own drawing. It is well to be reminded that the masters with whom this book is concerned generally passed this test much more easily than did the tyros, that Leonardo or Rubens knew how the head sits on the shoulders while lesser artists were less sure.

Professor Rosenberg knows as well as anyone that this mastery alone does not suffice as a criterion of excellence. Moreover, representational skill cannot always be tested with the same degree of objectivity. Our judgment becomes increasingly subjective the further we move from representation to suggestion. All representational drawings contain such an element of suggestion. The marks on the paper can be read as indications of solid shapes and of space, but they need not be so read. The less redundancy there is in the drawing (for instance through shading), the more ambiguous will the artist’s message become. Thus a perverse or stubborn reader may decline to follow the author when he is asked to “admire Rembrandt’s mastery in creating a highly expressive scene with a minimum of articulate strokes that build up…the structure of this sandy, rugged landscape…and a sundrenched atmosphere” (p. 184). The same non-cooperative reader might wonder whether the charge leveled against a drawing by van Borssom is not inherent in the medium itself, i.e., that “his spatial image appears incomplete and vague in the distances” (p. 190). In sober fact not even a Rembrandt can ever give us enough information in a landscape drawing to establish accurately the size and distance of the objects represented, nor, for that matter, can a photograph. Not that such criticism necessarily invalidates Professor Rosenberg’s observations: they remain illuminating when applied to particular drawings, but they lose some of their force when they are made to serve for the formulation of general principles. The general qualities Professor Rosenberg demands for “Form” and “Space” in works of art of proven excellence (p. 204) are therefore both too elusive and too subjective. There must be many run-of-the-mill drawings where the forms can be described as showing “solidity, organic character, coherence [and] clear distinction of planes,” and where the space can be read as “articulate, continuous and comprehensive.” There must also be masterpieces which could not be so described.


In the author’s final list of criteria, however, these representational qualities do not come first. The list begins with “qualities of line and tone” which are specified as “sensitivity, articulateness, flexibility, ease, surety, spontaneity, and suggestiveness.” Leaving aside the last (which really belongs to form and space) we have a number of laudatory terms none of which would be easy to explain with precision. Not that they are meaningless, any more than is the word “quality” they are meant to characterize, but one could think of countless instances where their exact application would be in doubt. Perhaps that criterion of excellence they are meant to convey could best be subsumed under the heading “control.” The great artist, as Professor Rosenberg shows in so many instances, is in superb control of his medium. He can do what he wants, largely because he only wants what he can do. He is intimately conversant with the potentialities of his means and responsive to the finest shades. Though he may not always be able to predict the exact shape or strength of the mark his pen or brush will leave on the paper, he will still remain in control by adjusting the next to the observed effect.

To talk in modern jargon, there is a constant “feedback” between the great artist and his medium when he is on top of his form. Professor Rosenberg often brings out this quality in his best descriptions, particularly in those of virtuosos like Rubens or Watteau. But clearly there also are artists who like to wrestle with their medium and who would avoid the impression of “ease”—Cézanne being the most exalted example. It looks once more as if the criteria of excellence here were also more elusive than the author allows. Moreover, how are we to recognize them? How are we to convince the unwilling skeptic that in the works which displease the author certain virtuoso effects are “overdone” (pp. 183-189)? And what if our doubting Thomas counters in another instance that he has no objection to “meticulous execution,…mechanical shading [and] hardness of contours” (p. 146)? The truth is surely that every critic is compelled to use such loaded terms in conveying his impression of artistic quality, but that he deceives himself if he thinks that the impression is derived from these observations. The description is often little more than a rationalization of the response.

MANY of Professor Rosenberg’s most persuasive characterizations, in other words, might be translated into exclamations of delight or of disapproval. They are gestures or pointers rather than the “concepts” he is after. This applies in particular to the last separate criterion he wishes to establish, that of “formal organization” which includes “integration,…sense of balance, and richness of relationships.” He is anxious to show that here he has a tool which will also allow him to assess works of non-figurative art, but the author’s approach here is particularly cautious. There is a comparison between a cubist Picasso and a painting by Albert Gleizes in which he claims for the former that “there is hardly a detail that could be changed without disturbing the balanced order of the whole” (p. 225)—a criterion taken from Aristotle’s Poetics which is both somewhat obvious and hard to verify. He then presents a Kandinsky which equally contains fragments of the visible world in a jumble that is hardly convincingly described as a “coherent whole.”

Not that anyone need doubt that we can and do sometimes experience an arrangement of shapes or marks as balanced or coherent. Contemporary painters, of course, frequently prefer the disturbing to the restful and the explosive to the organic. In some respects more “abstract” art would have provided an interesting test case for the author. For whatever one may think of such experiments it is clear that some of these works exhibit more “quality” than do others. Perhaps, also, it would be easier to show in such instances that the words we use in describing our reactions can never be more than loose metaphors, no more and no less so than are the descriptions of flavors in wine-seller’s catalogues. Such descriptions are certainly not meaningless either. They convey something about the global character of the experience. But their sense easily evaporates when we are tempted to locate them more precisely. Professor Rosenberg is not the first author, nor is he likely to be the last, who has fallen into this trap when trying to give reasons for his reactions to individual works of art. Thus he justly exclaims in front of an enchanting drawing: “How subtly and expressively Degas contrasts shapes and accents,” but tries to account for this intuitive conviction of quality by asking us to “notice the rectangular corner of the chair with the oval of the head, or the central vertical on the skirt with the diagonals formed in the center. There are variety and integration” (p. 198). This is the language of the art-historical seminar which derives its only justification from the need to keep a lantern slide a little longer on the screen.

True, this type of description is also sometimes claimed to teach beginners to “see,” but is there no danger that it deteriorates into a mild form of brain-washing? For those who know the vocabulary it is fatally easy to apply that technique also to a worthless work. We here read of a fine drawing by Matisse of a nude in an armchair in which “there is a balance of attraction between the head, framed by the arms, and the legs so interestingly set one above the other” (p. 209). There must be plenty of pinups which could be described in the same terms without exhibiting “quality.” More dangerous even is the possibility of this rhetoric being used to run down a great work. The cover of Professor Rosenberg’s book is graced by a ravishing Picasso drawing of a nude in a landscape. Would it be hard to spoil someone’s pleasure by describing the outline as “wiry,” the anatomy as “superficial,” the head as “merely pretty,” and the landscape as “vague”? Was it not in some such way that great artists (including Picasso) were in fact condemned out of hand by critics of the past? Was not Cézanne found clumsy and Rembrandt’s chiaroscuro “overdone”?

IT IS TO BE FEARED that Professor Rosenberg overrates the immunity of his method from such failures. He has preceded the main body of his book with five introductory chapters on great critics of the past, each representing a century since the Renaissance: Vasari, de Piles, Reynolds, Thoré, and Roger Fry. These richly illustrated chapters, which are mainly confined to quotations interspersed with a few words of commendation or mild criticism, are frankly disappointing. No effort is made to place these great figures into the context of their time, no hint is given of the relation between art criticism and the powerful theories of rhetoric from which the Academic theory of art derived so many of its concepts. But perhaps the basic flaw in Professor Rosenberg’s account is his apparent conviction that this Academic theory was little more than an obstacle on the road to the discovery of artistic quality. On the whole the critics of the past get good marks whenever they appear to transcend the alleged limits of the academic dogma. But much of this rests on a misunderstanding of the issues involved. True, academic doctrine took it for granted that an artist had to learn and master the skills of representation and dramatic evocation, and that he should not offend against the rules of decorum. But ever since Vasari it was part of the tradition that perfection demanded not only a knowledge of the rules but “a certain license within the rules” and a “grace that exceeds measurement.” This doctrine of grace is the important correlate of the doctrine of genius and of ideal beauty. Human effort alone can never create a work of art. It also needs grace, a term that hovers between our idea of “gracefulness” and that favor from heaven which is the free gift of God.

In the last analysis this is the very doctrine Professor Rosenberg applies for the discernment of excellence, though what was known as grace he describes in words such as “vitality, originality or inventiveness,” characteristics which are equally beyond the reach of conscious effort. In trying to show the presence of this divine spark in the work of the masters he has earned our gratitude for inculcating a fresh respect for their towering achievement. But it would be a pity if this respect were gained at the expense of journeymen who used such talents as they had to the best advantage of their art. For where the emphasis on vitality and originality becomes too exclusive it may lead precisely to those attitudes which, in Professor Rosenberg’s view, have undermined “quality” in contemporary art. In the long run it may turn out to be more profitable to probe the traditions of excellence in craftsmanship that produced a Lorenzo di Credi or a Cornelis de Vos than to look for that mysterious quality which may enable a rare genius to absorb, transfigure, and transcend these skills.

Yet it would be wrong to conclude that Professor Rosenberg’s pilgrimage in search of the sources of grace was in vain because he has not quite succeeded in plotting the life-giving spring on the map of rational knowledge. Would he really want us to follow its waters to the source only to bottle and label it? If we could specify in precise words what characteristics combine in that supreme excellence he has taught us to sense in the works he loves most, we could program a computer to match them ad libitum. He is the last man to want us to make such an attempt, but even if we wanted to do so, we would surely fail. Our language which was developed for quite different needs is notoriously unable to specify sensory qualities, let alone that resultant “quality” of which we need not doubt the existence because we cannot pin it down in words.

The story goes that a manufacturer of a mechanical piano approached Artur Schnabel with a request to play into his machine. “Ours,” he assured the artist, “is not one of those crude contraptions you may well dislike. In fact it has ten gradations of loudness.” “What a pity,” replied the pianist, “for unfortunately I have eleven.”

This Issue

February 1, 1968