Why Are Our Pictures Puzzles?

On the Modern Origins of Pictorial Complexity

James Elkins

• Publisher: Routledge
• Year: 2004
• ISBN:
  9781135963569
• Edition: 1
• Topic: Art
• Subtopic: History of Art

With bracing clarity, James Elkins explores why images are taken to be more intricate and hard to describe in the twentieth century than they had been in any previous century. Why Are Our Pictures Puzzles? uses three models to understand the kinds of complex meaning that pictures are thought to possess: the affinity between the meanings of paintings and jigsaw-puzzles; the contemporary interest in ambiguity and 'levels of meaning'; and the penchant many have to interpret pictures by finding images hidden within them. Elkins explores a wide variety of examples, from the figures hidden in Renaissance paintings to Salvador Dali's paranoiac meditations on Millet's Angelus, from Persian miniature paintings to jigsaw-puzzles. He also examines some of the most vexed works in history, including Watteau's "meaningless" paintings, Michelangelo's Sistine Ceiling, and Leonardo's Last Supper.

part 1
CONSIDERING THINGS

1
THE EVIDENCE OF EXCESS

IT MAY BE UNFAIR TO COMPARE VASARI’S ACCOUNT WITH THESE MODERN INTERPRETATIONS…. BUT WHY DOES HE SAY SO MUCH LESS ABOUT PIERO THAN PRESENT-DAY COMMENTATORS?

—DAVID CARRIER¹

It is best to begin by setting out some pertinent facts about the growth of the literature, simply to show that contemporary writing is historically anomalous in a potentially significant fashion. That condition, I think, may be especially transparent to working art historians. Writing art history, and reading it from the age, say, of fifteen, it may appear that what gets said about pictures is about right, and that the detail of art historical writing is more or less commensurate with the actual detail of the artworks, or perhaps even insufficient. Certainly, working on a specialized subject—for example on a dissertation—fosters the conviction that the existing literature is inadequate, and that a great deal remains to be said: but it does not usually lead to the conclusion that the amount or the kind of writing is fundamentally strange.

To see the writing itself as a problem, it is helpful to take a step back from the concerns of any one subject or specialty, and consider the shape of the discipline in a more abstract sense. In this chapter I offer some stories, facts, and figures about the growth of art historical writing, and especially about the length of our texts. At first I want to do this without drawing any conclusions, taking the numbers as what scientists call raw data. Afterward I will consider several of the explanations that might be, or have been, proffered to explain why contemporary historians write so much more than past generations. I am not entirely satisfied with any of the initial explanations, and this opening chapter is principally intended to unsettle the more common convictions about what is happening (or not happening) to the discipline.

THE SHEER VOLUME OF WRITING

According to George Steiner, there are on the order of 30,000 dissertations written each year on modern literature alone.² His estimate may be right, because American art history departments account for almost 400 dissertations each year, so it would not be unreasonable to guess that there are nearly 1,000 dissertations worldwide.³ The two principal indices of artwriting, the Art Index and the Bibliography of the History of Art (BHA) each record a total of around 30,000 entries each year, written by an average of 22,500 different art historians and critics.⁴ Certainly the discipline has grown precipitously. The earliest index of art historical literature is the Répertoire de l’art et d’archéologie; its first volume was published in 1910 and has a combined subject and author index which runs to about 4,000 entries. A combined subject and author index for the 1995 Art Index would have on the order of 30,000 entries.⁵ As the discipline grows, the number of subjects also increases; the Art Index has 4,773 subject headings in its 1995 volume, up from 2,258 in 1989.

Another tangible sign of the discipline’s growth is the list of journals that are classified as art history. The 1995 Art Index collates 268 journals explicitly on art history, and the BHA includes entries from about 1,125 journals in various fields; by contrast, the first volume of the Art Index, for the years 1929–1932, lists only 150 journals. Most large libraries subscribe to fifty or so journals in art history and criticism—less than a fifth of those listed in the Art Index. Even fifty is too many to read in a month, and an average art historian might subscribe to two or three, and flip through perhaps a dozen more: a first sign that our writing has outgrown us.

Given the volume of writing, it is inevitable that a great deal of attention has been paid to less well-known artists. The Art Index and the BHA list references to approximately 10,000 artists in their indices each year.⁶ Non-canonical, minor, non-Western, or otherwise neglected artists are increasingly likely to be the subjects of single volumes, monographs, and catalogues raisonnées. As of 1995, on the order of 3,000 artists had been described in monographs—substantially more than the 1,000 or so artists described in Janson’s History of Art.⁷ Along with this interest in what were once called second- and third-rate artists, the expanding literature has also treated “first-rate” artists more thoroughly than ever before. These are two different forms of interest, which I want to mark with two different phrases: I’ll call writing on less well-known artists extensive research, and writing on better-known artists and artworks intensive research. The difference is interesting and crucial, since a discipline that writes extensively is expanding its purview and experimenting with new subjects, while a discipline that grows intensively is consolidating its interests and focusing on canonical figures. Art history does both, but it largely advertises itself as doing the first; intensive research goes on more quietly, but may account for just as much of the discipline’s growth.

WRITING OUTSIDE ART HISTORY

Consider, by contrast, the state of artwriting before the later nineteenth century, and before the inception of art history as a discipline. In Greek and Roman texts, writing on individual pictures is uncommon or rare, and always succinct. Callistratus’s Descriptions and Philostratus’s Imagines, the two best-preserved sources for the ancient art of ekphrasis, each take about three hundred words to describe most images, and never more than seven hundred—that is, in most editions, from a single paragraph to about three pages per image.⁸ The Imagines itself is a short work. Outside Callistratus and Philostratus, descriptions of pictures are even more laconic; Pliny is a perfect example of extensive writing, since he covers a lot of ground but describes most images in a phrase or two. The majority of ancient texts mention paintings rather than describe them, and few offer interpretations. The exceptions, such as Pausanius’s famous description of Polygnotus’s paintings of the Nekyia and the Iliupersis, prove the rule: and even they are not long in comparison to modern writing.⁹ Pausanius’s “lengthy” account of the Iliupersis is roughly 2,000 words long; by comparison a recent review of the problem of the Iliupersis by Mark Stansbury-O’Donnell runs to nearly five times that length.¹⁰ The extra length is partly due to Stansbury-O’Donnell’s analysis of Pausanius’s prepositions (in order to reconstruct the painting, it is essential to understand what Pausanius meant by “next to,” “above,” “below,” and associated terms); but a fair amount of StansburyO’Donnell’s essay is devoted to analyzing the political meaning of Polygnotus’s composition, and showing how it balances the transgressions of the Danaans and Trojans. As always it is far from simple to decide if Pausanius thought such meanings were implicit in his text, and therefore went without saying. The structure of his narrative echoes the painting’s structure well enough so that he may have thought he had already done what Stansbury-O’Donnell did for him two millennia later. But whatever the explanation, the difference remains, and it only becomes more stark if we begin looking at other modern reconstructions of the Iliupersis.¹¹

The ekphrasis of the Iliupersis may be the longest surviving ancient description of a single image; the best-known example, Homer’s description of Achilles’s shield, is only 120 lines long.¹² There are also marginal examples such as Solensis Aratus’s poem about the constellations (which was illustrated in medieval and Renaissance editions), but the Phenomena is not a discourse on pictures in the sense I mean here. Aratus does not describe the forms of the constellations as much as the actions of the gods; and similarly, the “pictorial” descriptions in Bion, Moschus, Virgil, and other writers have only indirect and problematic relation to actual pictures.¹³

Much the same—and at the same level of generality—can be said of medieval practice. As Hans Belting and others have noted, there are numerous medieval texts on the subject of images, but they tend to consider pictures as vehicles of worship or idolatry, or as possessions of the Church or of individual churches. There are long texts associated with the Iconoclastic Controversy, some of which refer to individual images.¹⁴ Most longer documents concern the cost and commissioning of paintings or the use and abuse of religious images, but even the most intricate texts are not about individual pictures in the sense we might take that phrase today. It is commonly noted that the Renaissance was also the renascence of writing on art. As Belting puts it, when “the image became an object of reflection,” artists “found themselves driven by new kinds of arguments” that had to take account of aesthetic concepts, optical veracity, skill, invenzione, and a number of other generative concepts.¹⁵ E.H.Gombrich has argued somewhat differently that writing on art is a “leaven” of the art criticism that began with public competitions in Florence in the beginning of the fifteenth century.¹⁶ Gombrich’s and Belting’s accounts are commensurate with those that stress the rise of art academies, with their attendant literature.¹⁷ There are other ways of explaining the renascence of artwriting; for Panofsky, the increase is linked to the dawning sense of historical change, and the concomitant necessity of locating each generation’s place in the movement of culture—”le età,” as Vasari said, of art.¹⁸

These explanations partly serve different purposes, but they each give evidence of a shift in the volume of writing at the end of the middle ages, and especially of a new emphasis on intensive description. In many respects the Renaissance is the origin of our ways of writing about art: it is the period in which pictures first became the objects of intensive intellection, and were first able to bear the weight of philosophic and critical concepts. By the midfifteenth century, pictures had begun to be able to carry the burden of arguments about such seminal concepts as naturalism, relief, grazia, disegno, colorito, perspective, contrapposto, fantasia, and invenzione.¹⁹ Pictures and other artworks were widely discussed, and there were public debates, open letters, invited conferences, academies, public speeches, and philosophic defenses on the arts and on individual artists and artworks. In this context, the period from around 1435—when Alberti finished writing De Pictura—to about 1563—when the Accademia del Disegno was founded with Michelangelo as its tutelary head—is the beginning of the modern understanding of painting. After 1600, with the histories by Karel Van Mander, Joachim von Sandrart, and others, the literature on painting and drawing begins to grow exponentially, so that even Julius Schlosser’s encyclopedic Kunstliteratur can barely mention the major titles.²⁰

Still, most Renaissance writing about pictures is extensive rather than intensive, in that the treatises, letters, and dialogues become longer even though the passages on individual images remain short by modern standards. Again there are exceptions—one might name Martino Bassi’s dialogue on a relief sculpture proposed for the Milan cathedral, and Gregorio Comanini’s dialogue centered on the works of Arcimboldo—but for the most part attention is focused on the idea of painting or on the achievements of artists, rather than the meanings of individual pictures.²¹ Vasari is fairly generous in his descriptions; but even pictures that he judges to be important, exceptionally skillful, or complicated only occupy him for a page or two. Although the Vite is longer than almost any monograph that a university press might publish, comparatively little space is allotted to individual pictures: it is primarily extensive, not intensive. It is important to bear in mind just how markedly unmodern Vasari is in this regard; today two or three hundred words is scarcely enough to state the basic facts about a picture—to say something about its leading meanings, or to rehearse its title, date, size, medium, and provenance. In an average catalogue raisonné that basic information occupies a short paragraph, perhaps on the order of a hundred words (including dates, page and volume numbers, and dimensions). Vasari’s average descriptions are significantly shorter—on the order of a sentence or two.

So while Renaissance writers were the first to feel it necessary to speak at some length about pictures, their texts fall significantly short of modern standards of intensive interpretation. Over the next several centuries the extensive literature continued in full force (as Schlosser’s bibliography shows), while intensive writing became increasingly prominent. By and large the late Renaissance and Enlightenment texts that matter to art history are those that pay intensive attention to at least a few works. Winckelmann, Le Brun, De Piles, and others continue to be indispensable to the discipline’s sense of itself, while any number of “lesser” authors have fallen into oblivion at least in part because they do not engage the peculiarities of individual artworks. Thomas Da Costa Kaufmann has recently demonstrated this in the case of Winckelmann, whose immediate predecessors, writing in German and Latin in the century preceding the History of Greek Art, sometimes make strikingly “modern” points but fail to engage the reader’s interest in individual artworks.²² Reading in the large secondary and tertiary literature on art in the seventeenth and eighteenth centuries is rewarding for the light it casts on art history’s sense of its own history, but it seldom uncovers intensive engagements with artworks.

The growth of writing on art accelerated in the nineteenth century, and in the last hundred years there has been a surge in both extensive and intensive writing. The contrast with even the more extended examples of Renaissance writing has become extreme. Consider that Vasari allots a little over two pages to Leonardo da Vinci’s Last Supper, while Leo Steinberg’s essay on the Last Supper fills 113 pages—it is over sixty times longer than Vasari’s account.²³ Giorgione’s Tempesta, another painting we will be considering later in the book, attracted only two sentences—or rather, a faulty sentence and a sentence-fragment—during the sixteenth century: Marcantonio Michiel’s annotation “A little landscape on canvas with [a] storm [and a] gypsy and [a] soldier was by Giorgione”; and “another picture of a gypsy and a shepherd in a landscape with a bridge.”²⁴ Today the painting has been the principal subject of at least 3 full books, and on the order of 150 essays and other notices.

Contemporary writing is also textually intricate. Vasari does not use footnotes or illustrations aside from woodcut portraits of the artists, but Steinberg requires 29 pages of footnotes and 51 illustrations, including plans, fictive restorations of the space, cardboard models, and even a photograph of a bill-board of the Last Supper (plate 5). Vasari gives about ten pages to another central image, Masaccio’s frescos in the Brancacci Chapel. By contrast, in the last hundred years there have been at least three book-length studies on the Chapel.²⁵ In a rough count, the Brancacci Chapel inspired approximately 50 pages of writing before 1800 and on the order of 1,500 pages since then.²⁶ Steinberg’s essay—which we will read in Chapter 3—is not at all typical, but the gross amount of writing on the Brancacci Chapel is. In general, we write much more than past centuries: not just more pages, but also more pages on individual works. In many ways (and not least in the raw statistics) our practice is importantly different from anything in the past: different enough, I mean to suggest, so that it is puzzling that we do not experience the difference as a problem.

BRIEFLY, ON NON-WESTERN WRITING

The phenomena I am trying to explain in this book are Western in origin and meaning: art history developed from Western assumptions about pictures, and the voluminous responses I am going to consider have only shown themselves in writing done in Europe, America, and places influenced by their versions of art history. In another context it would be possible to expand on the hypothesis that the relevant properties of contemporary non-Western art histories are founded on Western models, and that the West continues to project its sense of Western art and history onto new material.²⁷ For such reasons the workings of pictorial complexity have resonance outside Western art, although I will not be pursuing non-Western art histories here. Even so, it is helpful to look quickly at some representative examples of non-Wes...