The Artist Historian

A decade after his death, Meyer Schapiro is remembered as the most inspiring and imaginative American art historian of the past century. Although a distinguished professor of medieval art at Columbia, he was best known as a public intellectual who was as deeply interested in the contemporary art scene as in the art of the past. His legendary lectures attracted not only students and scholars from neighboring fields but also—and more importantly—many well-known New York artists. He was immensely learned, but followed his own idiosyncratic approach. Asked how he managed to move so effortlessly in his work from the early Middle Ages to the nineteenth and twentieth centuries, he is said to have answered: “I skipped the Renaissance.” He never wrote a single great book, but his numerous articles and essays made enormous contributions to art scholarship. During the Thirties he had been drawn to Marxism—more precisely to Trotskyism—but he was never dogmatic and strongly opposed the excesses of Stalin. Above all, he believed in the autonomy of art, and viewed abstract painting as the last expression of uninhibited creativity and fantasy in a social world dominated by technology and industrial production.

I’ll never forget my first long conversation with him. It was early in 1962 on the campus at Columbia, during a snowstorm of which he seemed oblivious. I happened to mention Zeitbilder, a recent book by Arnold Gehlen, the conservative German sociologist,¹ in which Jackson Pollock’s much-discussed technique—his “all-over” method of dripping paint on a large canvas stretched out on the floor—was likened to the automated process of a factory assembly line. This provoked an explosive reaction from Schapiro, who angrily defended Pollock: his work, he said, was fundamentally about spontaneity. I learned more about Schapiro himself in that moment—even about Schapiro the medievalist—than from reading and studying all his admirable essays.

At the time Schapiro began his graduate work in the 1920s, there had long been in the United States a naive notion of the “Romanesque” style, which was associated with castles and tournaments, monasteries and pilgrimage. In the second half of the nineteenth century, neo-Romanesque buildings with round arches, heavy stonework, short squat columns, and recessed entrances—all seen in contrast to the high, pointed architecture of the Gothic churches that came later—were erected throughout New England by architects such as H.H. Richardson. In 1904, the publication of Henry Adams’s astonishing book Mont-Saint-Michel and Chartres did much to spread a violent and gloomy picture of the period in which Romanesque and Gothic art was produced. A decade later, George Barnard installed his picturesque collection of Romanesque fragments in New York. This collection, which he had assembled mainly from ruins and lapidaries in the south of France, became the core of the Cloisters, the museum constructed by the Rockefellers in Washington Heights in the 1930s that included architectural elements from several important Romanesque and Gothic buildings in France.

After World War I American interest in Romanesque art was greatly heightened by the work of Arthur Kingsley Porter, a self-taught art historian, who took up the study of Romanesque sculpture with almost religious fervor. Like a medieval pilgrim—but a pilgrim with car and camera—he roamed all over Europe, hunting Romanesque monuments. In 1923 Porter published ten volumes under the title Romanesque Sculpture of the Pilgrimage Roads from Bari in southern Italy to Santiago de Compostela. His parochial European colleagues were astonished; America was rapidly becoming a center of Romanesque studies and collections. It was in this climate that Schapiro chose Romanesque sculpture as his special field as a young art history student at Columbia.

1.

In 1926, at the age of twenty-two, Schapiro traveled to France to collect material for a doctoral dissertation on the sculpture at Moissac, the twelfth-century church near Toulouse known for its elaborate sculptures. In Paris he met French and German scholars who were interested in the same works, although they had little interest in the broader social issues that fascinated him. After his return to New York, he began a correspondence with Porter, by then a distinguished professor at Harvard, who shared his remarkable knowledge of Romanesque monuments and enormous collection of photographs with this brilliant young colleague. In 1932, Porter invited Schapiro to give a guest lecture at Harvard on medieval art.

Schapiro presented his dissertation in 1929 and two years later published the greater part of it in two path-breaking articles in The Art Bulletin. They were a stroke of genius. For the first time a great body of Romanesque sculpture became the object of a true art-historical analysis. Schapiro was not interested in conventional problems of chronology, influence, and attribution, or, as Porter had been, in relating Moissac to the sculptures on the pilgrimage road to Spain. Nor was he drawn to the religious iconography of medieval art, a subject which had been masterfully explored in the early years of the twentieth century by the great French art historian Émile Mâle. As a son of Jewish immigrants from Lithuania who grew up in modest circumstances in Brooklyn, he had had little chance to come into contact with Romanesque art and this, as it turned out, was a great advantage. With his fresh eyes, the capitals of the cloister and the sculptures of the portal at Moissac were strange and bewildering, requiring not only archaeological, antiquarian, or iconographic analysis, but also an acute sensitivity to their aesthetic qualities.

Nothing is in my view more misleading than to describe Schapiro’s approach to Moissac as merely formalist. The young art historian, who would soon become a perceptive critic of modern art, was stunned by the inventiveness and experimental character of these sculptures. Why, he asked, were they so primitive in their simplicity and at the same time so refined in their expression? By way of answer, he wrote in his first article on Moissac in 1931:

I find the essence of the style in the archaic representation of forms, designed in restless but well-coordinated opposition, with a pronounced tendency towards realism.²

Among professional medievalists Schapiro remained an exotic figure, unpredictable, independent, and inspired by elements of creative discord and surprise that had been immune to the tools of conventional analysis. He had been a member of the Young People’s Socialist League since age twelve, and his ecumenical interests constantly took him to fields and issues that were far from medieval cloisters. Within a year of publishing his “Romanesque Sculpture of Moissac” he also wrote—under a pseudonym—a very critical paper, “Architecture under Capitalism,” for the Marxist journal New Masses, in which he argued that the new approaches to social housing advocated by many architects were subject to the same capitalist forces they were ostensibly intended to neutralize.

He had become an active member of the New York left, but he sometimes seemed torn between his political convictions and his unwavering belief in the autonomy of art. During the early years of the New Deal, he took part in the debate about the social function of art, but he objected to those who advocated a rigid adherence to the doctrines of social realism.³ In his papers on Moissac and other monuments he tried to link art to surrounding social situations. Yet in a 1936 essay, he warned, “When I speak…on the social bases of art I do not mean to reduce art to economics or sociology or politics.”⁴

Soon after receiving his Ph.D., Schapiro joined the art history faculty of Columbia and initially began to publish rather dry academic articles on Romanesque sculpture. But in 1939 he made a radical break from earlier scholarship in two brilliant articles, “The Sculptures of Souillac” and “From Mozarabic to Romanesque in Silos,” in which he sought to analyze French and Spanish Romanesque sculpture in relation to social change in France and Christian Spain. Earlier, writers such as Victor Hugo and Eugène Viollet-Le-Duc had attempted to combine studies of art and society, but their ideas had been influenced by anticlericalism and were long forgotten by Schapiro’s time.

The lengthy articles on Souillac and Silos are Schapiro’s most innovative and widely praised writings on Romanesque sculpture. Both were based on the fieldwork he had undertaken years before in his dissertation research; but they also reflected a great deal of imaginative thinking done far from the medieval monuments in Europe, in the intellectual and political ferment of 1930s New York. The article on Souillac begins with a brilliant and subversive description of the large relief on the inner west wall of the church, which depicts the legend of Theophilus, an early Christian cleric who is said to have made a pact with the devil to regain his position as vicedominus (bishop’s administrator) and was saved by the intervention of the Virgin. With its chaotic intermingling of human figures, saints, and beasts, Schapiro observed,

The expected Romanesque adherence to an embracing architectural frame is violated throughout the work. But a more detailed analysis will show that the apparently “accidental” design is a deeply coherent arrangement, even systematic in a sense, and similar to other mediaeval works.

In short, Schapiro argued, the whole sculptural ensemble was such that corresponding elements were “discoordinate”—they directly opposed the usual principles of order, symmetry, and organization around a single axis. With his powerful imagination and persuasive rhetoric Schapiro catapulted Romanesque sculpture into the conflicts, restlessness, and chaos of the modern age. He was particularly drawn, for example, to what he called the sculptors’ “passionate feeling for expressive pantomime.” And he viewed the sculptures themselves as powerfully and imaginatively drawn from nature, as for example in his description of the animal imagery of the famous trumeau, or carved pillar:

A remarkable intrigue animates these opposed predacious beasts. They are of different species, the one a monstrous quadruped with the wings and head of a bird, the other a kind of Romanesque lion. Their foreparts are mounted on each other’s backs, but their bodies are turned in opposite directions, as if awaiting other prey. Then reversing the divergent movements of their bodies, they turn their heads and find a common prey behind their crossed backs.

Such prose explains the extraordinary, sensational success of the Souillac study. No one else had ever written on Romanesque sculpture with this intensity.

But there was a specific polemic in the essay that has until now been overlooked. Schapiro’s insistence on “discoordination” was also an attack against the most authoritative explanation of the principles governing Romanesque sculpture. In a book published in 1931, Henri Focillon, then the dean of French art historians, argued that the organization of sculptures was predetermined by what he called the loi du cadre, the “law of the frame”: sculptural ensembles were seen as composed in strict relation to the shape of the spaces they occupied. Schapiro’s “discoordination” boldly challenged Focillon’s rappel à l’ordre. It was New York against Paris! Schapiro had created in a single stroke a genuinely American vision of Romanesque sculpture.

Still, the article’s true ideological underpinnings—and its real provocation—lay elsewhere. With astonishing boldness, Schapiro applied his earlier Marxist interpretations of contemporary American society to ecclesiastical art from the feudal age. He regarded “discoordination” as not only a principal of formal organization but also a “conception of content” that captured an emerging tension between secular and religious culture in twelfth-century France. On the large relief at Souillac, for example, he observed that the Virgin, angels, and saints in the story have been displaced from the central field, which is occupied instead by Theophilus and the devil:

Theophilus is a layman, whose peace with the devil and change of fortune have an essentially secular nucleus, comparable to the later adventure of Faust. The antitheses of rank and privation, of the devil and the Virgin, of apostasy and repentance, create a psychological depth—the counterpart of a world of developing secular activity and freedom, more complex than the closed field of Christian piety represented in the dogmatic images of the majestic Christ on Romanesque portals.

Seen in these terms, “discoordination” became an expression not only of secular society but of the new “freedom of the artist.” In the rapacious beasts on the trumeau Schapiro discerned a “non-religious fantasy” whose wild composition reflected the “role of violence at this point in the history of feudal society.” This sounds a lot like the world of castles and knights, battles and tournaments, that had been popularized in earlier American literature—only now it was a historical vision with critical rather than nostalgic undertones. No wonder his study of Souillac has been admired by art historians for decades and contributed much to his enduring fame.

The view that art captures the physiognomy of the age in which it is produced—the general character and conflicts of a particular time and place—was widespread in the early decades of the twentieth century, particularly in Europe. Schapiro, as a voracious reader in many disciplines, was well aware of this approach, and in 1936 wrote an illuminating article on the “New Viennese School” of art history, with Hans Sedlmayr as its most brilliant but also dangerously enchanting mind.⁵ The perspicacity of Schapiro’s formalist analysis remains unsurpassed. But the “discoordination” he observed on the large relief owes—as we now know—not to the violent imagination of its sculptors but to the random reassembly of the surviving fragments after the church was sacked by the Huguenots in 1562. With this painful error Schapiro paid the price for his imaginative New York deskwork. It amounted to a kind of wishful art-historical illusion.

Of much greater interest are Schapiro’s daring conclusions about the social content of the sculptures their relation to the society in which they were produced. His argument that these twelfth-century monastic sculptures reflect a growing secular ethos and a new sense of freedom is alluring to nineteenth- and twentieth-century notions of enlightenment and social progress. But against such an interpretation stands the forbidding reality that the wall at Souillac is also a stern and frightening sermon in stone whose message is thoroughly religious—a warning against the sin of simony, the buying and selling of ecclesiastical offices, which had been a serious concern since the eleventh century. The incompletely preserved story of Theophilus, for example, refers to Mariolatry—not surprising in a church dedicated to the Assumption of the Virgin. The terrifying beasts on the pillar arise not from some secular fantasy but from the metaphors of the psalms, which formed the main part of the daily prayers of the monks in the choir stalls.

What we encounter in the sculptures of Souillac, then, is not an emerging secularism but rather religious art expressed with new ardor and narrative intensity. Schapiro’s article remains a landmark, but it teaches us at least as much about Marxist debates in New York during the 1930s as about the eight-hundred-year-old sculptures that were his subject. Schapiro himself later turned away from such dramatic social interpretations. But there was one aspect of this analysis to which he clung right up to the end: the idea that artists in the Romanesque period expressed a profound new sense of freedom and autonomy. So we are back in the snowstorm of 1962 and to Schapiro’s insistence about Jackson Pollock’s spontaneity.

Schapiro’s 1947 article “On the Aesthetic Attitude in Romanesque Art” displayed a new lightness and grace. During World War II he had written less frequently on politics, and his interest in studying art in relation to broad patterns of social change seems also to have diminished. Although he wrote important articles on medieval art, he did nothing substantial on Romanesque sculpture until this essay, at which point his focus had changed. Now he stressed what he saw as a changing understanding of beauty and creativity that bore similarities to contemporary, twentieth-century ideas about art and the artist. “I shall try to show,” he wrote,

that by the eleventh and twelfth centuries there had emerged in western Europe within church art a new sphere of artistic creation without religious content and imbued with values of spontaneity, individual fantasy, delight in color and movement, and the expression of feeling that anticipate modern art. This new art, on the margins of the religious work, was accompanied by a conscious taste of the spectators for the beauty of workmanship, materials, and artistic devices, apart from the religious meanings.

Despite the religious purpose for which the art was created, Schapiro argued that “the respect for the artist sometimes outweighed the fundamental religious rules.” These statements seem to echo his earlier thesis about secularization, but secularization was now viewed less with respect to social than to aesthetic liberation. The art historian who “skipped the Renaissance”—who took Fernand Léger to the Morgan Library to show him a medieval Beatus Manuscript—celebrated Romanesque sculpture as the beginning of modern artistic expression. This article seemed to be Schapiro’s final word on medieval art. With the exception of a short paper on a relief in Rodez, which he gave in 1961 (and which may have been a last morsel from his early expedition to Moissac) he published not a single line on the subject over the next twenty years.

2.

In 1967, at the age of sixty-two, Schapiro was invited to give the Charles Eliot Norton Lectures at Harvard. At this point in his career one might have expected him to speak about Cézanne or Seurat, artists about whom he had written brilliantly. Instead, he decided to return to the subject that had animated his early work: he dedicated the seven lectures, which he gave in February and March of that year, to “Romanesque architectural sculpture.” As was his habit, he spoke without a manuscript, using only a few handwritten notes and his marvelous gift for oral performance. As a result, there was no written text of the lectures, only recordings. These were transcribed, and beginning in the 1970s Schapiro set out to expand and annotate them for publication. At his death in 1996 he left this formidable task unfinished. Like Cézanne, whom he admired so much, Schapiro never believed that a life’s creative efforts could culminate in a single all-encompassing work. All his papers were to some degree experiments, or works in progress. With the Norton Lectures he faced the additional problem that the field had grown much larger since the 1930s and continued to grow rapidly in the years after he gave them. This growing body of scholarship does not obscure the originality of Schapiro’s own ideas, but it has greatly expanded and deepened our knowledge of the monuments treated in the lectures.

We must be grateful to Dr. Lillian Milgram Schapiro, his wife, for persisting in the plan to publish the lectures, a project that meant so much to Schapiro. It was a difficult process. In 2000, after Harvard University Press turned down the manuscript, Professor Linda Seidel, a distinguished scholar of Romanesque art, took over the delicate task of editing the text, which has now been published by the University of Chicago Press. Seidel herself had attended Schapiro’s lectures as a junior member of the Harvard faculty and had a cherished and enthusiastic memory of them. The challenge was how to render them, and Schapiro’s subsequent work on them, into a text that could be read in printed form. Seidel decided to make no changes to Schapiro’s language, although there are frequent repetitions and the grammar is more than once—charmingly—casual.

Her decision was certainly right. In reading these pages it seems possible to hear Schapiro’s own voice, even his laughter or the applause of the audience. It is almost like listening to a recording. But the book will not be easy going for nonspecialists. There are no footnotes indicating the sources Schapiro used for his numerous quotations from medieval texts, and the publisher—apparently for financial reasons—has only included a small number of illustrations. It would have been helpful to provide a list of easily accessible reproductions of the many monuments mentioned in the text but not illustrated. Still, at least we have Schapiro’s final words on the subject that first inspired his interest in art history seventy years ago.

The late contributions, or Alterswerk, of a prominent scholar often have the effect of seeming to echo earlier research. Many of the ideas that Schapiro developed in his Norton Lectures are variations of thoughts and insights he had first elaborated in the years between the completion of his dissertation in 1929 and his 1947 essay “On the Aesthetic Attitude in Romanesque Art.” But the art scene in New York—and the general social and political climate—had changed enormously. By 1967, his reflections on art and social change had long been relegated to the past and Schapiro himself had written variously on Leonardo, Freud, and “The Liberating Quality of the Avantgarde.”

In places, the lectures seem to express a kind of nostalgia for intellectual experiences that took place long before. He invites his listeners into a fantastic world, one that seems to have become somewhat distant to Schapiro himself. The old themes are still there—the spontaneity and imaginativeness of Romanesque art among them—but his metaphors have become more detached, less concise, and softer. He moves agilely around a larger group of monuments, but France is the main focus, with Moissac, Saint-Gilles, and—astonishingly—also Chartres, usually regarded as a classic work of Gothic architecture. For readers familiar with Schapiro’s earlier studies, little of this is new. But there is something else: some of the most sensitive descriptions of Romanesque sculptures ever written.

In his first lecture Schapiro deals with “The Rebirth of Monumental Sculpture in the West: Disappearance and Rebirth.” This fundamental problem had not occupied his earlier studies, but since Ghiberti in the fifteenth century, it had been one of the great enigmas of art history. Why did the production of architectural sculpture come to a halt in late antiquity and why did it reappear in the late eleventh century? Discounting the idea that its abandonment was a result of the early condemnation of pagan idolatry or of a collapse of artistic values caused by the barbarian invasions, Schapiro argued that the revival resulted from a devaluing of the civic spaces in which the Roman imperial state had formerly erected public monuments such as triumphal arches or columns. In other words, the decline of sculpture was owing not to religious or aesthetic developments, but to social and political change.

Similarly, the rebirth—or return—of monumental sculpture during the Romanesque period, half a millennium later, he argued, was the result of a new conception of public space. The outer walls of a church, for example, now belonged to an “exterior, public world” in which

people are free to move, to regard things according to their own inclination or position at the moment, and to shift their point of view and to experience them without regard to a prescribed occasion and liturgy.

Thus liberated from controlled religious settings such as altar antependia, ciboria, candlesticks, and crosses in the interior of early medieval monuments, sculptors were free to create works such as church portals “that seem to be of an artisan character which have for us the highest finesse and power of artistic invention and sensibility within that time.” One immediately recognizes Schapiro’s old ideas from the 1930s, but he appears to be no longer certain of his earlier conclusions. Only once does he use the term “secular,” and he even makes reference to the “sermon” of a Romanesque portal. But he continues to put forward the notion that much of Romanesque sculpture is imbued with nonreligious content.

In the next two lectures, Schapiro spoke about “Field, Figure, and Frame,” a subject he had dealt with in his paper on Souillac twenty-eight years before. But now he shapes his argument with reference to his intervening work on modern and contemporary art. He starts by discussing the “principle of freedom”—the idea that the relation of images of sculptures to their frame or surrounding architecture is one that encourages rather than limits artistic expression. This is a principle, in Schapiro’s view, that is commonly expressed in modern and contemporary painting. He cites Renoir, who “told his son that Cézanne was the greatest French artist since the Romanesque sculptors.”

These two chapters are filled with scintillating descriptions of Romanesque monuments. Nobody has written a more illuminating analysis of the tympanum—or sculpted recess—in the narthex of Vézelay, which Schapiro likens to “a sort of planetary system in which there is rotation as well as revolution with reference to a dominant or a center.” No less seductive is his description of the tympanum at Autun, with its representation of the Last Judgment:

…The whole has a rather panoramic and episodic character, with a scattering of figures of many different sizes and with axes which are not those of architecture, but axes of human passion, fear, reverence, struggle, anxiety.

From these and other observations, Schapiro concludes that Romanesque sculptors enjoyed a latitude in their aesthetic choices and expression that looked forward to the work of modern artists many centuries later. “The forms of buildings, boundaries, and frames,” Schapiro says,

are a material with which he works freely. He can bend them, twist them, shorten them, lengthen them, amputate them, drape the figure about them, place them in and out of the other figures. The sculptor has, then, what Horace [called] the potestas audendi, the power of daring….

The following two lectures, on “The Programs of Imagery (i): Themes of Action and Themes of State” and “(ii): Tradition, New Reality, and Nature” do not have the same power and strength. “Programs of Imagery” is a revealing title, and it has a sharp polemic edge. In contrast to religious iconographers such as Émile Mâle and also to Erwin Panofsky, who sought to connect the image with the word, Schapiro argued that Romanesque depictions sprang from the imagination of the artists themselves. Unlike many of his fellow scholars, he had a sensitivity to nonliterary qualities that allowed him to appreciate the wildness and violence of Romanesque sculpture and its particular vein of humor. He was fascinated by “representations of festivity, play, hunting, and war, also themes of parody and humor.” He concludes these two lectures by describing the Romanesque “Programs of Imagery” as

a totality of ideas and images in change, in movement, in process, and assimilating, in a bold way, ideas and experiences from the actuality of the time and introducing new norms of adequacy of images, norms which require also action, individualization, contemporary reality.

This overloaded sentence gives an impression of the breathlessness; we sense the drive of Schapiro’s rhetoric but also the difficulty of transcribing his words in cold print.

More than anyone else Schapiro opened our eyes to the expressive freedom of Romanesque sculpture. But with such generalizations as the passage just quoted, his argument risked becoming shallow and redundant. With all his modernist enthusiasm for the freedom of the artist, Schapiro seemed to forget that Romanesque sculpture was fundamentally a religious, ecclesiastical, and monastic art. Its violent and fantastic images were an expression of the moral tension between the severe strictures of the Gregorian Church order and a growing worldliness—a tension between sanctity and sin, in which the sinful was defined religiously and should not be confused with the secular.

In lecture VI, Schapiro complains about the poverty of language for describing the contradictory postures, movements, and gestures that can be observed in Romanesque figures. “We are constantly confronted by the difficulty,” he reflects, “of distinguishing those features in a representation of the human figure which belong to nature and those which belong to style or to artistic form.” Schapiro is here returning to problems which he had tackled in his dissertation on Moissac and to the lessons he had learned from The Rendering of Nature in Greek Art (1907), a well-known book by the Viennese archaeologist Emanuel Löwy. Löwy had argued that in Greek art primitive conceptual images were gradually adjusted to natural appearance, a point that Schapiro elaborates and expands in his lecture by analyzing the bizarre poses of Romanesque sculptural figures. Regarding the famous figure of Eve at Autun he observes that the different elements of her story that are brought into the image

are also adapted to one another in terms of qualities that are not simply those of representation, but belong to an independent artistic structure of line as melodic in quality, as sinuous, as wavy.

He is entranced by what he calls the “duality” of “the Romanesque saint, who [often] appears unstable, torn, divided, the head moving one way, the body the other way, the knees bent in different fashion.” Such sensitivity for the expressive qualities of sculpture seems at times almost to turn Schapiro’s prose into verse, transforming the academic lecturer into a dithyrambic bard. In art-historical scholarship rapture is a dangerous, even devastating device, but Schapiro had the rare capacity to balance precision and passion. Yet from today’s perspective, this remarkable style of performance seems dated; academic lecturers now strive to be both measured and telegenic.

Schapiro’s last lecture was devoted to “Animal Imagery in Romanesque Sculpture.” It is lively but challenging. He dismisses the common interpretation that the many beasts and animals on Romanesque monuments are taken from religious texts. In some cases, such as the tympana at Moissac or at Jaca in northeastern Spain, he concedes that these images are derived in part either from the text of the Bible or by moralizing inscriptions identifying sins such as “unchastity and avarice.” But he cites the example of the extraordinary beasts on the portal of the church at Aulnay, in southwestern France, to argue against a general theory of literary derivation:

Above, a figure, somewhat obscene looking, facing outward, resting upon a post; a bird in whom there is a human head stuck in its own neck and peering outward…. We are confronted, then, by a process of free creation, of fantastic objects with qualities of the hybrid, in some cases of the violent, and which go beyond nature or any text that is familiar to us.

Schapiro was much too learned not to know that most of these hybrid animals can indeed be traced to ancient texts and reappear in medieval encyclopedias and bestiaries. But he was so concerned to resist reducing such wild or fantastic images to mere representations of texts that he simply dismissed such connections.

It would be absurd to deny what Schapiro calls “the taste for violence or strength, for power and conflict” in these images. Romanesque portals, capitals, consoles, and manuscript pages are covered with frightening, voracious creatures. The question is: What kind of “terror” is expressed in this bestial imagery? What kind of anxiety does it reveal? Schapiro was surely correct in his view that these aggressive beasts are not symbols. There is a strong basis, however, for supposing that they express the fear and threat of hell, of demons, of the devil, and of divine retribution for sin. They belong, then, to the dark side of Romanesque religious imagery and not—alas—to new “fields of freedom, a space in which fantasy is unconstrained by the requirements of a religious text,” as Schapiro wanted to believe.

Since its rediscovery in the eighteenth century, medieval art has often provided a fertile ground for the modern imagination. The German Nazarenes shed tears about the piety and the innocence of late medieval painters and the English Pre-Raphaelites followed them on this path. Émile Mâle began the introduction to his book on French religious art of the thirteenth century with the words: “Le moyen âge eut la passion de l’ordre” (“The middle ages had the passion of order”). German art historians exclaimed about the power of the Kaiserdome (imperial cathedrals) and of the fortresses erected by the Hohenstaufen. But all of these various accounts were to some extent based on the idealistic perspective afforded by the notion that one belonged to an enlightened modern age and could look back with nostalgia on the closed world of the Middle Ages. Nothing could be more opposed to these concepts than Schapiro’s idea of “discoordination” in Romanesque sculpture. At the center of his vision was his enthusiasm for artistic freedom. He saw the sculptures of Moissac and Vézelay in the light of modern art. That may have been the illusion of a particular moment in the past century. Much of it does not stand up to criticism. But I know of no more moving, more sympathetic reading of medieval art.