Cover image for The Surviving Image: Phantoms of Time and Time of Phantoms: Aby Warburg's History of Art By Georges Didi-Huberman and Translated by Harvey Mendelsohn

The Surviving Image

Phantoms of Time and Time of Phantoms: Aby Warburg's History of Art

Georges Didi-Huberman, and Translated by Harvey Mendelsohn


$99.95 | Hardcover Edition
ISBN: 978-0-271-07208-1

$40.95 | Paperback Edition
ISBN: 978-0-271-07209-8

432 pages
7" × 10"
93 b&w illustrations

The Surviving Image

Phantoms of Time and Time of Phantoms: Aby Warburg's History of Art

Georges Didi-Huberman, and Translated by Harvey Mendelsohn

“A painstakingly detailed examination of Warburg’s writing, showing its connection with works of Friedrich Nietzsche, Jakob Burckhardt, and other notable pioneers of a psychologically oriented analysis of culture. Didi-Huberman offers insightful readings and criticism of such scholars as Ernst Cassirer, Erwin Panofsky, and Ernst Gombrich, accused of trafficking in ‘canons, ideal entities and transcendental notions,’ not to mention a more general denunciation of unnamed practitioners of a ‘positivist’ art history.”


  • Description
  • Reviews
  • Bio
  • Table of Contents
  • Sample Chapters
  • Subjects
The Surviving Image, originally published in French in 2002, is the result of Georges Didi-Huberman’s extensive research into the life and work of foundational art historian Aby Warburg. Warburg envisioned an art history that engaged with anthropology, psychoanalysis, and philosophy in order to understand the “life” of images. Drawing on a wide range of Warburg’s unpublished letters and diaries, Didi-Huberman demonstrates unequivocally the complexity and importance of Warburg’s ideas and the ways in which his legacy was both distorted and diffused as art history became a “humanistic” discipline. The Surviving Image takes Warburg as its main subject but also addresses broader questions regarding art historians’ conceptions of time, memory, and symbols and the relationship between art and the rational and irrational forces of the psyche.

Faithfully and thoughtfully translated by Harvey Mendelsohn, this first English-language edition of Didi-Huberman’s masterful study of Warburg is a stirring and significant treatise on the philosophical nature of art history.

“A painstakingly detailed examination of Warburg’s writing, showing its connection with works of Friedrich Nietzsche, Jakob Burckhardt, and other notable pioneers of a psychologically oriented analysis of culture. Didi-Huberman offers insightful readings and criticism of such scholars as Ernst Cassirer, Erwin Panofsky, and Ernst Gombrich, accused of trafficking in ‘canons, ideal entities and transcendental notions,’ not to mention a more general denunciation of unnamed practitioners of a ‘positivist’ art history.”
“For anyone interested in the intersections of art, material and visual culture, anthropology, and philosophy, The Surviving Image stands as essential reading, long overdue to any English-language reader familiar with Giorgio Agamben, Philippe-Alain Michaud, and Christopher Johnson’s writings on Warburg. Didi-Huberman’s book may itself feel ‘untimely’ in a risk-averse intellectual age; this fact only serves to recommend it, and Aby Warburg’s original work, all the more.”
“It is wonderful to have this book in English. Didi-Huberman's magnum opus is an impassioned, critical love letter to the discipline of art history, which he considers as ossified by ‘systematic and reassuring’ methodologies, ‘territorialized,’ and bound by institutional expectations, but since Warburg also potentially open to the numinous, unsystematic, undependable, inconstant moments in which representation and desire present themselves as art history. Of the many attempts to revive Warburg, this is the only one that takes the consequences of a full engagement seriously.”
“Didi-Huberman argues that Warburg offers offers us a ‘psycho-history’ of culture: a model of historical understanding as a response to images possessed of vital power and emotional force. This book fills the need for a better understanding of Warburg’s contribution to the discipline of art history, and will draw the attention of anyone who teaches its history and methods as well as that of students who seek to understand the intellectual life of their chosen field of study.”
“When Georges Didi-Huberman’s The Surviving Image was first published in French, it transformed the image of Warburg as the scholar immersed in the arcana of Renaissance magic, art, and philosophy, and he gained his place as a theorist, urgently questioning the nature of inquiry into art history and visual culture. Ostensibly about Warburg, the range and significance of this key work of art theory and historiography is far wider, for it deals with important philosophical questions to do with art, memory, time, and the construction of art-historical knowledge. Harvey Mendelsohn has created a lucid and elegant translation, an admirable accomplishment that will ensure that this book gains the wider readership it deserves.”

Georges Didi-Huberman is on the faculty of the École des hautes études en sciences sociales in Paris. His books in English include Fra Angelico: Dissemblance and Figuration; Invention of Hysteria: Charcot and the Photographic Iconography of the Salpêtrière; and Confronting Images: Questioning the Ends of a Certain History of Art, the last also published by Penn State University Press.

Harvey Mendelsohn is a professional translator of French and German and the proprietor of H. L. Mendelsohn Fine European Books, an antiquarian book shop specializing in scarce and out-of-print books on the history of architecture, city planning, landscape, and the decorative arts.


List of Illustrations


Translator’s Note

Chapter 1: The Image as Phantom: Survival of Forms and Impurities of Time

Chapter 2: The Image as Pathos: Lines of Fracture and Formulas of Intensity

Chapter 3: The Image as Symptom: Fossils in Motion and Montages of Memory




Chapter 1

The Image as Phantom: Survival of Forms and Impurities of Time

<1>Art Dies, Art is Reborn: History Begins Again (From Vasari to Winckelmann)

One may ask if art history—the type of discourse that goes by that name, Kunstgeschichte—was really “born” one day. At the very least, let us say that it was never born all at once, on one or even on two occasions that could be considered “birth dates” or identifiable points in a chronological continuum. Behind the year 77 and the dedicatory epistle to Pliny the Elder’s Natural History there stands, as is well known, an entire tradition of Greek historiography. Similarly, behind the year 1550 and the dedication in Vasari’s Lives there stands, revealing its residues, a whole tradition of chronicles and eulogies written for the uomini illustri of a city like Florence.

Let us boldly assert the following: historical discourse is never “born.” It always recommences. And let us observe this: art history—the discipline which goes by that name—recommences each time. This happens, it seems, each time that the very object of its inquiry is experienced as having died … and as undergoing a rebirth. This is exactly what happened in the 16th century, when Vasari established his entire historical and aesthetic enterprise upon the observation that ancient art had died. He writes of the voracità del tempo in the proemio of his book, and then points to the Middle Ages as the real guilty party in this process of forgetting. But, as is well known, the victim was “saved,” miraculously redeemed or ransomed from death by a long movement of rinascità that, in broad outline, begins with Giotto and culminates with Michelangelo, recognized as the great genius in this process of recollection or resurrection. It appears that starting from this point—starting from this renaissance that itself emerged from a state of mourning—something could emerge calling itself the history of art (Fig. 1).

<insert fig. 1 about here>

Two centuries later, the whole thing starts again (with some substantial differences, of course). In a context that is no longer that of the “humanist” Renaissance but rather that of the “neoclassical” restoration, Winckelmann invents art history (Fig. 2), by which we mean art history in the modern sense of the word “history.” In this case, art history emerges from an age of Enlightenment, soon to become the age of the grand systems—Hegelianism in the first place—and of the “positive” sciences in which Michel Foucault sees the two concomitant epistemological principles of analogy and succession at work. That is to say, the phenomena are systematically apprehended according to their homologies, and the latter, in turn, are interpreted as the “fixed forms of a succession which proceeds from analogy to analogy.” Winckelmann, whom Foucault unfortunately does not discuss, may be seen as representing, in the domain of culture and beauty, the epistemological change within thinking about art in the age of history —a history that is now authentic, already “scientific.” The history under consideration here is already “modern,” and already “scientific” in the sense that it goes beyond the simple chronicle of the Plinian or Vasarian type. It aims at something more fundamental, which Quatremère de Quincy, in his eulogy of Winckelmann, rightly called an analysis of time:

The scholar Winckelmann is the first to bring the true spirit of observation to this study. He is the first who dared to decompose Antiquity, to analyze the times, the peoples, the schools, the styles, the nuances of style. He is the first who laid out the road and set out the markers in this unknown land. He is the first one who, in classifying the various periods, included the history of the monuments, and compared the monuments with each other, discovered reliable characteristics, principles of criticism, and a method that, in correcting a mass of errors, prepared the discovery of a mass of truths. And, finally, when turning from analysis to synthesis, he succeeded in creating a body out of what had been only a pile of debris.

<insert fig. 2 about here>

The image is significant. While the “piles of debris” continue to be strewn around on the ground and underground in Italy and Greece, Winckelmann, in 1764 published a book, his great History of Ancient Art, that, in Quatremère’s phrase, “creates a body” from this whole dispersion. A body: an organic joining of objects, the anatomy and physiology of which amounts to something like the joining of artistic styles and the law of their biological functioning, that is to say, of their evolution. And also a “body”: a corpus of knowledge, an organon of principles, in fact a “body of doctrine.” Winckelmann thus invented art history, in the first place, by going beyond the simple curiosity of the antiquarians and constructing something like an historical method. The historian of art will no longer be satisfied just to collect and admire its objects. As Quatremère writes, he will analyze and decompose, exercise his observational and critical powers; he will classify, bring together and compare, and “turn from analysis to synthesis” in order to “discover the reliable characteristics” that will give to every analogy its law of succession. And this is how the history of art assumes the form of a “body,” of a methodical domain of knowledge, of a genuine “analysis of time.”

The majority of commentators have been aware of the methodological, even doctrinal aspect of this elaboration. Winckelmann establishes an art history less through what he discovers than through what he constructs. It is not sufficient simply to see Winckelmann the “aesthetic critic” of the Thoughts on the Imitation of Greek Works as being succeeded by Winckelmann the “historian” of the History of Ancient Art; for there is no doubt that the “aesthetic crisis” of the Enlightenment affected even the manner in which he assembled his basic archaeological material. In reading the exegeses of this body of work, one also senses a certain theoretical discomfort caused by a contradictory figure who, on the one hand, is supposed to be the creator of a history, and, on the other, is a zealous advocate of an aesthetic doctrine. One can not simply say that this contradiction “is only apparent.” One must say, rather, that it is constitutive. As Alex Potts has demonstrated, the History of Ancient Art establishes the modern perspective of knowledge of the visual arts only by means of a series of paradoxes in which the historical position is, time and again, composed of “eternal” postulates, and in which the general conceptions, in turn, are overthrown by their own historicization. These contradictions are far from delegitimizing the historical enterprise underway here—that is something that only a positivist or naïve historian would believe, imagining the existence of a history drawing its presuppositions only from its own objects of study. Rather, these contradictions are literally the enterprise’s foundation.

How should this framework of paradoxes be understood? It seems to me insufficient, indeed impossible, to separate within Winckelmann’s work various “levels of intelligibility”; for they are so different that, in the end, they would form a major contradictory polarity, with one term being the aesthetic doctrine, the atemporal norm, and the other, historical practice, the “analysis of time.” This separation, conceived literally, would wind up rendering the very expression “art history” incomprehensible. At the very least, one can easily sense the eminently problematic nature of that expression: what conception of art must it imply in order that one might write its history? And what conception of history must it imply in order that one might apply it to works of art? The problem is difficult because the whole thing hangs together; that is to say, the position one takes with regard to a single element obliges one to take a position on all the others. There is no history of art without a philosophy of history—be it merely spontaneous and not really thought out—and without a certain choice of temporal models; just as there is no history of art without a philosophy of art and without a certain choice of aesthetic models. One must try to see how, in Winckelmann, these two types of models work in concert. This would allow one, perhaps, better to understand the dedication placed at the end of the prologue to the History of Ancient Art—“This history of art is dedicated to art and to time”—the almost tautological character of which remains something of a mystery to the reader.


Books are often dedicated to the dead. Winckelmann first dedicated his History of Art to ancient art, because, in his eyes, ancient art had long been dead. Likewise, he dedicated his book to time, because the historian, in his eyes, is the one who walks in the realm of things that have passed, that is to say, that have passed away. Now, what happens at the other end of the book, after several hundred pages in which ancient art has been recollected for us, reconstructed—in the psychological sense of the term—put in the form of a story? We sense a kind of depressive latching on to a feeling of irremediable loss and to a terrible suspicion: is the object whose history has just been told not simply the result of a phantasmic illusion by means of which this feeling, or the loss itself, may well have misled us.

In meditating upon its downfall [I] have felt almost like the historian who, in narrating the history of his native land, is compelled to allude to its destruction, of which he was a witness. Still I could not refrain from searching into the fate of works of art as far as my eye could reach; just as a maiden, standing on the shore of the ocean, follows with tearful eyes her departing lover with no hope of ever seeing him again, and fancies that in the distant sail she sees the image of her beloved (das Bild des Geliebten). Like that loving maiden we too have, as it were, nothing but a shadowy outline left of the object of our wishes (Schattenriss ... unserer Wünsche), but that very indistinctness awakens only a more earnest longing for what we have lost, and we study the copies (Kopien) of the originals more attentively than we should have done the originals (Urbilder) themselves if we had been in full possession of them. In this particular we are very much like those who wish to have an interview with spirits (Gespenster), and who believe that they see them when there is nothing to be seen (Wo nichts ist).

A formidable page—its very beauty and its poetry are formidable. And a radical page. If art history recommences here, it is defined as having for its object a fallen object, one that has disappeared and been buried. Ancient art—the art of absolute beauty—thus shines in its first modern historian by a “categorical absence.” The Greeks themselves, at least in Winckelmann’s mind, never produced a “living” history of their art. That history begins, first revealing its necessity, at the very moment in which its object is conceived as a dead object. Such a history, therefore, will be experienced as a work of mourning (the History of Ancient Art is a work of mourning for ancient art) and a hopeless evocation of what has been lost. Let us immediately stress this point: the phantoms that Winckelmann speaks of will never be “convoked” or even “invoked” as powers that are still active. They will simply be evoked as past powers. They will be the equivalent of “nothing” that is existent or actual (nichts ist). They represent only our optical illusion, the lived time of our mourning. Their existence (be it only spectral), their survival, or their return are quite simply not envisioned.

This, then, is what the modern historian becomes: someone who evokes the past, saddened by its definitive loss. He no longer believes in phantoms (soon, in the course of the 19th century, he will no longer believe in anything but the “facts”). He is a pessimist, often employing the word Untergang, which signifies decline or decadence. In fact, his entire enterprise seems to be organized according to the temporal schema of greatness and decline. No doubt Winckelmann’s enterprise should be set in the context of the “historical pessimism” characteristic of the 18th century. We should note, too, that, in the aesthetic domain, Winckelmann’s ideas inspired innumerable nostalgic writings on the “decadence of the arts,” and even the “revolutionary vandalism” linked to the successive destructions of ancient masterpieces. The temporal model of greatness and decline proved to be so potent that it long informed the definition of art history, as can be seen, for example, in the Brockhaus Real-Encyclopädie: “Art history is the representation of the origin, the development, the grandeur and the decadence of the fine arts.” This is no different from what Winckelmann said:

The History of Art is intended to show the origin (Ursprung), progress (Wachstum), change (Veränderung), and downfall (Fall) of art.

Careful attention to this schema reveals that it is linked to two types of theoretical models. The first is a natural model, more precisely, a biological one. In Winckelmann’s sentence the word Wachstum should be understood as “growth,” vegetable or animal, and the word Veränderung also takes on the vitalist connotation implied by any notion of “mutation.” What Winckelmann means by art history is thus not fundamentally very different from natural history. It is known, of course, that he read the one by Pliny, but he also read Buffon’s; just as he read the treatise on physiology by J. G. Krueger, and the medical manual by Allen. And, as we learn from a letter dated December 1763, he wanted some day to move from “studies of Art” to “studies of Nature.” From all that, Winckelmann drew a conception of historical science articulated not only around the typical classificatory problems of Enlightenment epistemology, but also around a temporal schema that is obviously biomorphic, constructed between the poles of progress and decline, birth and decadence, life and death.

The other side of this theoretical configuration is better known: it is an ideal model, and, more specifically, a metaphysical one. It thus accords very well with the “categorical absence” of its object. Think of Solon’s famous saying, reported by Aristotle —to ti en einai—which assumes the prior death of that about which one wants to pronounce the truth, or, better, the “quiddity.” In this sense, we could say that the very disappearance of ancient art founds the historical discourse capable of telling its ultimate quiddity. The history of art as Winckelmann conceives it, therefore, is not satisfied with describing, or classifying, or dating. While Quatremère de Quincy speaks of a simple movement backwards “from analysis to synthesis,” Winckelmann himself radicalized his position from the philosophical point of view: the history of art (die Geschichte der Kunst) should be written in such a way that it makes explicit the essence of art (das Wesen der Kunst).

The History of Ancient Art which I have undertaken to write is not a mere chronicle of epochs, and of the changes which occurred within them. I use the term History (Geschichte) in the more extended signification which it has in the Greek language; and it is my intention to attempt to present a system (Lehrgebäude).... The History of Art (die Geschichte der Kunst) in a more limited sense, is [the history of its development] as far as external circumstances were concerned, but only in reference to the Greeks and Romans.... However, the principal object is the ‘essential of art’ (das Wesen der Kunst).

Reading this text one sees that it is not exactly true to say, as it often is, that the historicity of art as Winckelmann conceives it emerges “from a compromise through which history would find a field within or at the margins of the norm.” To speak in this way gives too much credit to historical discourse as such. It is to suppose that a history becomes normative only by leaving its own domain, by straining against its “natural” philosophical neutrality, in short, by betraying its “natural” modesty in the face of pure and simple observational facts. This, however, is to fail to see that the norm is internal to the narrative itself, and even to the simplest description or mention of any phenomenon that the historian considers worth retaining. The historical narrative, it goes without saying, is always preceded and conditioned by a theoretical norm concerning the “essence” of its object. Art history, therefore, is conditioned by the aesthetic norm that determines the “right objects” for its narrative, those “beautiful objects” whose conjunction constitutes, in the end, something like an essence of art.

Winckelmann was thus right to claim for his history the status of a “system” (Lehrgebäude) in the philosophical and doctrinal sense of that term. To varying degrees his enterprise resonates with those of a Montesquieu, of a Vico, of a Gibbon or of a Condillac. This status of Winckelmannian history, moreover, was clearly recognized in the 18th century. Herder writes that “Winckelmann has most certainly proposed this grand, true and eternal system (Lehrgebäude)” as the quasi-Platonic undertaking of an “analysis bearing on the general, on the essence of beauty.” As a thinker deeply concerned with historicity, Herder soon raised the following question: “Is this the goal of history? The goal of a history of art? Are not other forms of history possible?” But he readily recognized the need for a history of art that, going beyond the historical collections of Pliny, of Pausanius and of Philostrates, was theoretically established; and this he called, with Winckelmann, an historical system.

He also termed it an “ideal construction.” Ideal in the sense that it was conceived in the first place to harmonize with the metaphysical principle par excellence, namely the ideal of beauty, that “essence of art” that the great artists of antiquity were able to realize in their works. “Ideal beauty,” of course, constitutes the cardinal point of the whole Winckelmannian historical system as of neoclassical aesthetics in general. It provides the essence, and therefore the norm. The history of art is simply the history of its development and its decline. It also appears to confirm aesthetic thought’s long-term allegiance to philosophical idealism.

The word “ideal” suggests that the essence—here, the essence of art—is a model: a model to be attained according to the “categorical imperative” of classical beauty; and, yet, it is presented as a model that is impossible to attain as such. It is very significant that the chapter Winckelmann devotes to the “Essence of art” is really devoted instead to the detours that our mind must take in order to recollect for itself the ideal beauty of Greek statues.

Since the first chapter of this book is only an introduction to the latter, I will pass, after these preliminary observations, to the essence itself of art.... I imagine myself, in fact, appearing in the Olympia Stadium, where I seem to see countless statues of young, manly heroes, and two-horse and four-horse chariots of bronze, with the figures of the victors erect thereon, and other wonders of art. Indeed, my imagination has several times plunged me into such a reverie, in which I have likened myself to those athletes …. I would not, however, wish this imaginary flight to Elis to be regarded as a mere poetic fancy, but as real contemplation of the objects. And this fiction will take on a sort of reality as if I were to conceive as actually existing all the statues and images of which mention has been made by [ancient] authors [Translation modified—Trans.].

This is really very strange. The ideal is apprehended and recognized by means of a “real contemplation of objects,” as Winckelmann phrases it. But not by the contemplation of real objects. The latter have disappeared and have been replaced by later copies. All that remain are the mediations of the mind in search of that point outside of time that is the ideal. And, meanwhile, the most necessary of these mediations—textual reconstruction and ideal restoration—will be called art history. An art history understood as standing in the service of the Idea, and presented as the narrative of avatars, of the moments of greatness and decline with respect to the norm of art: “beautiful nature,” “noble contour,” and “spiritual archetype” in the outlining of female bodies, elegant draperies, and so on. The History of Ancient Art is obviously composed of constant appeals back to the aesthetics presented ten years earlier in the Thoughts on the Imitation of Greek Works.

Here, then, is our inventor of the history of art, plunged into mourning over his object, crying over the death of ancient beauties; here is our aesthete with his “esprit de système,” our historian who does not believe in phantoms, paradoxically constructing the absent objects of his narrative— or, as he believes, of his science—“representing [them to himself] as if they existed,” on the basis of old Latin and Greek descriptions to which he is obliged to give credence. Here he is, finally—he who has assailed us with the “essence of art,” with a principled panegyric of “good taste” (der gute Geschmak) and an absolute rejection of “any deformation of the body”—in an astonishing passage of the Reflections in which he expresses his horror of “venereal diseases and their daughter, the English malady,” those evils that he assumed were unknown to the ancient Greeks. And, as if those things were linked by some obscure common pathology, Winckelmann expresses just as radically his rejection of pathos, that malady of the soul that deforms bodies and thus ruins the ideal, which presupposes the calm that characterizes greatness and nobility of soul:

The more tranquil the state of the body the more capable it is of portraying the true character of the soul. In all positions too removed from this tranquillity, the soul is not in its most essential condition, but in one that is agitated and forced. A soul is more apparent and distinctive when seen in violent passion, but it is great and noble when seen in a state of unity and calm.

What was presented in the Reflections as a general postulate will be applied, in the History of Art, to the specific domain of Greek art. Instead of saying “one must” (point of view of the norm), Winckelmann henceforth is satisfied to write that the Greeks “were accustomed to.” This point of view is “historical,” of course; but it is the same essence that is expressed in it, or, should I say, that shows itself in it:

Expression, in its limited as well as more extended signification, changes the features of the face, and the posture, and consequently alters those forms which constitute beauty. The greater the change, the more unfavorable it is to beauty. On this account, stillness was one of the principles observed here, because it was regarded, according to Plato, as a state intermediate between sadness and gaiety; and, for the same reason, stillness is the state most appropriate to beauty, just as it is to the sea. Experience also teaches that the most beautiful men are quiet in manners and demeanor…. Besides, a state of stillness and repose, both in man and beast, is that state which allows us to examine and discover their real nature and characteristics, just as one sees the bottom of a river or a lake only when their waters are still and unruffled, and consequently even Art can express her own peculiar nature (das Wesen der Kunst) only in stillness.


This introduction is sufficient, I believe, to show the eminently problematic nature of the moment of thought represented by the History of Ancient Art and its overall legacy. It erects a system, but one that constantly fails to reach completion: every affirmation of a thesis or theoretical proposition is quickly followed by a contradiction. Thus, Winckelmann contrasts art history with simple judgments of taste, but an aesthetic norm informs every step of his historical narrative. He claims his history is an objectification of the “debris” of the past, but a powerful subjective element never ceases to guide his scholarly writing: “I travel in my mind to the stadium on Olympus.” The art history that Winckelmann advocates oscillates ceaselessly between essence and becoming. In it the historical past is invented as much as it is discovered.

What should one make of this state of affairs? It has been said since Quatremère de Quincy, and it is still repeated today, that Winckelmann invented art history in the modern sense of the term. Is not this one more contradiction? The sociologist of images, the iconologist, the archaeologist using an electron microscope, the museum curator who is familiar with spectrometric analyses – are they still burdened with such philosophical problems? The status of art history as a “scientific” discipline seems so well established that it is hard to see what we could still owe to such a world of thought. Yet, one is often unaware of the heritage of which one is the legatee. What knot of problems does this History of Ancient Art continue to offer us?

The very title of Winckelmann’s work introduces and imposes a triple knot, a knot tied three times: a knot of history (how can we construct it, how write it?), a knot of art (how can we distinguish it, look at it?), and a knot of antiquity (how are can we recollect it, restore it?). Winckelmann’s “system,” of course is not philosophical in the strict sense and therefore could not be considered anything like a dialectical construction. But there exists a crucial notion, a word that holds together the three bows of the knot. It is a kind of magic word, resolving all the contradictions, or rather causing them to pass unnoticed. It is the word imitation, and it forms the central element of the Winckelmannian system, the hinge, the pivot thanks to which all the differences are linked together and all the abysses can be crossed.

In the conclusion of his work, cited above, Winckelmann seems to have opened a chasm: a depressive chasm, linked to the loss of ancient art and to the impossibility of the return of that “object of wishes”; a chasm separating mourning from desire (Wunsch); a chasm separating the “originals” (Urbilder) of the Greek statues and their Roman “copies” (Kopien). But elsewhere in his work, beginning with the Thoughts, of course, imitation throws a bridge across these chasms. Imitation of the ancients, as practiced by the neoclassical artist, has the capacity to rekindle desire beyond the mourning. It creates a link between the original and the copy that enables the idea, the “essence of art” to revive, as it were, and to traverse time. It is thanks to imitation that the “categorical absence” of Greek art, to use Alex Potts’ expression, becomes capable of a renaissance, and even of an “intense presence.”

For it is truly a matter of presence and of the present: the present time of an imitation “revives the lost original,” thus restoring to the origin an active or current presence. That turns out to be possible only because the object of imitation is not an object, but rather the ideal itself. While the depressive side of Winckelmannian history made Greek art into an object of mourning, impossible to attain—“we have, as it were, nothing left but a shadowy outline of the object of our wishes” —a maniacal side, if I may be so bold as to put it that way, makes this art into an ideal to be grasped, into the categorical imperative of the “essence of art,” something which becomes possible only through the imitation of the ancients. Imitation, of course, is a highly paradoxical concept. But its paradox is precisely what allows Winckelmann to execute his famous pirouette: “The only way for us to become great, or, if this be possible, inimitable, is to imitate the ancients.”

This is a considerable achievement, and its consequences will be considerable, as well. They affect the very framework, the temporal architecture of the entire enterprise; for the art history that Winckelmann constructs winds up overlaying the natural time of Veränderung [alteration] on the ideal time of the Wesen der Kunst. And this is what allows him to have the schema of “life and death” and of “greatness and decline” coexist with the intellectual project of a ”renaissance” or “neoclassical” restoration. We must stress the crucial element of this achievement: imitation allows this renaissance to imitate only the ideal. How can one fail to recognize here, reconfigured but still carried along, the three basic “magic words” of Vasarian idealism. How can one fail to recognize, in the overlaying of natural time on ideal time, the thing that constitutes the ambivalence of the humanist concept of imitation itself? Moreover, would the modern imitation of the inimitable ancients have been possible without the middle term represented, in Winckelmann’s eyes, too, by the Renaissance imitation—by Raphael, in the first place—of these same ancients?

A difficult knot (with a tangled-up solution) now becomes one that is properly tied (with an obvious solution), consisting of three loops. The knot of Antiquity comes undone, forming a notion of the ideal; the knot of art comes undone, forming a notion of imitation; and the knot of history comes undone forming a notion of the Renaissance. This is how Vasari’s humanist history had already been constructed, and this is how Winckelmann’s neo-classical history recommences. Let us, however, raise Herder’s question again: “Is [this] the goal of history? The goal of a history of art? Are not other forms of history possible?”

And let us be clear about what is still at stake here, given the Winckelmannian heritage so unanimously claimed among art historians. Consider, in the first place, the “analysis of time.” Might there not exist a time of images which is neither “life and death” nor “greatness and decline,” nor even that ideal “Renaissance” whose values historians constantly put to their own uses? Might there not be a time for phantoms, a return of the images, a “survival” (Nachleben) that is not subject to the model of transmission presupposed by the “imitation” (Nachahmung) of ancient works by more recent works? Might there not be a time for the memory of images—an obscure game of the repressed and its eternal return – that is not the one proposed by this history of art, by this narrative? And, as for art itself: might there not be a “body” of images that escapes the classifications established in the 18th century? Might there not be a type of resemblance that is not the one imposed by the “imitation of the ideal,” with its rejection, in Winckelmann’s formulation, of pathos? Might there not be a time for symptoms in the history of the images of art? Was this history truly “born” one day?

<1>Warburg, Our Phantom

A century and a half after Winckelmann wrote his monumental History of Ancient Art, Aby Warburg published, not in Dresden but in Hamburg, a miniscule text—actually, a five-and-half-page summary of a lecture—on “Dürer and Italian Antiquity.” The image that opens this text is not that of a Christian resurrection, as in Vasari (Fig. 1), or of some Olympian glory, as in Winckelmann (Fig. 2). Rather, it is one in which a human being is torn apart, a passionate and violent scene, frozen at a moment of extreme physical intensity (Fig. 3).

<insert fig. 3 about here>

The dissymmetry between these moments of thinking about history, about art and about antiquity, appears to be very radical. In his brief text, which is shorter than a single “Life” in Vasari, as is the case with all his published work—just as all his published work is less voluminous than the History of Ancient Art by itself—Warburg surreptitiously decomposes and deconstructs all the epistemic models employed in Vasarian and Winckelmannian history of art. He thus deconstructs what the history of art still today considers to be its initiatory moment.

For the natural model of cycles of “life and death” and “greatness and decline,” Warburg substituted a resolutely non-natural, symbolic model, a cultural model of history in which temporal periods are no longer fashioned according to biomorphic stages, but, instead, are expressed by strata, hybrid blocks, rhizomes, specific complexities, by returns that are often unexpected and goals that are always thwarted. For the ideal model of “renaissances,” “good imitations,” and the “serene beauties” of antiquity, Warburg substituted what might be termed a phantasmal model [modèle phantomal] of history, in which temporal periods are no longer fashioned according to the academic transmission of knowledge, but are expressed, rather, by hauntings, “survivals,” residues, and the persistent return of forms—that is to say, by notions that do not constitute knowledge, that are unthought, and by unconscious aspects of time. In the final analysis, the phantasmal model that I am speaking of is a psychicological model [modèle psychique], in the sense that the psychicological point of view would not constitute a return to the point of view of the ideal but to the possibility of its theoretical decomposition. What we have here, then, could be called a symptomatic model [modèle symptomal], in which the emergence and change of forms is to be analyzed as an ensemble of processes characterized by tensions —or example, tensions between the desire for identification and the constraint of alteration, between purification and hybridization, the normal and the pathological, order and chaos, and between characteristics that can be seen and others that remain unthought.

Admittedly, I have asserted all this in a very abrupt and condensed manner. We need to start over again from the beginning in order to construct this working hypothesis. But it was necessary to state the following right away: with Warburg, thinking about art and thinking about history took a decisive turn. And after him we no longer confront or stand before the image [devant l’image] or before time [devant le temps] in the same way as we did previously. [L’Image Survivante was preceded by two other books by Didi-Huberman in which he explored themes in the methodology and epistemology of art and its history using these same two phrases as their respective titles: Devant l’image. Question posée aux fins d’une histoire de l’art, 1990 (translated as Didi-Huberman, G. 2005) and Devant le temps. Histoire de l’art et anachronisme des images, 2000.—Trans.]. All the same, art history does not “commence” with him in the sense of a systematic re-foundation, as one might well have expected. Beginning with him, though, art history becomes relentlessly worried about itself; the history of art becomes unsettled and confused, which is a way of saying—if one recalls the lesson taught by Walter Benjamin—that it comes close to reaching an origin. Warburg’s art history is very much the opposite of an absolute beginning, of a blank slate. It is more like a vortex in the river of the discipline—a troublemaking moment beyond which the course of things is inflected, indeed, profoundly changed.

Just how profoundly changed is not easy to determine, even today. Elsewhere I have attempted to characterize certain tensions, both in the history of the discipline and in its current state, that have impeded the recognition of a change of this magnitude. Let me add to that the following persistent impression: Warburg is our obsession [hantise]; he haunts us. He is to art history what an unappeased ghost—a dybbuk—might be to the house we live in. What is such an obsession? It is something or someone that always comes back, survives everything, reappears at intervals, and expresses a truth concerning an original state of affairs. It is something or someone that one cannot forget, and yet is impossible to recognize clearly.


Warburg, our phantom: located somewhere in us, but beyond our grasp, unknown. Upon his death, in 1929, the obituaries devoted to him—penned by scholars as distinguished as Erwin Panofsky and Ernst Cassirer—displayed the great respect due to ancestors who really matter. He was recognized as the founding father of a substantial discipline, iconology; but his work was soon eclipsed by that of Panofsky, so much clearer and more distinct, so much more systematic and reassuring. Since that time, Warburg has been wandering through the history of art like an unmentionable ancestor, a ghostly father of iconology—without anyone ever saying exactly what should not be mentioned, nor what should be disavowed in his work.

Why ghostly? First of all because one does not know where to get hold of him. In his obituary of Warburg, Giorgio Pasquali wrote, in 1930, that the historian, during his lifetime, “already disappeared behind the institution he had created” in Hamburg, the famous Kulturwissenschaftliche Bibliothek Warburg, which, following the exile precipitated by the Nazi menace, was able to survive and revive in London. In order to tell who Warburg was, and what Warburg was, Ernst Gombrich—to whom the task fell of writing a book that was first conceived by Gertrud Bing—resolved to produce an “intellectual biography” in which he would deliberately self-censure certain psychological aspects of Warburg’s life and personality. This decision was accompanied by a somewhat disembodied “elaboration” of an oeuvre in which the dimension of pathos, indeed of the pathological, proved to be essential, as much with respect to the objects studied as to the view that was brought to bear on them. Edgar Wind severely criticized this prudish reassembly, this watered-down version. One should not, he thought, separate a man from his pathos—his empathies and his pathologies—one should not separate Nietzsche from his madness nor Warburg from those losses of self that led to his confinement behind the walls of a psychiatric asylum for almost five years. The symmetrical danger exits, of course: that of neglecting the work in favor of a cheap fascination with a destiny worthy of a Gothic novel.

Another source of this ghostly character is the impossibility, even today, of discerning the exact limits of Warburg’s oeuvre. Like a spectral body, it remains without definable contours: it has not yet found its corpus. It haunts every book in his library—and even every interval between the books, on account of the famous “law of the good neighbor” that Warburg instituted for their classification. But, above all, it is spread out in the vast maze of still unpublished manuscripts—all those notes, sketches, schemas, and journals, along with the correspondence that Warburg tirelessly kept up, never throwing anything away, and that the editors have so far been unable to bring together in a methodical fashion, so perplexing is the “kaleidoscopic fashion” of all this material. Certain of these writings, moreover, were explicitly envisaged as proposing fundamental principles, namely the Grundlegende Bruchstücke zu einer monistischen Kunstpsychologie (1888-1905) and the Allgemeine Ideen (1927). Thus, given our ignorance with respect to such a mass of texts, all our reflections about Warburg remain subject to a certain indecisiveness. To write about this oeuvre today, we must accept that our own working hypotheses may one day be modified or brought into question by an unanticipated piece of this floating corpus.

But that is not all. There is a third, still more fundamental cause of the ghostly aspect of this thought, namely style—and, therefore, time. In reading Warburg we confront the difficulty of seeing the tempo of the most exhausting, or the most unexpected erudition—such as the sudden appearance, in the middle of an analysis of the Renaissance frescoes of the Schifanoia Palace in Ferrara, of an Arab astrologer of the 9th century, Abu Ma’sar — combined with the almost Baudelairian tempo of rockets: thoughts that simply burst out, uncertain thoughts, aphorisms, permutations of words, experiments with various concepts. All this, Gombrich assumes, is liable to put off the “modern reader,” but this is precisely what signals Warburg’s modernity.


From what position, then, from what place and from what time does this phantom speak to us? His vocabulary is drawn alternately from German romanticism and from Carlyle, from positivism and from Nietzsche’s philosophy. He alternately displays a meticulous concern for historical detail and the uncertain inspiration of prophetic intuition. Warburg himself described his style as being like an “eel broth” (Aalsuppenstil). Taking our cue from this remark, let us imagine a mass of serpentine, reptilian bodies, somewhere between the dangerous circumvolutions of the Laocoon—which obsessed Warburg for his entire life, no less than the snakes that the Indians he studied inserted into their mouths (Fig. 37)—and the unformed mass, without head or tail, of a way of thinking that always stubbornly resists “cutting itself,” that is to say, defining for itself a beginning and an end.

Let us observe, in addition, that Warburg’s vocabulary itself seems destined to assume the status of a specter. Gombrich notes that the most important words of this vocabulary—such as bewegtes Leben, Pathosformel, and Nachleben—have difficulty making their way into English. It would be more to the point to say that postwar Anglo-Saxon art history, which owed such a large debt to German émigrés, deliberately gave up the use of German philosophical language. The unappeased ghost of a certain philological and philosophical tradition, Warburg thus wanders through a twofold and elusive time. On the one hand, he speaks to us from a past that the “progress of the discipline” seems to have rendered outmoded. It is characteristic, in particular, that the term Nachleben—“survival,” a concept which is crucial to the whole Warburgian enterprise— fell into complete neglect and, if by chance it has been cited at all, has not been the subject of any serious epistemological critique.

On the other hand, Warburg’s oeuvre can be read as a prophetic text and, more precisely, as the prophecy of a knowledge that will come to us in the future. In 1964, Robert Klein wrote concerning Warburg: “[He] created a discipline that, contrary to so many others, exists but has no name.” Taking up this formulation, Giorgio Agamben has shown the degree to which the “science” envisaged by such an oeuvre has “not yet been established”—indicating thereby not so much a lack of rationality as the considerable ambition and disruptive nature of this way of thinking about images. Warburg said of himself that he was created less to exist than to “remain [I would say: persist] as a beautiful memory.” Such is indeed the sense of the word Nachleben, this word that signifies “afterlife” or “living afterwards”; a being from the past never ceases to survive. At a certain moment, its return into our memory becomes urgent, possessing the anachronistic urgency of what Nietzsche termed the untimely.

Such is Warburg today: a survivor whose presence is urgent for art history. He is our dybbuk, the ghost of our discipline, speaking to us simultaneously of his (of our) past and of his (of our future). With respect to the past, we should rejoice in the philological work that, especially in Germany, has been devoted for some years now to Warburg’s oeuvre. With respect to the future, things are obviously rather trickier: the value of Warburg’s efforts as a “stimulus” having now been recognized, interpretations of his work are beginning to diverge. Not only was the heritage of the “Warburgian method” questioned right from the moment scholars began to employ it; the current profusion of references to this supposed “method” can truly make one dizzy. Warburg, one might say, is redoubling his ghostliness at the very moment when everyone is beginning to invoke him as the guardian angel of the most diverse theoretical approaches. He is guardian angel of the history of mentalités, of the social history of art, and of micro-history, guardian angel of hermeneutics, guardian angel of a so-called anti-formalism, guardian angel of a so-called “retro-modern post-modernism,” guardian angel of the New Art History, and even a major ally of feminist critique etc., etc.

<1>Forms Survive: History Opens Up

What remains certain is that, as Ernst Gombrich wrote—but how could he not feel himself targeted by his own statement?—“the [current] fascination exerted by Warburg’s legacy may also be viewed as a symptom of a certain dissatisfaction” with art history as it has been practiced since the end of the Second World War.

In his time, Warburg himself had displayed this kind of dissatisfaction, which is a way of expressing a demand that has not yet been fully formulated. In 1888, when he was only twenty-two years old, Warburg, in his private journal, was already castigating art history for “cultivated people,” the aestheticizing” history of art of those who are content to evaluate figurative works of art in terms of beauty. He was already calling for a Kunstwissenschaft, a “science of art,” writing that there would come a day when, without it, it would be as futile to talk about images as for a non-physician to comment on a symptomatology.

And, in 1923, Warburg still recalled that it was on account of his “downright disgust with aetheticizing art history” (ästhetisierende Kunstgeschichte) that he suddenly left for the mountains of New Mexico. Throughout his life he insisted that the serious study of images required a much more radical questioning than all that “curiosité gourmande” of the attributionists—such as Morelli, Venturi, and Berenson—whom he termed “professional admirers.” He likewise demanded much more than the vague aestheticism of the disciples of Ruskin and Walter Pater (when they were of the vulgar kind, that is to say “bourgeois”), and of those of Burckhardt and Nietzsche, too. Thus, in his notebooks he sarcastically evokes the “tourist-superman on Easter vacation” who comes to visit Florence “with Zarathustra in the pocket of his loden coat.”

Responding to this dissatisfaction, Warburg evinced a constant displacement: a displacement in thought, in philosophical points of view, in fields of knowledge, in historical periods, in cultural hierarchies, and in geographical locations. And this very displacement continues to make him phantom-like: Warburg was, in his time—though never more so than today—the will-o’-the-wisp, the passe-muraille of art history. His displacement towards art history, towards scholarship and images in general, had already been the result of a critical reaction to his family’s world: a malaise with respect to upper-middle-class business circles and to Orthodox Judaism. But, above all, his displacement through the history of art, to its borders and beyond, created in the discipline itself a violent critical reaction, a crisis, and a real deconstruction of disciplinary frontiers.

This reaction is already evident in the choices the young Warburg made as a student between 1886 and 1888. He studied with classical archaeologists—classical in all senses of the term—such as Reinhard Kekulé von Stradonitz and Adolf Michaelis. With the latter he studied the Parthenon friezes; in the former’s course he discovered the aesthetics of the Laocoon and, in 1887, produced his very first analysis of a Pathosformel. He became a disciple of Carl Justi, who initiated him into classical philology and introduced him to Winckelmann, but also to Velazquez and to Flemish painting. On the other hand, he developed an enthusiasm for the “anthropological” philology of Hermann Usener, with all the philosophical, ethnographic, psychological, and historical problems that it raised in its wake. Then, in Karl Lamprecht’s lectures on history conceived as a “psychological science” he encountered several of the basic elements of his future methodology.

With regard to the Renaissance, the teaching of Riehl and of Thode served mainly as foils. (The latter had made Italian artistic development a result of the Franciscan spirit, pushing the return of pagan antiquity into the background.) But Hubert Janitschek led him to understand the importance of theories of art— those of Dante and of Alberti—as well as the role played by the social practices that are linked to all forms of the production of figurative art. As for August Schmarsow, he quite simply initiated Warburg into the Florentine terrain, if I may put it that way: it was in Florence itself that the young historian took the former’s courses on Donatello, Botticelli, and the relationship between Gothic and Renaissance in Quattrocento Florence, all of them subjects we recognize today as eminently Warburgian.

Furthermore, Schmarsow advocated a Kunstwissenschaft resolutely open to anthropological and psychological questions. He elaborated a specific concept of visual communication and “information” (Verständigung), but, above all, he understood the fundamental role of what, in that period, was called the “language of gestures.” Taking up again the topic of the expressiveness of the Laocoon, and going beyond Lessing, Schmarsow sought to elaborate a theory of the corporeal empathy of images, utilizing a binomial schema of “gesture” (Mimik) and “modelling” (Plastik). It is not so astonishing, therefore, to see the young Warburg going from the ancient Psychomachias to the reading of Wilhelm Wundt, or from Botticelli to medical courses, or even to a course on probabilities, where, in 1891, he gave a presentation on “The logical foundations of games of chance.”

More than a domain of knowledge in formation, it was really a domain of knowledge in motion that, little by little, was emerging through the seemingly erratic play of the all these methodological displacements. Born in 1866, Warburg was a member of a generation of distinguished art historians (Emile Mâle was born in 1862, Adolph Goldschmidt in 1863, Heinrich Wölfflin in 1864, Bernard Berenson in 1865, Julius von Schloesser in 1866, Max J. Friedländer in 1867, and Wilhelm Vöge in 1868), but his epistemic position and his institutional situation totally distinguished him from these others. In 1904, nearing his fortieth birthday, he once again failed to pass his habilitation examination for a post as professor in Bonn. He had already written as early as 1897, half lucidly and half anxiously: “I have decided once and for all that I am not suited to be Privatdozent.” He was to decline, subsequently, offers of chairs in Breslau, in Halle, and in general all public positions, refusing, for example, to represent the German delegation at the International Congress in Rome (1912) even though he had been one of its most active promoters. He was to remain a “private researcher”—and we should understand the word in all its possible senses—whose very project, the “science without a name” was unable to find a satisfactory home in the various existing disciplinary enclosures and other academic arrangements.

This, then, was his initial dissatisfaction: the territorialization of the study of images. In 1912, concluding his address to the Rome Congress on the astrological motifs in the frescoes of Francesco del Cossa at Ferrara, Warburg pleaded, to use his own words, for an “enlargement” of the discipline:

The isolated and highly provisional experiment that I have undertaken here is intended as a plea for an enlargement of the methodological borders of our study of art (eine methodische Grenzerweiterung unserer Kunstwissenschaft) [Translation modified—Trans.].

It would be correct, but very incomplete, to understand this plea as a call for “interdisciplinarity,” or as the philosophically-motivated extension of a point of view about images to areas beyond the factual and stylistic problems that the traditional historian or art chooses to consider. It is certain that Warburg’s desire was always to reconcile a philological concern (and thus the prudence and competence that it presupposes) and a philosophical concern (and thus the risk and even impertinence that it presupposes). But there is still more: Warburg’s demands respecting art history stem from a very precise position concerning each of the two terms “art” and “history.”

Warburg, I believe, felt dissatisfied with the territorialization of the study of images because he was sure of two things at least. First, we do not stand confronted with or before an image the way we do before a thing whose exact boundaries we can trace. The ensemble of definite coordinates—author, date, technique, iconography, etc.—is obviously insufficient for that. An image, every image, is the result of movements that are provisionally sedimented or crystallized in it. These movements traverse it through and through, each one having its own trajectory—historical, anthropological, and psychological—starting from a distance and continuing beyond it. They oblige us to think of the image as an energy-bearing or dynamic moment, even though it may have a specific structure.

Now, that entails a basic consequence for art history, which Warburg announces in the words immediately following his “plea”: we stand before the image as we do before a complex time, namely the provisionally configured, dynamic time of these movements themselves. The consequence, indeed the stakes in play, of a “methodical enlargement of the frontiers” is no less than a deterritorialization of the image and of the time in which its historicity finds expression. This clearly means that the time of the image is not the time of history in general, i.e. time that Warburg defines here in terms of the “universal categories” of evolution. What, then, is the urgent task he envisages (the one that is untimely, not current)? The history of art needs to reestablish an “evolutionary theory of its own,” its own theory of time, which, one immediately notes, Warburg orients towards an “historical psychology”:

Until now, a lack of adequate general evolutionary categories has impeded art history in placing its materials at the disposal of the—still unwritten—“historical psychology of human expression” (historische Psychologie des menschlichen Ausdrucks). By adopting either an unduly materialistic or an unduly mystical stance, our young discipline ... gropes toward an evolutionary theory of its own (ihre eigene Entwicklungslehre), somewhere between the schematisms of political history and the dogmatic faith in genius.

For now, we must seek to follow Warburg in his attempt to “pass through walls”: to “decompartmentalize” the image and the time that it bears within itself or that bears it. To follow the organic movement involved here, without omitting anything, would be an overwhelming task. One can at least begin to undertake an epistemological critique of this scope by considering the way or ways in which Warburg goes about initiating movement in and displacing art history. Once again, we observe that everything involved in this enterprise is a matter of style —whether style of thinking, of making a decision, or of coming to know something; which is to say that is a matter of time, of tempo.


One way to displace things is to take one’s time, to postpone (différer). In Florence, Warburg is already “postponing” the history of art: he makes it take on another time than the Vasarian time of the self-glorifying “histories,” or the Hegelian time of the “universal meaning of history.” He creates a novel type of relationship between the particular and the universal. In order to do that, he traverses and overturns the traditional domains of art itself. As the Uffizi galleries are no longer enough for him, he decides to immerse himself in the un-hierarchical world of the archives, of the Archvio, with its innumerable private ricordanze, its account books, its notarized wills, and such like. Thus, the notice of a payment for a votive image made in 1481 and based on the donor’s own countenance, or the last wishes of a Florentine bourgeois become, in his eyes, elements of a body of material, both moving and unlimited, suitable for reinventing the history of the Renaissance. This is a history that could already be called “phantasmal,” in the sense that in it the archive is treated as a material vestige of the sounds of the dead. Warburg writes that his aim in using the “archival documents that have been read” is “to restore the tone and timbre of those unheard voices” (den unhörbaren Stimmen wieder Klangfarbe zu verleihen)—voices of the deceased, yet voices that still lie waiting, coiled up, as it were, simply in the writing itself or in the particular turns of phrase of an intimate Quattrocento journal exhumed in the Archivio.

Looked at from this point of view, which might be called that of ghostly return, the images themselves are considered to be what survives of a dynamic process of anthropological sedimentation that has become partial, or virtual, having been largely destroyed by time. Thus, as a first approximation, the image—starting with those portraits of Florentine bankers that Warburg examined with a particular fervor—is viewed as what survives of a population of ghosts. Ghosts whose traces are scarcely visible and yet are disseminated everywhere: in an astrological theme concerning birth, in a business letter, in a garland (guirlande) of flowers (the very one from which Ghirlandaio took his name), in a detail relating to the fashions of the time, a belt buckle, say, or the particular curl of a woman’s chignon.

This anthropological dissemination obviously calls for a multiplication of points of view, approaches, and competencies. The impressive Kulturwissenschaftliche Bibliothek Warburg, in Hamburg, was destined to assume the burden of such an epistemological displacement, a burden demanding infinite patience, and one which was constantly enlarged and altered. Conceived by Warburg as early as 1889 and built between 1900 and 1906, the library constituted a kind of magnum opus in which its author, although assisted by Fritz Saxl, probably got lost as he went about constructing his “thinking space” (Denkraum) in it. In this rhizome-like space, which by 1929 contained 65,000 volumes, art history as an academic discipline underwent an ordeal of regulated disorientation: everywhere that there existed frontiers between disciplines, the library sought to establish links.

But this space was still the working library of a “science without a name”: a library, thus, for work but also a library that was a work in progress. Fritz Saxl put it very well when he said that the library was, above all, a space of questions, a place for documenting problems, a complex network at the summit of which—and this is extremely significant for our purposes—stood the question of time and of history. “It is a library of questions, and its specific character consists precisely in the fact that its classification obliges one to enter into its problems. At the library’s summit (an der Spitze) is located the section on the philosophy of history.”

Salvatore Settis, in a remarkable article, has reconstituted the library’s practical models—beginning with library of the University of Strasbourg, where Warburg was a student—as well as the theoretical context provided by the debates over the classification of knowledge at the end of the 19th century. Above all, he has traced the many stages in Warburg’s incessant ruminating on the trajectories and “places” in the library, showing them to be a function of the way in which he grappled with the fundamental problems signaled by such crucial terms as Nachleben der Antike (“survival of antiquity”), Ausdruck (“expression”), and Mnemosyne.

This helps us understand how a library conceived in this manner was able to produce its displacement effects. A heuristic attitude—that is to say, a thought experiment that doesn’t assume one knows ahead of time the axiom on which the answer depends—guided the ceaseless efforts involved in its reconfiguration. How was one to go about organizing interdisciplinarity? That presupposed, once again, the difficult encounter of philological cogwheels and philosophical grains of sand. It also presupposed creating a real archaeology of the fields of knowledge that are linked to what today are called the “human sciences”—a theoretical archaeology centered, from the start, on the twofold question of forms and symbols.

At the same time, however, an enterprise of this kind generated what might be called an aporetic situation. In the beginning, it involved one person and one universe of questions. And, as one can still verify today, wandering among the bookshelves at the Warburg Institute in London, one has a very strange feeling using a working tool that clearly bears the mark of its builder. If Warburg’s library has resisted the effects of time so well, it is because the phantoms of the questions he raised have found neither a stable home nor any rest. In his funeral oration for the historian, Ernst Cassirer wrote a magnificent page on the auratic character of a library at once so private and so open, “inhabited” by those “original spiritual configurations,” as he put it, from which there seems to have emerged, specter-like and still “without a name,” a possible archaeology of culture. It is undeniable, however, that this kind of strangeness carries with it something like the stigmata of an aporia: Warburg multiplied the links between the fields of knowledge, that is to say, between the possible responses to the insane overdetermination of images. And with respect to this multiplication, he probably dreamed of not choosing, of postponing, of cutting nothing out, of taking the time to take everything into account—surely a kind of insanity. How does one orient oneself in the midst of this knot of problems? How does one orient oneself in the “eel soup” constituted by the problem of the determinism of images?

There is another way of posing the question, of displacing things. Another style, another tempo. Namely, to lose—or rather seem to lose—one’s time. It is to proceed along the edges of an issue, to act by impulse. It is to bifurcate, to branch off all of a sudden, to no longer put anything off. It is to directly confront the differences involved in the matter. It is to start out, as it were, at ground level. Not that the Archivio or the library are pure abstractions, floating above the terrain. To the contrary, these treasure houses of knowledge and civilization bring together a great number of different strata, and one can, in fact, follow their movements in the terrain, from one archive to another, from one field of knowledge to another. But to bifurcate is something else: it means moving towards the terrain, traversing the ground, and accepting the existential ordeal provoked by the questions one raises.

In fact, it requires one to undergo a displacement in one’s point of view, more specifically, the displacement of one’s position as subject in order to give oneself the means of displacing the definition of one’s object. Warburg offered reasons for his trip to New Mexico that he himself labeled “romantic” (der Wille zum Romantischen), above all a powerful reaction to the inanity of the modern civilization (die Leerheit der Zivilisation) that he observed on the East coast of the United States during a trip with his family. But he also invoked properly “scientific” reasons (zur Wissenschaft) connected with his “downright disgust with aestheticizing art history” and with his quest for a “science of art” (Kunstwissenschaft) open to the symbolic field—or, as he put it at that time, to the cultural field in general (Kulturwissenschaft).

Although Warburg’s “Indian trip” has often been studied, the question of what exactly he was looking for—and of what he found there—has to some extent remained unanswered. There is agreement on the methodological importance of such a displacement, setting aside the readings of those who, bewildered or even shocked by it, interpret it as the purely negative and inappropriate [déplacé] act of an art historian in the midst of a psychological crisis. Still, one must ask what type of object Warburg encountered during this trip: what type of object suitable for displacing the “art” object contained in the very expression “art history.” Let us ask, symmetrically, what type of time Warburg experienced there that was suitable for displacing “history,” as that term is generally understood in the expression “art history.”

What sort of object, then, did Warburg encounter in the course of this experience? Something that, in 1895, probably still remained unnamed. Something that was an image but also an act (i.e. corporeal and social) and a symbol (i.e. psychological and cultural). A theoretical “eel soup,” in short. A pile of snakes—the very thing that actually was swarming in the Oraibi ritual, and the very thing that shot forth symbolically its celestial lightning strokes (Figs. 37 and 73 to 76); and, likewise, that which appeared as an image in the vision of the reptilian-like stalactites of New Mexico and the torsades of a Baroque retable before which Indians were observed praying at Acoma.

The problem raised by this “concretion” of acts, images and symbols is not really one of knowing whether Warburg was looking to establish a parity or disparity between them and the Western European objects from the Florentine Renaissance he was working on. Was he there to establish an analogy with the Renaissance, with its festivals, representations of Apollo and of the serpent Python, and its Dionysian and pagan elements, as Peter Burke thinks? Or was he perhaps there to carry out a complete reversal of the Western, classical point of view, as Sigrid Weigel contends? The answer must be dialectical. It is in the “visible incorporation of strangeness,” to use Alessandro Dal Lago’s expression, that Warburg no doubt looked for the foundation of the polarities that, according to him, are manifested in every cultural phenomenon; but for him this foundation should be understood not as communal and archetypal but rather as differential and comparative.

What, then, made this object suited for displacing the “art” object that the discipline of art history traditionally studied? Precisely the fact that it was not an object, but rather a complex of relations—indeed, a pile, a conglomeration, or a rhizome of relations. This is undoubtedly the main reason for the passionate engagement with anthropological questions that Warburg displayed throughout his life. Anchoring the images and works of art in the field of anthropological questions was a first step in displacing art history, but also a way of leading it to confront its own “fundamental problems.” As an historian, Warburg, like Burckhardt before him, refused to pose these problems at the most basic levels of knowledge, as a Kant or a Hegel would have done. For him, to pose “fundamental problems” was not a matter of seeking to derive the general law or the essence of a human faculty (the capacity to produce images) or of a domain of knowledge (the history of the visual arts). It was, rather, a matter of multiplying the pertinent singularities. In short, it meant to enlarge the field of admissible phenomena in a discipline whose attention until then had been riveted on its objects—to the detriment of the relationships that these objects establish, and by which they are established—like a fetishist on his shoes.


Anthropology, therefore, displaces and defamiliarizes—one might almost say, disquiets—art history. Not in order to disperse it into some eclectic interdisciplinarity without a point of view, but in order to open it up to its own “fundamental problems,” which, in large measure, remain unexamined within the discipline. It is a matter of doing justice to the extreme complexity of the relationships and determinations, or, better, overdeterminations, of which images are constituted. But it is also a matter of offering a new formulation of the specificity of these relationships and of the formal work of which the images themselves are constitutive elements. It is completely foolish, though often enough done, to see Warburg as someone whose sole concern was the discovery of historical “facts” and iconographic “contents,” a so-called anti-formalist incapable of distinguishing between a mass-produced image and a unique masterpiece. What he attempted, rather, as his final project, the Mnemosyne Atlas, clearly attests, was to reformulate the problem of style, that problem of linkages and formal efficacy, by always joining the philological study of the singular case with the anthropological approach to the relationships that render these singularities historically and culturally viable.

It would require a whole book to determine precisely what Warburg found in the anthropology of his time that was capable of transforming his attitude as an art historian; for this involves a vast field encompassing specialized ethnographic studies and grand, philosophically-inspired systems. It would require, in particular, reconstructing the substantial impact on him of the thought of Hermann Usener, whose courses Warburg attended in Bonn between in 1886 and 1887, and whose aim of establishing a “morphology of religious ideas” profoundly marked Warburg’s methodology. He had approached ancient myths in the same spirit as Warburg was soon to do in his study of Renaissance frescoes, linking philological inquiry—with its emphasis on details, specifics, and singularities—to the most fundamental problems of psychology and anthropology. For example, in studying the forms of Greek metrics, Usener conceived of the latter as a symptom of overall culture, seeking for survivals up through the period of medieval music; and, reciprocally, he approached acts of belief generally as forms that, in every specific case, had to be addressed with the tools of the philologist.

One could also look for what Warburg borrowed from the overly-general anthropology of images that Wilhelm Wundt attempted in his gigantic Völkerpsychologie. Or one could trace Warburg’s references to Lucien Lévy-Bruhl, for example with respect to the “law of participation,” the “survival of the dead” and the notion of causality in “primitive mentality.” But it is the important to bring to light not only what Warburg owes to the anthropology of his day; one must also ask, reciprocally, what anthropology in general and historical anthropology in particular owe to an approach of this type.

For various reasons—mainly historical reasons, of course, linked to the long years of the two world wars—French scholarship displayed a particular ignorance regarding this German tradition. Hermann Usener, whom Mauss, however, read very closely, remains unknown to Jean-Pierre Vernant and Marcel Detienne. As for Warburg, he has been ignored not only by positivist art historians, but also by historians sympathetic to structuralism, even by the best scholars of the Annales school. Thus, Jacques Le Goff generously accords Marc Bloch sole credit for the “creation of historical anthropology”; observing, moreover, that the latter’s Rois thaumaturges includes only about ten pages—and not very analytical ones, at that—of “iconographic material” (“dossier inconographique”), he concludes that the “renovation of art history is one of the priorities of historical research today.”

Rereading Warburg today requires that one invert one’s customary perspective. His way of practicing art history, of opening it up, which is so particular and so radical, has had the effect, it seems to me, of raising anew the questions of historical anthropology—a discipline he conceived of in the form he inherited from Jacob Burckhardt and Hermann Usener—on the basis of an inquiry into the symbolic efficacy of images. It is not art history that has to “renew” itself on the basis of “new” questions raised by the overall discipline of history in the domain of the imaginaire; it is the discipline of history itself that has to recognize that a given moment in its own history the “guiding” notions, the “novelty” comes from thinking in specific ways about the powers inherent in the image.

For Warburg, in fact, the image constituted a “total anthropological phenomenon,” a particularly significant crystallization or condensation of what a “culture” (Kultur) was at a given moment in its history. This is what one must understand, first of all, by the idea, dear to Warburg, of the “mythopoetic power of the image” (die mythenbildende Kraft im Bild). And that is why, in his work on the “emotive formulas” of the Renaissance—the Pathosformeln, those gestures that are intensified in representation through the artists’s recourse to the visual formulas of Classical antiquity—he felt that there was no “disciplinary” contradiction in orienting his research towards such topics as social mimicry, choreography, fashions in clothing, behavior during festivals, and codes governing the way people greet each other.

In short, the image should not be dissociated from the overall actions and way of acting (agir) of the members of a society; nor from the knowledge and ways of thinking [savoir] of an epoch; nor, or course, from beliefs and ways of believing [croire]. Here resides another essential element of Warburgian invention, which was that of opening up art history to the “dark continent” of the magical efficacy of images—but also to their liturgical, juridical, and political efficacy:

It is one of the prime duties of art history (Kunstgeschichte) to bring such forms out of the twilight of ideological polemic and to subject them to close historical scrutiny. For there is one crucial issue in the history of style and civilization (eine der Hauptfragen der stilerforschenden Kulturwissenschaft)—the influence of antiquity on the culture of Renaissance Europe as a whole—that cannot otherwise be fully understood and resolved.

The slippage in the vocabulary is significant: we move from art history (Kunstgeschichte) to a science of culture (Kulturwissenschaft), and this move simultaneously opens up the field of objects to be studied and sharpens the formulation of the fundamental problems. For example, Kunstgeschichte tells us that a genre of fine arts called the “portrait” emerged in the Renaissance thanks to the humanistic triumph of the individual and to progress in mimetic techniques; but Warburg’s Kulturwissenschaft tells another story, involving the much more complex time of the intersection—an interlacing, an overdetermination—of ancient and pagan magic (survivals of the Roman imago) and of medieval and Christian liturgy (the practice of ex voto in the form of effigies), as well as of the specific circumstances of artistic and intellectual activity in the Quattrocento. As a result, the portrait is transfigured before our eyes, becoming the anthropological support of a “mythopoietic power” that the Vasarian version of art history had shown itself to be incapable of explaining.

The Kunstwissenschaft, the “science of art” that Warburg so ardently wished for during his youth, thus took the form of a specific investigation of images within the framework of a non-specific, endlessly open Kulturwissenschaft. It was necessary to open up the field of objects capable of interesting the art historian, inasmuch as the work of art was no longer envisaged as an object fully enclosed its own history, but rather as the dynamic point of encounter—Walter Benjamin will later call it the lightning flash—of heterogeneous and over-determined historical factors. In a magisterial article on Warburg’s concept of Kulturwissenschaft, Edgar Wind wrote that “any attempt to detach [even the artistic] image from its ties to religion and poetry, to cult and drama, amounts to cutting it off from its own lifeblood.” Countering any notion of an autonomous history of images—which does not mean that their specific formal qualities must be ignored—Warburg’s Kulturwissenschaft, therefore, ultimately opens up the time in which this history occurs. By having the Greek word for memory (Mnemosyne) engraved in capital letters above the entry door to his library, Warburg indicated to the visitor that he was entering into the territory of another time.

<1>Nachleben, or the Anthropology of Time: Warburg with Tylor

The name of this other time is “survival” (Nachleben). We know the key expression, the mysterious watchword of Warburg’s entire enterprise: Nachleben der Antike. It is the fundamental problem, the one for which he gathered all that material in archives and libraries, seeking to understand the sedimentations and shifts that occurred in the many different terrains involved. It is also the fundamental problem that Warburg tried to confront, in the brief time he had there, on the terrain itself of his American Indian experience. Thus, before examining the notion of survival in the context of the “science of culture” that Warburg patiently elaborated on the basis of images of antiquity and of the modern Western world, it seems worthwhile to look at the emergence of this problematic in its experimental stage on the specific, “displaced” terrain of his travels in Hopi country. The theoretical and heuristic function of anthropology—its capacity to de-territorialize the fields of knowledge and to reintroduce difference into objects and anachronism into history—will thereby appear all the more clearly.

The “survival” that Warburg invoked and investigated throughout his life was, originally, a concept of Anglo-Saxon anthropology. When, in 1911, Julius von Schlosser, who was a close friend of Warburg’s and who in many respects shared an interest in his problematic, referred to the “survival” of figurative practices associated with wax, he did not employ the vocabulary that would naturally come to him from his own native language. He did not write Nachleben, or Fortleben or Überleben, but, rather, survival, in English, as Warburg also did on several occasions. This is a significant indication of a citation, of a borrowing, indeed of a conceptual displacement: what is cited by Schlosser, and what before him Warburg had already borrowed, or displaced, is nothing other than the survival of the great British anthropologist Edward B. Tylor. When Warburg suddenly left Europe for New Mexico, in 1895, he was not undertaking “a journey to the archetypes,” as Fritz Saxl believed, but rather “a trip to the survivals”; and his theoretical reference point was not James G. Frazer, as Saxl wrote, but Edward B. Tylor. Commentators on Warburg, as far as I can determine, have not really paid much attention to this anthropological source. At most, they have considered only the differences. Ernst Gombrich, for example, argued that the “science of culture” called for by Tylor could never find favor in the eyes of a disciple of Burckhardt’s concerned primarily with Italian art. And yet, this “science of culture” was enthroned at the beginning of Primitive Culture (published in London in 1871), a work that had such an impact that by the end of the 19th century anthropology was called “Mr Tylor’s science.” Of course, a work’s fame, even if it is immense, as in this case, is not sufficient to guarantee its status as a theoretical source. It is, above all, in the establishment of a specific link between history and anthropology that a point of contact exists between Warburg’s Kulturwissenschaft and Tylor’s science of culture.

Each of them, in fact, aimed at overcoming the virtually never-ending opposition between the model of evolution required by any kind of history and the absence of a temporal dimension that is often attributed to anthropology, an opposition that Levi-Strauss was still criticizing a century later. Warburg opened up the field of art history to anthropology, not only in order to discover in it new objects for study, but in order to open up time to a new approach, as well. Tylor, for his part, wanted to carry out a strictly symmetrical operation. He began by asserting that the fundamental problem of any “science of culture” was that of its “development”; that this development was not reducible to a law of evolution that could be formulated according to the model of the natural sciences and that the anthropologist could not understand what “culture” means except by establishing its history, and even its archaeology:

In working to gain an insight into the general laws of intellectual movement [of culture in general], there is practical gain in being able to study them rather among antiquarian relics of no intense modern interest.

Warburg was certainly never to disavow this methodological principle concerning the importance of studying objects devoid of interest at the current moment: what creates meaning in a culture are often its symptomatic, unthought, or anachronistic aspects. Here we are already in what we may call the phantasmic time of survivals. Tylor introduced it at the level of theory at the beginning of Primitive Culture, observing that the two rival theories of “development of culture”—“progress-theory” and “degeneration-theory”—need to be treated dialectically, the one intertwined with the other. The result will be a kind of temporal knot that is difficult to understand because within it there occur ceaseless intersections of movements tending towards evolution and movements resistant to evolution. Within the space of these intersections there soon appears, as the differential of two contradictory temporal statuses, the concept of survival. In fact, in his attempts to establish a theoretical foundation for his work, Tylor devoted a major portion of his efforts to this concept.


But he had already used the word, as if spontaneously, in another context, and in the midst of another kind of temporal experience: during a displacement, namely on a trip to Mexico. Between March and June 1856, Tylor had scoured Mexico on horseback, making observations and taking thousands of notes. In 1861 he published his travel diaries—his own Tristes tropiques—where there appear, one after the other and seemingly to his own surprise, mosquitoes and pirates, alligators and missionary fathers, slave trading and Aztec vestiges, Baroque churches and Indian costumes, earthquakes and the use of firearms, table manners and ways of counting, museum objects and street fights, etc., etc. Anahuac is a fascinating book because it displays its author’s constant astonishment: astonishment that a single experience in the same place and at the same time could encompass this knot of anachronisms, this mélange of things past and present. Thus, during the Mexican Holy Week festivals he witnessed a number of heterogeneous commemorations, half-Christian and half-pagan. And at the Indian market in Grande he saw a system of numeration that he had previously thought could be found only in pre-Columbian manuscripts. A further example was the coexistence of ornaments of ancient sacrificial knives with the ornaments on the spurs worn by the Mexican vaqueros (Figs. 4, 5).

<insert fig. 4 about here>

<insert fig. 5 about here>

Confronted with all that, Tylor discovered culture’s extreme variety and vertiginous complexity (which one also senses in going through Frazer); but he also discovered something even more overwhelming (which one never senses in going through Frazer): the vertiginous play of time in the present, in the current “surface” of a given culture. This vertigo finds expression, first of all, in the powerful sensation that the present is woven from multiple pasts. (This is something which is obvious in itself, but its methodological consequences are less so.) That is why Tylor thinks that the anthropologist must become the historian of each of his observations. The “horizontal” complexity of what he sees is rooted, above all, in the “vertical” or, to use a linguistic term, “paradigmatic” complexity of time:

Progress, degradation, survival, revival, modification, are all modes of the connexion that binds together the complex network of civilization. It needs but a glance into the trivial details of'our own daily life to set us thinking how far we are really its originators, and how far but the transmitters and modifiers of the results of long past ages. Looking round the rooms we live in, we may try here how far he who only knows his own time can be capable of rightly comprehending even that. Here is the ‘honeysuckle’ of Assyria, there the fleur-de-lis of Anjou, a cornice with a Greek border runs round the ceiling, the style of Louis XIV and its parent the Renaissance share the looking-glass between them. Transformed, shifted, or mutilated, such elements of art still carry their history plainly stamped upon them; and if the history yet farther behind is less easy to read, we are not to say that because we cannot clearly discern it there is therefore no history there.

It is characteristic that this example of survival—one of the very first offered in Primitive Culture—pertains to the formal elements of ornamentation, those “primitive words” found in all discussions of the notion of style. It is likewise characteristic that this survival of forms is expressed in terms of an imprint or stamp. To say that the present bears the mark of multiple pasts is above all to assert the indestructibility of the stamp of time—or of several time periods—on the forms themselves of our present life. Thus, Tylor writes of “the strength of these survivals,” by means of which, as he states, using another metaphor, “old habits hold their ground in the midst of a new culture ... which presses hard to thrust them out.” He also compares the tenacity of the survivals to “a stream once settled in a river bed [that] will flow on for ages,” expressing, again in terms of stamping, what he calls the “permanence of culture.

Here, then, we see a “fundamental problem” raised in which Warburg could have recognized his own investigation of the “permanence” and “tenacity” of ancient forms during long stretches of the history of Western art. But that is not all. Tylor might have explained such permanence in terms of the “essence of culture”—as did many 19th century writers on philosophical anthropology. The central interest of his thought concerning this point, as well as the closeness of his position to Warburg’s approach, is due to a further, decisive point: the “permanence of culture” is not expressed as an essence, a global characteristic, or an archetype, but, on the contrary, as a symptom, an exceptional characteristic, something displaced. The tenacity of the survivals, their very “power,” as Tylor says, comes to light in the tenuousness of miniscule, superfluous, derisory, or abnormal things. It is in the recurrent symptom, in games, in the pathology of language, and in the unconsciousness of forms that survival as such is to be found. Accordingly, Tylor paid great attention to children’s games (bows, slings, rattles, knuckle bones, playing cards—survivals of very serious old practices, stemming from war and divination), just as Warburg would later pay great attention to the practices of Renaissance festivals. He examined the characteristics of language—adages, proverbs, and ways of greeting, just as Warburg later wanted to do for Florentine culture. Most importantly, however, in examining survivals Tyler considered them specifically in terms of superstitions. For him, the very definition of this anthropological concept could be inferred from the traditional Latin meaning of the term superstitio:

Such a proceeding as this would be usually, and not improperly, described as a superstition; and, indeed, this name would be given to a large proportion of survivals generally. The very word “superstition” in what is perhaps its original sense of a “standing over” from old times, itself expresses the notion of survival. But the term superstition now implies a reproach…. For the ethnographer's purpose, at any rate, it is desirable to introduce such a term as “survival,” simply to denote the historical fact.”

This passage allows us to understand why the analysis of survivals in Primitive Culture culminates in a long chapter dedicated to magic, astrology, and all the various forms they assumed. How can we fail to recall here that high point of the Nachleben der Antike reconstructed by Warburg in his analysis of the astrological activities found in the Ferrara frescoes and even in the writings of Martin Luther? In both cases—and this is before the work of Freud—it is a split within consciousness, a logical error, or a nonsensical aspect of an argument that opens a breach in the current state of some historically-produced factor, allowing its survivals to appear. Tylor, before Warburg and Freud, liked to study “trivial details” because of their capacity to make sense of—or rather serve as symptoms—of their own insignificance. (He also called them “landmarks.”) Before Warburg and his interest in the “animism” of votive effigies, Tylor attempted, along with others, it is true, to construct a general theory of this power of signs. Before Warburg and his fascination with the expressive phenomena of gestures, Tylor sought to create, again, like others, a theory of “emotional and imitative language.” And, in his own fashion, again before Warburg and Freud, he made a case for the exceptional capacity of the symptom—whether it be absurdity, a lapsus, illness, or madness—to act as guide to the vertiginous temporal dimensions of the survivals existing within a given culture. Might the path indicated by the symptom prove to be the best way of hearing the voice of the phantoms?

It may perhaps be complained that ... throughout the whole of this varied investigation ...of the dwindling survival of old culture its illustrations should be so much among things worn out, worthless, frivolous, or even bad with downright harmful folly. It is in fact so, and I have taken up this course of argument with full knowledge and intent. For, indeed, we have in such inquiries continual reason to be thankful for fools. It is quite wonderful, even if we hardly go below the surface of the subject, to see how large a share stupidity and impractical conservatism and dogged superstition have had in preserving for us traces of the history of our race, which practical utilitarianism would have remorselessly swept away.


In the domain of the historical and anthropological sciences, the notion of survival, located between those of phantom and symptom, may be considered a specific expression of the trace. Warburg, as is well known, took a great interest in the vestiges of classical antiquity: vestiges which were in no way reducible to the objective existence of material remains, as they subsisted just as often in a society’s forms, styles, behaviors, and psyche. One easily understands his interest in Tylor’s survivals. In the first place, they referred to a negative reality, namely what appears to be a discarded element in a culture, something which is no longer of its time and no longer of any use. (For example, the Florentine bòti, in the 15th century, bore testimony of a practice already cut off from the present and from the “modern” concerns behind the creation of Renaissance art.) In the second place, according to Tylor the survivals refer to a masked reality; something persists and testifies to a vanished stage of a society’s history, but its very persistence is accompanied by an essential modification—a change in status and change in meaning. (For example, to say that the bow and arrow of ancient warfare have survived in children’s games is, obviously, to say that their status and meaning have completely changed.)

In this respect, the analysis of survivals clearly appears to be a matter of analyzing symptomatic manifestations as much as phantasmal ones. Survivals refer to a level of reality that we might call “breaking and entering” [réalité d’éffraction], a tenuous, even imperceptible reality; and thus one could also say they refer to a spectral reality. Thus, astrological survivals appear in Luther’s writings as “phantoms,” phantoms whose efficacy Warburg was able to detect thanks to their nature as intruders—and as the intrusion of a symptom– in the logic of Luther’s argumentation. It is not surprising that the first area in which Tylor’s notion of survivals found employment was in the study of beliefs: its most numerous applications were in the history of religion. Even so, some archaeological studies concerned with long periods of time—anticipating what André Leroi-Gourhan later named “technical stereotypes”—have also succeeded in approaching the history of objects in terms of this notion of survival.

<1>Evolution’s Destinies, Heterochronous States

It must be said, however, that the notion of survival has never had a very good press—and that is true not just in art history. In Tylor’s period, survival was accused of being a concept that was too structural and abstract, a concept completely resistant to any precision or factual verification. The positivist objection consisted in asking: how do you go about dating a survival? This showed a complete misunderstanding of a concept meant, precisely, to describe a kind of temporality that is not “historical,” at least not in the trivial, factual sense of the term. Today, one would more likely accuse survival of being insufficiently structural: a concept, in short, that bears the evolutionist stamp. Accordingly, it is considered outmoded and irrelevant, an old scientistic phantom typical of the 19th century. This is what one tends to assume, without giving the matter much thought, in the light of modern anthropology, which, from Marcel Mauss to Claude Lévi-Strauss, has supposedly produced the necessary reorientation of ethnological concepts that were too deeply marked by essentialism (as in Frazer) or by empiricism (as in Malinowski).

When one begins to examine the question more closely, however, it becomes obvious that matters are more nuanced and complex than first appeared. What is really under debate is not the notion itself of survival, but rather the use to which it was put by several late 19th century Anglo-Saxon ethnographers. Mauss, for example, has no hesitation in employing the term in his own work: chapter three of his Essay on the Gift is entitled “Survival of those principles [that establish “the exchange of gifts”] in Ancient Laws and in Ancient Economies.” There he explains that the principles of the gift and of the counter-gift are to be considered “survivals” by the historian as well as by the ethnologist. They have a general sociological value, since they allow us to understand a stage in social evolution. But there is more to them than this: they also have a bearing on social history. Institutions of this type have really provided the transition towards our own forms of law and economy. They can serve to explain historically our own societies. The morality and the exchange practiced in societies immediately preceding our own still retain more-or-less important traces of all the principles we have just analyzed [in the framework of so-called “primitive” societies”].

Elsewhere, Mauss will go as far as to extend the notion of survival to the “primitive” societies themselves:

There is no known society that has not evolved. The most primitive men have an immense past behind them; diffuse tradition and survival therefore play a role even among them.”

This amounts to saying not only that “primitive” societies have a history—which was long denied by some, and is reflected in the expression “peoples without history”—but also that this history can be as complex as our own. It, too, is composed of conscious transmissions and of “diffuse traditions,” as Mauss writes. It, too, is constituted through the play of—or in a knot of—heterogeneous temporal phases: a knot of anachronisms. It is just that this is hard to analyze in the absence of written archives. When Mauss critiques the uses made of the notion of survivals, it is, therefore, not in order to question the appropriateness of employing models of time characterized by this kind of complexity. On the contrary, it is in order to refute ethnological evolutionism as an over-simplification of the required models of time. Thus, when Frazer describes “survivals” of the “ancient confusion of magic and religion,” and Mauss responds that “the hypothesis really explains very little,” we need to understand that what Mauss is objecting to is the following hypothesis: that the confusion between magic and religion was followed by the emergence of the latter as an autonomous sphere, one which was more rational, more moral, in short, more “evolved.”

Mauss also critiqued, clairvoyantly, what continues to be the other basic trap of any analysis of survivals: one could call it archetypism. It terminates not in the simplification of the models of time, but in their negation, pure and simple, and in their dilution in an essentialist view of culture and of the psyche. The key element in this trap is the decoy of analogous perception. When the resemblances become pseudomorphisms, and when they serve, beyond that, to produce a general and non-temporal meaning, then, of course, “survival” becomes involved in a myth-making process, and turns into an epistemological obstacle. Let us note right here that Warburg’s Nachleben has, on occasion, been interpreted in this manner and employed to such ends. But Warburg’s philological effort, his perception of singularities, and his constant attempt to keep track of all the various strands, to identify each thread—even though he knew that some strands slipped from his hands, had been broken, and ran in underground channels—all of that distances his notion of Nachleben from any essentialism. What we might term symptomatic anamnesis clearly has nothing in common with archetypal generalization.

Lévi-Strauss’s critique, set forth in the introductory chapter of his Structural Anthropology, seems a good deal harsher. That is because it is more radical but, at the same time, more one-sided and, at times, burdened with inaccuracies and possibly even a hint of bad faith. He begins, following Mauss, by criticizing archetypism and its erroneous use of substantialized analogies and of pseudomorphisms in the service of universalism. Looking for traces of this approach in Tylor himself, he notes that the bow and the arrow do not form one “species,” as Tylor states in language based on the biological link of reproduction: for, “there will always be a basic difference between two identical tools, or two tools which differ in function but are similar in form, because one does not stem from the other; rather, each of them is the product of a system of representations.” Let us remark, in passing, that Warburg would have unhesitatingly subscribed to this first assertion; for it amounts to making the organization of symbols the foundational structure of the empirical world.

Lévi-Strauss takes a further, less well advised, step when he writes that studies focused on the problematic of survivals “do not teach us anything about ... unconscious processes in concrete ... experiences.” He himself invalidates this assertion a few pages later in according Tylor the status of virtual founder of the analysis of the “unconscious nature of collective phenomena.” Yet, in his eyes, Tylor’s anthropological work remains devoid of any concern with history, and in this regard he simply cites a brief passage of Researches into the Early History of Mankind (1865), without paying attention to the book’s very title and, above all, without recognizing the ideas on the historicity of primitive societies that Tylor elaborated six years later in Primitive Culture—ideas that Lévi-Strauss clearly wished to credit exclusively to Franz Boas. In 1952, the author of Structural Anthropology asserted that the historicity of primitive peoples is “beyond our reach” (“hors d’atteinte”), which is a completely unconscious paraphrase of the passages from Tylor cited above.

None of this alters the basic question facing us; we still don’t know what “survival” means. The first thing to establish is to what extent, if any, this concept derives from evolutionist doctrine—in terms of both content and of what is at stake. When, in the seventh chapter of his book Researches into the Early History of Mankind, devoted to the “Development and decline of civilization,” Tylor sprinkles his text with references to Darwin, the stakes are clearly polemical: he needs, at this juncture, to play off human evolution against divine destiny, that is to say, the Origin of Species against the Bible itself. He needs to rehabilitate “developmentalism” and its links to the notion of the species against the religious theories of degeneration and their links to the notion of original sin.

A further observation should be made here: at the moment when Tylor starts making these references in his texts, he has not yet elaborated a vocabulary concerning “survival.” Even though the debate over evolution does constitute his overall epistemological horizon, Tylor, in constructing his notion of survival, clearly displays his independence with regard to the doctrines of Darwin and Spencer. Where natural selection speaks of the “survival of the fittest,” which guarantees biological innovation, Tylor approaches survival in an inverse manner, from the angle of the most “unfit and inappropriate” cultural elements, the bearers of a bygone past rather than of an evolving future.


In short, survivals are only symptoms, bearers of temporal disorientation. They are in no sense the initial indicators of a teleological process, of any “evolutionary direction” whatsoever. They do bear witness, certainly, to a more original, and repressed, state, but they say nothing concerning evolution itself. They doubtless possess some diagnostic value but have no prognostic value at all. Let us recall, finally, that, according to Tylor, a theory of culture ought no more to be based in biology than in theology. For him, “savages” are no more the fossils of an original human group than they are degenerate examples of God’s image. His theory aimed rather at an historical and philological point of view, which is sufficient to explain its attractiveness to Warburg.

One thing is certain: Warburg’s concept of survival (Nachleben) was initially sketched out in an epistemic field in which anthropological subjects and the central preoccupations of evolutionist theories were major elements. In this regard, Ernst Gombrich asserts, Warburg remains a “man of the 19th century”; accordingly, he concludes, his art history has aged, its basic theoretical models having become outmoded. The simplification is brutal, and not free of bad faith. At best, it shows how difficult it was for iconologists of the second generation to administer a patrimony that was clearly too “phantasmal” to be “applied” in the form in which they received it. At worst, this simplification aimed at blocking off again precisely the theoretical paths the notion of Nachleben had opened up.

Warburg the “evolutionist”—what can that mean? That he read Darwin? There is not a shadow of doubt about that. That he promoted an “idea of progress” in the arts and adopted a “continuist model of time”? Nothing could be further from the truth. The theory of evolution, of course, introduced the question of time into the life sciences, moving beyond that “long cosmic duration”—as Georges Canguilhem put it—that still constituted the framework of Lamarck’s thinking. But, to raise the question of time is to raise the question of times, that is to say, of the different temporal modalities manifested, for example, by a fossil, an embryo, or a rudimentary organ. Patrick Tort, has shown, moreover, that it is a complete mistake to consider the philosophy of Herbert Spencer—which automatically comes to mind when one hears the word “evolutionism”—as being closely based on the Darwinian theory of biological evolution. The latter is a bio-ecological theory of transformation, in which the emergence of living species is subject to the process of variation; while the former is a doctrine, or better an ideology, of the meaning of history whose conclusions—widespread among the ruling classes and the industrial milieux of the 19th century—are opposed in many respects to those of the Origin of Species.

The misunderstanding is rooted, precisely, in the notion of survival. It was only in the fifth edition of his book that Darwin introduced the Spencerian phrase of “the survival of the fittest.” Today, students of epistemology see only theoretical confusion in the association of these two words (which, Tylor, as we have seen, carefully dissociated). To speak in this manner amounts to linking selection very tightly to survival: the fittest, the strongest survive and multiply. The idea that this law might be relevant to the historical and cultural world comes from Spencer, not from Darwin, who, instead, saw in civilization a way of opposing natural selection, of becoming “unfit.” In this sense, Warburg was no doubt a Darwinian, but not an evolutionist in the Spencerian sense.

For Warburg, Nachleben made sense only if it was used to complexify the notion of historical time, to recognize in the world of culture the existence of specific, non-natural temporal modes. To base a history of art on “natural selection,” i.e., the successive elimination of the weaker styles, with this elimination bestowing on change or becoming its perfectibility and on history its teleology—nothing could be further from his basic project or from his temporal models. The surviving form, in Warburg’s understanding of the term, does not triumphantly survive the death of its competitors. Quite to the contrary, it survives, as symptom and as phantom, its own death. Having disappeared at a certain point in history; it reappears much later, at a moment when, perhaps, it is not expected; it has survived, therefore, in the still poorly defined limbo of a “collective memory.” Nothing could be further from this idea than the “synthetic,” authoritarian, and highly systematic notions of Spencer’s so-called “social Darwinism.” On the other hand, one could trace links between this idea of survival and certain Darwinian statements concerning the complexity and the paradoxical interpenetration of biological times.

From this point of view, Nachleben could be compared—although not assimilated—to models of time which allow for a symptomatic interpretation of certain cases within the framework of the theory of evolution, that is to say, models which create difficulties for all schemas of adaptation that stress continuity. Theorists of evolution have spoken of “living fossils,” those creatures that have survived but are completely anachronistic. They have spoken of “missing links,” those intermediate forms in a series of variations situated between older stages and recent ones. With the concept of “retrogression” they have indicated their refusal to oppose a “positive” evolution to a “negative” regression. They have also spoken not only of “panchronistic forms”—living fossils or surviving forms, i.e. organisms that have been widely found in the fossil state and that were believed to have disappeared but that are suddenly discovered, in certain conditions, in the state of living organisms —but also of “heterochronies,” those paradoxical states of a living organism in which heterogeneous phases of development are found combined. When the normal processes of natural selection and genetic mutations cannot account for the development of a new species, they have even spoken of “hopeful monsters,” “non-competitive” organisms that are nevertheless capable of engendering a radically divergent, original evolutionary line.

In its own way, Warburg’s Nachleben is really only concerned with “living fossils” and “retrogressive” forms. It allows for “heterochronies,” and, indeed, “hopeful monsters”—like the prodigious sow of Landser’s, with two bodies and eight trotters, that Warburg, after seeing it in an engraving by Dürer, discussed in terms of what he called a “world of prophetic freaks” (Region der wahrsagenden Monstra). But one also sees how a misunderstanding can arise when the label “evolutionist” is applied to a body of work as experimental—and also as unsettled and heuristic—as Warburg’s.


In order to get a better grasp of the anachronistic and extraordinary object of his quest, Warburg proceeded like all pioneers: he cobbled together a system of disparate borrowings, reorienting them using the “good neighbor” approach for each one with respect to all the others. Ernst Gombrich revealed, but also over-estimated, his use of the heterodox evolutionism of Tito Vignoli. This positivist source should really be placed alongside the romanticism of Carlyle, for example, from which Warburg drew further arguments in favor of that questioning of history that always arises from recognizing the phenomena of survival. His influence on Warburg was not limited to just the “philosophy of the symbol,” and of clothing, found in the strange book entitled Sartor Resartus, to which we shall return. In the same context, Carlyle sketched a veritable philosophy of history in dialogue with the whole of German thought, including that of Lessing, of Herder, of Kant, of Schiller, and, of course, of Goethe.

It was a philosophy of distance (history as that which puts us in contact with what is distant) and of experience (history as philosophy teaching by example); it was a philosophy of the vision of times, at once prophetic and retrospective; it was a critique of prudent history, and an encomium of artistic history; it was a theory of the “signs of Times” that Carlyle himself defined as “hyperbolic-asymptotic,” always in search of limits and of unknown depths. Whereas he considered history in the trivial sense to be successive, narrative, and linear, Carlyle spoke of time as an eddy composed of innumerable and simultaneous acts and “solids,” which he wound up calling the “chaos of Being.”

It is not without interest to observe that in 1890 Wilhelm Dilthey commented on this philosophy of history in relation to his own “critique of historical reason.” In their very different ways, and though they disagreed on many points, Carlyle and Dilthey were thus able to furnish the young Warburg with several conceptual tools he later used in constructing, little by little, the temporal model of his own emerging Kulturwissenschaft. The opening up of art history to anthropology could not fail to modify its own schemas of intelligibility, its own determinants. And whether he wanted to or not, Warburg found himself taking part in a polemic that at the end of the nineteenth century was opposing the positivist historians or “specialists” to the proponents of an expanded Kulturgeschichte, such as Salomon Reinach and Henri Beer in France, and, in Germany, Wilhelm Dilthey and Warburg’s own teacher Karl Lamprecht.

What should we conclude from this play of borrowings and debates if not that evolutionism produced its own crisis, its own internal critique? In recognizing the necessity of enlarging the canonical models of history—narrative models, models of temporal continuity, models based on the assumption of the attainability of objectivity—and in moving slowly towards a theory of the memory of forms—a theory composed of jumps and latencies, of survivals and anachronisms, of desires and unconscious motives—Aby Warburg effected a decisive rupture with the very notions of historical “progress” and “development.” He thereby set evolutionism against itself, deconstructing it simply by identifying and recognizing the importance of those phenomena of survival, those cases of Nachleben to which we now must turn our attention, with the aim of understanding his specific elaboration of them.

<1>Renaissance and the Impurity of Time: Warburg with Burckhardt

Warburg elaborated the notion of Nachleben within a very specific historical framework, one which formed virtually the exclusive domain of his published studies. It encompassed, first of all, the Italian Renaissance (Botticelli, Ghirlandaio, and Francesco del Cossa, but also Pico della Mirandola), and, secondly, the Flemish and German Renaissance (Memling, Van der Goes, and Dürer, but also Luther and Melanchthon). If we consider such a notion today, it of course seems to offer us a theoretical lesson capable of “refounding,” as it were, several major presuppositions concerning our knowledge of images in general. But we should not forget that Warburg formulated the problem in the context of the Renaissance in particular. We should not expect him to provide something that he never promised (which is what Gombrich does, for example, when he reproaches him for having “virtually omi[tted] medieval art” when speaking of survivals).Whatever general value the notion of Nachleben may possess results from a reading of Warburg, and thus of an interpretation of him; we are the only ones responsible for that interpretation.

Let us agree that, in any case, Warburg has a certain taste—though subtle and surreptitious—for provocation. Is it not provocative of the historian-philosopher to juxtapose head on in his work two notions as different as “survival” and “renaissance”? Of course, in German, the word “Renaissance” means an historical period: it does not, as in French or Italian, spontaneously refer to a process that, at the time, referred to the “survival of antiquity”(Nachleben der Antike). But the impression persists that there is something irritating about the confrontation of these two words. We must observe, in fact, that neither of them emerges unaffected by this pairing. The Renaissance, as the Golden Age of the history of art, loses some of its purity and of its completeness. Reciprocally, survival, as an obscure evolutionary process, loses something of its primitive or prehistoric aura.

But why this context? Why the Renaissance? Why, in particular, begin or begin again—I am thinking of Warburg’s thesis on Botticelli, his first published work —with the Italian Renaissance? First of all, because that is precisely where art history, conceived as a branch of knowledge, had begun or begun again. Warburg and Wölfflin, before Panofsky, reinvented the discipline of art history by returning to the humanist conditions, that is to say, to the Renaissance conditions, of an order of discourse that had not always existed as such. Entering into the Renaissance—entering into art history by the royal road of the Renaissance—also meant, for a young scholar at the end of the nineteenth century, entering into a theoretical polemic about the very status, about the style and the stakes of historical discourse in general.

This polemic goes back to Jules Michelet, who, in several celebrated formulas, sketched, for the first time, a properly historical and interpretative notion of the Renaissance: “the discovery of the world and of man,” “the advent of modern art,” “the free flight of fantasy,” the return to antiquity conceived of as “an appeal to the living forces,” and so on. Let us try to relativize what today appears banal or even questionable in these expressions; for when Warburg followed the courses of Henry Thode at the University of Bonn, he probably heard a hundred recriminations concerning this “modern “ Renaissance, perceived essentially as the moment of the invention of an anti-Christian morality. Recriminations addressed less to Michelet himself than to two German thinkers guilty of having pushed such formulas to their extreme consequences. These two authors are none other than Jacob Burckhardt and Friedrich Nietzsche. The polemic, one suspects, was not only about the status, Christian or not, of the Italian Renaissance, but also about the status of historical knowledge itself, of its philosophical and anthropological ambitions. At the heart of this polemic lay nothing less than a struggle over the new Kulturgeschichte inaugurated by Nietzsche and by Burckhardt.

It is clear that between Thode’s “Franciscan” lectures and Burckhardt’s “modern” writings, Warburg did not hesitate a moment. The former’s name is not cited even once in the pages of the Gesammelte Schriften, whereas the latter’s influence is acknowledged throughout them. A single example will suffice to bring out this contrast: in his article of 1902 on the Florentine portrait, Warburg begins, precisely, with a topic involving Franciscan iconography—The Confirmation of the Rule of the order of St. Francis, portrayed by Giotto in the Church of Santa Croce and by Ghirlandaio in Santa Trinita—which renders the absence of any reference to Thode all the more flagrant. Indeed, Warburg simply left unmentioned the fact that his anthropological interpretation of Ghirlandaio’s cycle contradicted point by point the schema proposed by Thode in his own work on the Renaissance. In contrast, the same text opens with a vigorous theoretical statement dominated by the authority of Burckhardt:

With all the authority of genius, that model pioneer (vorbildlicher Pfadfinder), Jacob Burckhardt, dominated the field that he himself had opened up for scholarship: that of Italian Renaissance civilization (Kultur der Renaissance). But it was not in his nature to be an autocratic exploiter of the land (Land) he had discovered. Such, indeed, was his self-abnegation as a scholar (wissenschafatliche Selbstverleugnung) that, far from yielding to the temptation of tackling the cultural history of the period as a whole (Einheitlichkeit), he divided it into a number of superficially unrelated sectors (in mehrere äusserlich unzusammenhängende Teile), which he proceeded to explore and describe with magisterial poise and authority. On the one hand, in Die Cultur der Renaissance in Italien, he discussed the psychology of the individual in society without reference to visual art; on the other, in Cicerone, he undertook to offer no more than "an introduction to the enjoyment of works of art”…. Our perception of the greatness of Jacob Burckhardt must not deter us from following in his footsteps.

This “path” (Bahn) demands a methodological rigor that is extremely difficult to maintain. But it led Warburg’s “humility”—his Selbstverleugnung [self-denial] as he puts it here—to reach the level of humility he recognized in Burckhardt. This attitude could almost be called Stoic. On the one hand, it meant recognizing the unity (Einheitlichkeit) of all culture, its fundamentally organic nature. On the other hand, however, it meant refusing to assert it, to define it, or to claim one has grasped it as such: things are to be left in their state of division or of “disassembly” (Zerlegung). Like Burckhardt, Warburg always refused to complete [reclore] a system, which was his way of always postponing the moment of conclusion, the Hegelian moment of absolute knowledge. It was necessary, he thought, to push “humility,” or epistemological modesty, to the point of recognizing that an isolated researcher—a pioneer—can and should work only on singularities, as Warburg well expresses it on the same page, presenting the paradox of a “synthetic history” which, however, consists of “particular studies,” that is, studies that are not placed in any hierarchical order:

Even after his death, this connoisseur [Burckhardt] and scholar of genius presented himself to us as a tireless seeker: in his posthumous Beiträge für Kunstgeschichte von Italien, he opened up yet a third empirical path to the great objective of a synthesis of cultural history (synthetische Kulturgeschichte). He undertook the labor of examining the individual work of art within the immediate context of its time, in order to interpret as "causal factors" the ideological and practical demands of real life (das wirkliche Leben).

Wölfflin, too—the other great 20th century “re-inventor” of art history—admired in Burckhardt a master capable, precisely, of creating a “systematic history” in which the “system” was never defined, that is to say, completed, schematized, and simplified. With Burckhardt, his “sensitivity to the individual work” always predominates, leaving any conclusion an open one. Now, no one has been better able than Warburg to accomplish—if such a verb may be used here—the paradoxical task so well expressed in his text by the verb zerlegen, “to decompose” or “take apart.” No one in the field of art history has ever traveled with such daring along the path of this infinite analysis of singularities—an analysis which, because of its lack of completion, has wrongly been considered “imperfect” or “unfinished.”

The modesty and humility Warburg displayed with respect to the historical “monument” erected by Burckhardt are neither false nor simply a matter of politeness. They do not, however, mean that the later body of work is purely the offspring of the earlier one. In his personal notes, Warburg is quite willing to be more critical, readier to discuss certain issues and even take an opposing position. It should also be said that Warburg’s basic vocabulary—that of Nachleben, of Pathosformeln, and of the theory of “expression” (Ausdruck)—does not figure among Burckhardt’s own conceptual tools. Yet, one can not help thinking that Warburg’s famous Notizkästen—his multicolored cardboard file boxes—are so to speak, the three-dimensional incarnation of the Materialien that Burckhardt had assembled with a view towards writing a “History of Renaissance Art,” a project continually placed on hold and never published (Figs. 6 and 7). It is worthwhile, in any case, to determine what elements among the work of the great historian from Basel could have nourished the intuitions and intellectual constructions of the young Warburg.

<insert fig. 6 about here>

<insert fig. 7 about here>


Entering into art history by the “royal road” of the Florentine Renaissance, as Warburg did in 1902 (in the study of the portrait) and, earlier, in 1893 (in the study of Botticelli), meant taking a position with respect to the very concept that Burckhardt had forged throughout his livre-fleuve, The Civilization of the Renaissance in Italy. That book’s themes and theses have been endlessly glossed. Commentators have recognized its audacity, its ambitious scope, and its “brilliant,” animated presentation. Some have written admiringly of its way of unifying extraordinarily rich and highly varied historical material. On the other hand, every famous theme—the opposition of the Middle Ages and the Renaissance, the primacy of Italy, the “development of the individual”—has come in for criticism. It has also been observed that beyond all the critiques, the book still dominates historical debate concerning the notion of the Renaissance. The fact of this oppressive “domination” has been used to argue that, with his masterpiece, Burckhardt created a mythical Renaissance whose myths fostered a cult which ultimately yielded what Heinrich Mann censured in the expression hysterical Renaissance.

If there is indeed a myth of the Renaissance, this myth is intrinsic to Renaissance culture itself—and Burckhardt is the one who analyzed it as such. The “development of the individual” probably does derive from a mythical structure, in any case an ideological and political structure. It nonetheless generated effects in the realms of knowledge and style, and in the realms of truth and of history. If the “individual” is a Renaissance myth, it at least created those fascinating realities that are the Quattrocento Florentine portraits. Now, this is precisely where Warburg started from: to analyze a myth, to trace the ramifications of its aesthetic effects, required one to both gauge its fecundity (as a “science of the concrete”) and deconstruct it (as an ensemble of phantasms).

Burckhardt’s analysis, therefore, did not interest Warburg because it offered a few generalizations explaining how the Renaissance, as a culture or a period, might have emerged entirely pure and conceptually “armed,” like Athena emerging from the head of Zeus. Burckhardt did recognize a “development of the individual” in Renaissance Italy, but this “development” found a strange conclusion in an analysis of the symptoms and of the mental traits, of the parodies and of the defamations—in themselves obstacles to any trivially evolutionary model—of which the individual, from Franco Sacchetti to Aretino, was a constant victim. Burckhardt spoke, therefore, of the “development of the individual” not only as ever-increasing emancipation but also as a development of [the individual’s] perversity.

One can derive two very different interpretations from this analysis. The first is moralizing: it follows the “greatness and decline” model of the pessimistic outlooks of the 18th century. It rightly sees a connection between Burckhardt and Schopenhauer. But, in stressing the theme of decline it winds up viewing Burckhardt as nothing but a reactionary ideologue, an anti-democratic precursor of Spengler’s type of Kulturpessimismus, and even a partisan of the “conservative revolutions” which, in Germany, prepared the way for Nazism. The other interpretation is structural: it is more intent on detecting the workings of history than in the judgments of history. It has the advantage—which Warburg fully understood—of being dialectical and, for that very reason, epistemologically fecund. When Burckhardt castigated “modern ... culture” and its incapacity to “understand Antiquity,” he was not so much offering a “reactionary” judgment as he was drawing attention, in a critical fashion, to the more general problem of the relationship between a culture and its memory; for a culture which represses its own memory—its own survivals—is just as likely to become powerless as a culture immobilized in the perpetual commemoration of its past. Walter Benjamin’s view of this matter, it seems to me, was no different.

The “development of the individual” in the Renaissance thus contains within itself the development of the individual’s symptoms—encompassing perversions and negative qualities in general. What should one conclude from this proposition? The proponent of a moralistic view would speak of a “decline,” asserted in the name of a certain “purity,” though it is unclear whether or not one should, like Winckelmann, locate such purity exclusively in the time of the “Greek miracle.” A structural point of view understands that time—whatever that time may be, whether of Antiquity or the Renaissance—is impure. This is the kind of interpretation, I believe, that provided the starting point for all of Warburg’s subsequent work, inasmuch as he was able to use Burckhardt’s analyses to construct an incisive notion of this impurity of time—to construct, in short, the theoretical foundation of the notion of “survival.”


Right from the start, Burckhardt had decided to take the measure of the complexity of times that he saw as an essential characteristic of the Renaissance, finding it impossible—and historically pernicious—to sum up the period as consisting in the appealing science of a Leonardo, the angelic expression of a Raphael, or the genius of a Michelangelo. A half-century before Freud defined his “fundamental rule” of non-omission. Burckhardt wrote that historians “must not seal [them]selves off from anything past”; the lacunae, the dark areas, the counter-themes, and the aberrations are all part of his quest. That is why the famous “development of the individual” should be considered in terms of what Burckhardt called a “mixture of ancient and modern superstitions” characteristic of Renaissance Italy. (Warburg later undertook a similar analysis for the Germany of Luther and Melanchthon). Where Robert Klein saw in Burckhardt “a certain opposition between the two orientations of the Renaissance”—the positive spirit of the “discovery of man and of the world” and the fantastic spirit of esoteric fictions —we are tempted to recognize something like a dialectical clairvoyance, a way of thinking centered on tensions and polarities, which Warburg, for his part, went on to systematize at each level of analysis.

Given all this, it is hard to see how the famous “resurrection of Antiquity” could be thought of, with regard to its temporal aspects, as a pure and simple return of the “same” (the same “ideal of beauty,” for example). It is its relationship—inescapably anachronistic—to a specific time and place, Italy of the 15th century, which leads this return to be bound up with differences, with complexities, and with metamorphoses. It is the encounter of the long period of survivals—Burckhardt does not designate them by this term, writing instead that “…. this Antiquity had made its influence felt for a long time”—with the short period of stylistic decisions that makes the Renaissance such a complex phenomenon.

This is why Burckhardt, with regard to the historical concept of rebirth – which he noted in the verbal form (renaître) in French, in a manuscript of 1856—was able to describe a true dialectical movement: between the temps-coupure (or period of rupture) of what he called the “reprise” of the ancient past and the temps-remous (or period of slow stirring) of the “vital remains” (lebensfähige Reste), which had long remained latent, though efficacious in a sense, at the very heart of the “long interruption” that caused them to go unperceived. Antiquity is not a “pure object of time” which returns as such when called. It is a great movement of large domains, a silent vibration, a harmonic wave which traverses all the historical layers and all the levels of a culture:

The history of the ancient world, i.e., of all those peoples whose lives have flowed into ours, is like a fundamental chord that keeps sounding through the fields of human knowledge.

In this light, one is less astonished to find Burckhardt penning a proposition as radical—and as scandalous in the eyes of the aesthetic devotees of the Renaissance—as the following: “The Renaissance created no organic style of its own” (kein eigener organischer … Stil). What does that mean? That the Renaissance is impure—both in its artistic styles and in the complex temporality of its comings and goings between the living present and recollected Antiquity. One cannot imagine, in the 19th century, a more pointed critique of historicism (bent on unity of time) or of aestheticism (bent on unity of style).

The Renaissance is impure. Warburg never ceased to explore and deepen this observation, thanks to his specific concepts of Nachleben and Pathosformel. The Renaissance is impure. That perhaps limits it with respect to any ideal, but it is also the source of its very vitality. And this is exactly what Warburg wrote in 1920: the “heterogeneous mixture of elements” (Mischung heterogener Elemente) designates precisely what is “vital” (so lebendig) in the “civilization of the Renaissance” (Kultur der Renaissance). It designates the “composite” character of the Florentine style (Mischstil), and it implies the existence of a constant dialectic of “tensions” and “compromises,” with the result that, in the end, Renaissance culture appears to the historian to be a truly “enigmatic organism”:

When conflicting worldviews (Lebensanschauung) kindle partisan emotions, setting the members of a society at each other's throats, the social fabric inexorably crumbles (Verfall); but when those views hold a balance within a single individual—when, instead of destroying each other, they fertilize each other and expand the whole range of the personality—then they are powers (Kräfte) that lead to the noblest achievements of civilization. Such was the soil in which the Florentine early Renaissance blossomed. The citizen of Medicean Florence united the wholly dissimilar characters (heterogene Eigenschaften) of the idealist—whether medievally Christian, or romantically chivalrous, or classically Neoplatonic—and the worldly, practical, pagan Etruscan merchant. Elemental yet harmonious in his vitality (Lebensenergie), this enigmatic creature (ein rätselhafter Organismus) joyfully accepted every psychic impulse as an extension of his mental range,to be developed and exploited at leisure.

<1>Survival Renders History Anachronistic

The Renaissance is impure, and the notion of survival is Warburg’s way of designating the temporal mode of that impurity. Although not striking, the expression “vital remains” (lebensfähige Reste), in Burckhardt’s writings, seems to me decisive for understanding, going back earlier than Warburg himself, the paradox—and the necessity—of such a notion. It is the paradox of a residual energy, of a trace of past life, of a death barely evaded and almost ongoing: a phantasmal death to put it bluntly, one which gives to this triumphantly named “Renaissance” culture its own principle of vitality. But just what vitality and what temporality are we discussing here? How does survival impose a specific, fundamental way of understanding the “life of forms” and “forms of time” that this life displays?

Our working hypothesis will be that, beyond Burckhardt’s evocation of “vital remains,” Warburg’s Nachleben provides a model of time specifically suited to images, a model of anachronism which breaks not only with Vasarian filiations (those family novels) and Winckelmannian nostalgia (those elegies of the ideal), but also with all the usual assumptions about the meaning of history. The concept of Nachleben, as Warburg understands it, therefore, is linked to a whole theory of history; it is with respect to Hegelianism that we must ultimately take the measure of such a concept and judge it.

Let us observe, to begin with, that Warburg himself was well aware that the “survival of antiquity” was a “central problem” (Hauptproblem) in all his research. His closest collaborators and friends, such as Fritz Saxl and Jacques Mesnil have attested to this:

The library founded in Hamburg by Professor Warburg is distinguished from all other libraries by the fact that it is not devoted to one or several branches of human knowledge, that it does not fit into any of the usual categories, whether general or local, but rather that it has been formed, classified, and oriented with a view to solving a problem, or rather a vast ensemble of connected problems. This problem is the one which has preoccupied Warburg since his youth: what did Antiquity really represent for the men of the Renaissance? What was its significance for them? In what areas and in what ways did it exercise its influence? The question posed in this manner was not for him a purely artistic and literary question. In his mind the Renaissance evoked not only a style but also, and above all, the idea of a culture: the problem of survival and of the renaissance of antiquity is as much a religious and social problem as an artistic one.

The current classificatory scheme at the Warburg Library still testifies to this obsession: virtually every major section begins with a sub-section on the “survival of Antiquity,” encompassing the survival of the ancient gods, of astrological knowledge, of literary forms, of figurative motifs, etc. The volumes of lectures (Vorträge der Bibliothek Warburg), published between 1923 and 1932 by Fritz Saxl, likewise are all marked by this problem. Just opening the first volume we find an article on Dürer as interpreter of Antiquity (by Gustav Pauli) joined by a study on Hellenistic survivals in Arab magic (by Hellmut Ritter), Ernst Cassirer’s famous lecture on the concept of “symbolic form,” and an essay by Adolph Goldschmidt on the “Survival of ancient forms in the Middle Ages” (Das Nachleben der antiken Formen im Mittelalter). All of the bibliographic efforts of the Warburg Institute came together in a two-volume work devoted exclusively to the problem of the survival of Antiquity.

But was this problem really all that new? Had not the neo-classicism of Winckelmann and his followers already projected Antiquity (Altertum) all the way into the living present (Gegenwart) of the men of the 19th century? Ernst Gombrich has insisted on the influence of a text by Anton Springer—the first chapter of his book Bilder aus der neueren Kunstgeshichte, published in 1867—concerning “The Survival of Antiquity in the Middle Ages” (Das Nachleben der Antike im Mittelalter). In the margin of a passage in which Springer speaks of the ancient draped statue as a “perfect tool of expression,” Warburg noted his agreement with a laconic “bravo.”

Warburg, of course, was thoroughly knowledgeable about all the historical literature concerning the problem of the “ancient tradition.” But this knowledge, from our point of view, underscores all the more sharply the difference between his notion of Nachleben and all those others which, in varying guises, were under discussion at the time. How, then, was Warburg’s notion of survival able to break with all the preceding and contemporary ones? Essentially because it alone was not meant to be superimposed on any historical periodization. Springer’s Nachleben simplified history by periodizing it: it allowed one to see a “diminished” Antiquity existing in the form of its survivals in the Middle Ages, as opposed to the “triumphant” Antiquity of the Renaissance. Warburg’s Nachleben, in contrast, is a structural concept. It is as relevant to the Renaissance as it is to the Middle Ages: “Each age has the renaissance of Antiquity it deserves” (jede Zeit hat die Renaissance der Antike, die sie verdient), he wrote. But he could have just as well written, in a symmetric fashion, that each period has the survivals it deserves, or rather, that are necessary to it and, in a sense, underlie it stylistically.


According to Warburg, the notion of survival offers us no way of simplifying history: it confronts any urge we might have towards periodization with a formidable disorientation. It is a notion that cuts across any chronological scheme. It always describes an other time, and thus it disorients history and opens it up, making it more complex. In short, it anachronizes history. It creates the following paradoxical situation: the most ancient things sometimes come after less ancient ones. Thus, the Indian type of astrology—the most ancient there is—came to be used again in Italy in the 15th century after it had been supplanted and rendered out of date by Greek, Arab, and medieval astrology. This single example, developed at length by Warburg, shows how survival disorients history, revealing how each period is woven with its own knot of antiquities, anachronisms, present times, and tendencies towards the future.

Why does medieval knowledge survive in Leonardo? Why does the Northern Gothic survive in the classical Renaissance? Michelet already said that the Middle Ages are “all the more difficult to kill because they have long been dead.” It is the things which have long been dead, in fact, which haunt our memory the most effectively, and the most dangerously. For example, when today’s housewife works on her horoscope, she continues to manipulate the names of ancient gods in whom, it is assumed, no one any longer believes. Survival, therefore, opens up history, which is what Warburg encouraged when he spoke of a “history of art in the widest sense” (wohl zum Beobachtungsgebiet der Kunstgeschichte im weitesten Sinne): a history, namely, open to the anthropological problems of superstition and of the transmission of beliefs. This would be an art history informed by that “psychology of culture” in which Warburg began to take a passionate interest when studying under Hermann Usener and Karl Lamprecht.

To the degree that it enlarges the discipline’s objects, approaches, and temporal models, survival complexifies history: it frees up a kind of “margin of indeterminacy” in the correlation of historical phenomena. What comes “after” almost frees itself from what comes “before” when it joins that phantasmal, surviving “before the before.” This can be seen, for example, in the work of Rembrandt, which Warburg termed “more ancient and more classical”—more Ovidian, in short—than that of an Antonio Tempesta, which preceded it historically. The form almost frees itself from the content, as in the frescoes in Ferrara in which the Renaissance structure—the reciprocal position of the figures, and the astrological reference itself—coexists with an iconography which is still medieval, heraldic and knightly.

This makes it clear that the ideas of tradition and of transmission present a formidable complexity: they are historical (Middle Ages, Renaissance), but they are also anachronistic (the Renaissance of the Middle Ages, the Middle Ages of the Renaissance); they are constituted out of conscious and unconscious processes; of forgetting and rediscovery; of inhibitions and destructions; of assimilations and inversions; of sublimations and alterations—all terms, moreover, that Warburg himself used. A displacement of perspective, through which the historical model of the Renaissance and the anachronistic model of survival became dialectical, was sufficient to turn the very idea of a transmission into a problematic one. All the more so since this complexity, according to Warburg, is accompanied by a stubborn reference to an anthropology based on the linked questions of belief, of alienation, and of knowledge—and of the image, of course:

The history of the influence of Antiquity, as observed through the transmission, disappearance, and rediscovery (überliefert, verschollen und wiederentdeckt) of its gods, has some unexplored insights to contribute to the history of the meaning of anthropomorphic thought (eine Geschichte der Bedeutung der anthropomorphistischen Denkweise).…Thus understood, the images and words (Bilder und Worte) here discussed—a mere fraction of all that might have been brought to light—are to be regarded as hitherto unread records of the tragic history of freedom of thought in modern Europe (die tragische Geschichte der Denkfreiheit). At the same time the intention has been to show, by the example of a positive investigation, how the method of the study of civilization (kulturwissenschaftliche Methode) can be strengthened by an alliance between the history of art and the study of religion (die Verknüpfung von Kunstgeschichte und Religionswissenschaft).

Because it is woven of long stretches of time and of critical moments, of ageless latencies and of brutal resurgences, survival ends up by anachronizing history, thereby eroding any chronological notion of duration. In the first place, survival anachronizes the present: it violently contradicts the obvious facts presented by the Zeitgeist, that “spirit of the age” on which the definition of artistic styles is so often based. Warburg liked to cite Goethe’s statement that “what is called the spirit of the age (Geist der Zeiten) is, in reality, nothing more than the spirit of the worthy historian in whom this age is reflected.” Consequently, Warburg gauged the greatness of an artist or of a work of art—in opposition to what a too readily accepted sociological reading of his work would have us believe—according to its capacity to resist such a spirit, such a “spirit of the age.”

In the second place, survival anachronizes the past: if Warburg analyzed the Renaissance as an “impure time,” it was also because the past from which it summoned up its “living forces,” namely, classical Antiquity, was itself very far from having an absolute origin. Consequently, the origin itself is an impure temporality characterized by hybridizations and sediments, by protensions and perversions. Thus, in the pictorial cycles at the Schifanoia Palace, what survives is an Oriental model of astrology in which the more ancient Greek forms had already undergone a long process of alteration. As soon as the art historian takes the risk of recognizing the longues durées at work in the artistic monuments of the Renaissance—as Warburg did in presenting together a work by Raphael and the Arch of Constantine in Rome, created 1,200 years apart —he quite logically exposes himself to the risk of anachronism. Let us call this a decision to recognize anachronism at work in historical evolution itself.

For the notion of survival indeed opens a breach in the usual models of evolution, detecting within the latter paradoxes, ironies of fate, and non-linear changes. It anachronizes the future inasmuch as it is considered by Warburg to be a “the force which determines style” (als stilbildende Macht). The fact that Luther and Melanchthon reveal their interest in the “arcane survivals of paganism” (and den fortlebenden mysteriösen Praktiken heidnischer Religiosität) of course seems “a paradox in terms of any rectilinear view of history” (geradlinig denkende Geschichtsauffassung). But that is precisely what fully justified Warburg’s call for a model of time specific to the history of images: what he called, as we have seen, a search for “an evolutionary theory of its own” (ihre eigene Entwicklungslehre).


Now we are somewhat better prepared to understand the paradoxes of a history of images conceived as a history of phantoms, in which survivals, latencies, and returns [revenances], all take part in the most clearly marked developments of periods and styles. One of Warburg’s most striking formulations, dating from 1928, a year before his death, was his definition of the kind of history of images that he pursued as “ghost stories for grown-ups” (Gespenstergeschichten für ganz Erwachsene). But whose ghosts are these? when and where do they come from? Warburg’s admirable texts on the portrait, with their mixture of archaeological precision and melancholic empathy, at first make one think that these ghosts are a matter of persistence, of the survival of a post-death state.

At the time he was working on the portraits of the Sassetti family (a family of bankers, like his own family), Aby Warburg wrote his brother Max a moving letter in which he tried to describe how it was that all his archival work, however “arid” (eine trockene Arbeit) it might be, became “tremendously interesting” (colossal interessant) as soon as he was able to restore to a kind of life, even of palpitation, those “phantom-like images” (schemenhafte Bilder) of beings who had disappeared so long ago. With this in mind, we can better understand the paradoxical “liveliness” of the Florentine portraits (that is to say, their physical relationship with death) and, consequently, their very powerful “animism” (that is to say, their psychical relationship with the inanimate). After all, was it not on the ancient sarcophagi, those caskets of death, that the artists of the Renaissance—from Nicola Pisano to Donatello and beyond—scrutinized the classical formulas for representing life itself, that “life in motion” which survived, fossilized, as it were, in the marble of the Roman remains?

But that is not all. The phantoms of this history of images also emerge from an inchoate past: they can be seen as the survival of what we might call a “pre-birth.” Their analysis should teach us something decisive concerning what Warburg rightly termed the “formation of a style,” its “morphogenesis.” The model of Nachleben, therefore, is not applicable solely to a quest for disappearances; rather, it seeks the fecund element in the disappearances, that which yields a trace and, accordingly, is capable of becoming a memory, of returning, indeed, of a “renaissance.” With this scheme, we have, speaking in epistemological terms, something like a redefinition of the biomorphic model of evolution.

Life, death, and renaissance, progress and decline—in other words, the models habitually used since Vasari—are no longer sufficient for describing the symptomatic historicity of images. Darwin, of course, dealt with these same issues, as can be seen in his analysis of “accidental appearances”—truly symptoms, or malaises dans l’évolution—where he describes in a remarkable fashion the “return of lost characteristics” and the notion of the “latencies” through which the biological structure of the “common ancestor” survives”:

With pigeons, however, we have another case, namely, the occasional appearance in all tile breeds, of slaty blue birds with two black bars on the wing, white loins, a bar at the end of the tail, with the outer feathers externally edged near their basis with white. As all these marks are characteristic of the parent rock-pigeon, I presume that no one will doubt that this is a case of reversion, and not of a new yet analogous variation…. No doubt it is a very surprising fact that characters should reappear after having been lost for many, probably for hundreds of generations…. In a breed which has not been crossed, but in which both parents have lost some character which their progenitor possessed, the tendency, whether strong or weak, to reproduce the lost character might, as was formerly remarked, for all that we can see to the contrary, be transmitted for almost any number of generations. When a character which has been lost in a breed, reappears after a great number of generations, the most probable hypothesis is, not that one individual suddenly takes after an ancestor removed by some hundred generations, but that in each successive generation the character in question has been lying latent, and at last, under unknown favorable conditions, is developed.

<1>Exorcism of the NACHLEBEN: Gombrich and Panofsky

Before inquiring into the conditions under which, in the history of art, an ancient form becomes capable of surviving in certain cases and of undergoing a renaissance in others, let us attempt to determine how this problematic fared within the history of the discipline. Was Warburg’s Nachleben understood? By a few, certainly; but certainly not by the main stream, as a few examples will make clear.

When Julius von Schlosser published his History of the Wax Portrait in 1911, it became clear that the vocabulary of survival—borrowed from Tyler, but mainly from Warburg, who was a friend of Schlosser’s —had opened the only possible theoretical way of understanding the strangest phenomenon of wax sculpture, namely its persistence [longue durée], its resistance to the history of style, in other words, its capacity of surviving without significantly evolving. Schlosser understood that the history of images is not all a “natural history,” but rather an elaboration, a “methodological construction” (ein methodisches Präparat), and that it escapes from the laws of a trivial “evolutionism.” This is what justifies, at the end of the book, his critique and summary dismissal of “teleological pretensions” of the Vasarian sort.

Schlosser clearly left unexplored, undoubtedly more from modesty than from ignorance, a certain number of theoretical problems inherent in the model of survival. But a powerful idea was beginning to take shape. It is this: if art has a history, images, for their part, have survivals, which “declassifies” them, separating them from the usual domain of works of art. The price of this survival is the disdain in which they are held by a “high” history of artistic styles. That is why the History of the Wax Portrait has for many years been read more by anthropologists than by art historians.

With regard to models of time, Edgar Wind probably never risked making theoretical moves as radical and exploratory as those of Warburg and Schlosser. But he clearly understood that the word survival should be employed as more than a trivial “biological metaphor.” In 1934, he wrote that “when we speak of the ‘survival of the classics’, we mean that the symbols created by the Ancients have continued to exert their power over subsequent generations—but what do we mean by the word ‘continued’?” And Wind indicates that survival presupposes the harmonious working together of an entire ensemble of operations including forgetting, transformation of meaning, eliciting of memory, and unexpected rediscovery. This kind of complexity ought to remind us of the cultural, non-natural character of the temporality involved here. Wind is criticizing not only Wölfflin’s “immanent history” but also “historical continuity” in general, which is unaware of what is involved in all this kind of survival, the forces brought into play in every instance of survival: “pauses” and “crises,” “jumps” and “periodic reversions.” All of this forms a skein of memory (memory-mnemosyne), not a narrative history, resulting, therefore, not in a succession of artistic facts but in a theory of symbolic complexity.

One could not have a clearer statement of the critique of historicism contained in the very hypothesis of survival. Gertrude Bing rightly noted Warburg’s paradoxical situation regarding the epistemology of the historical sciences. (One could also, I believe, make an analogous observation concerning Michel Foucault.) On the one hand, he can be incomplete, biased, and even mistaken regarding certain historical facts; on the other hand, his hypothesis about memory—the specific type of memory presupposed by Nachleben—has profoundly altered our very understanding of what an historical phenomenon is. Significantly, Gertrude Bing stressed the way in which the notion of Nachleben transforms our whole conception of tradition: it is no longer a continuously flowing river in which things simply start from upstream and travel downstream, but a tense dialectical process, a drama played out between the river and its own eddies. Here we may again note that this way of conceiving historicity is not all that far from Walter Benjamin’s.


It must be said, however, that this approach found very few followers. Historians often prefer not to risk making a mistake: a fact, in their eyes, is worth more than a hypothesis, which is inherently uncertain. Let us call this scientific modesty—or perhaps cowardice, or even philosophical laziness. At worst, it is a positive hatred of all “theory.” Gombrich, in 1970, wanted to conclude his biography of Warburg by, as he termed it, “putting [the latter’s work] in perspective.” Here one detects a strange wish to “kill the father,” a definite desire to make sure that the ghost or revenant—as Warburg defined himself in 1924—no longer returns. And, with him, the “outmoded” hypothesis of survival will also cease for a time from its eternal return to the back of the art historian’s mind.

In order to arrive at this goal, two moves are necessary. The first consists in invalidating survival’s dialectical structure, that is to say, denying that a double rhythm, composed of survivals and renaissances, always organizes the temporality of images, rendering them hybrid and impure in the process. To this end, Gombrich does not hesitate to claim that Warburg’s Nachleben can be seen simply as the equivalent of what is called a revival. The second move consists in invalidating survival’s anachronistic structure, which is accomplished simply by returning to Springer and re-periodizing the distinction between survival and renaissance. In other words, the distinction is very simply reduced to a chronological one between the Middle Ages and the Renaissance. Gombrich thus ends up distinguishing the obscure “tenacity” of the medieval survivals and the inventive “flexibility” of the imitations all’antica that only a Renaissance worthy of that name could have produced, beginning only in the 15th century.

Sorting out the various transformations of the notion of survival would be a huge task, requiring anyone who undertook it to examine the whole history of the discipline since Warburg’s time. Let us, then, indicate only the most important landmarks. At the beginning of the 1920s, in the first volume of the Vorträge der Bibliothek Warburg, Adolph Goldschmidt published an article on “The Survival of ancient forms in the Middle Ages.” Aware right from the start of the paradox of the Nachleben, which is simultaneously an indicator of “continued life” (Weiterleben) and of “continued death” (Weitersterben), Goldschmidt attempted to extend into the Middle Ages what Warburg had observed in Botticelli, notably by pointing out the expressive role of drapery in Byzantine art. Twenty years later, Jean Seznec presenting the “survival of the ancient gods,” called the theme an argument that would be troubling for received views of chronology. In showing the interference between the Middle Ages and the Renaissance, he, too, demonstrated the broad scope of the field of survivals:

As the Middle Ages and the Renaissance come to better known, the traditional antithesis between them grows less marked. The medieval period appears “less dark and static,” and the Renaissance “less bright and less sudden.” Above all, it is now recognized that pagan antiquity, far from experiencing a “rebirth” in fifteenth-century Italy, had remained alive within the culture and art of the Middle Ages. Even the gods were not restored to life, for they had never disappeared from the memory or imagination of man.... The difference in styles acts as a further hindrance to our awareness of the continuity of tradition, for Italian art of the fifteenth and sixteenth centuries invests the ancient symbols with fresh beauty; but the debt of the Renaissance to the Middle Ages is set forth in texts. We shall attempt to show how the mythological heritage of antiquity was handed down from century to century, through which vicissitudes it passed, and the extent to which, toward the close of the Cinquecento, the great Italian treatises on the gods which were to nourish the humanities and art of all Europe were still indebted to medieval compilations and steeped in the influence of the Middle Ages.

This kind of respect for Warburg’s teachings and for the notion of the impurity of the temporality of images represents, it must be said, a minority position. Everywhere else one senses a desire to establish an ever clearer and more distinct periodization of the history of art, one more schematic and satisfying to the mind. In short, the procedure used to invalidate Warburg’s approach, so clearly expressed by Gombrich, was employed more surreptitiously in a whole series of theoretical moves by which the notion of Nachleben was reoriented towards various temporal schemas and deterministic models that it had the merit of challenging in the first place. Thus, survival was drawn towards the atemporal notion of the archetype, or toward the idea of eternal cycles; this was done in order to explain, but at little cost, the mixture of “continuities” and “variations” which inevitably stamps the history of images.

The notion of survival was also drawn in the more positivistic direction of the material remains of Antiquity, or of the more general question of sources. It was also drawn towards a more “formalist” point of view, that of influences. And it was used, as well, by scholars interested in iconographic traditions and, more generally, in those unexamined permanent elements which have characterized certain ancient artistic genres up to the modern period. Finally, all this has been looked at from the opposite direction by sociologically-informed theories of reception, or in terms of the “taste for the antique,” of imitation, or simply of “reference” to the “stylistic norms” of Antiquity. Considered outmoded, or else used as a passe-partout, but, in any case, stripped of all theoretical significance, Warburg’s Nachleben is thus no longer debated. That does not mean that it has been assimilated. Quite to the contrary. Let us say, rather, that it has been exorcized by the very discipline which is indebted to Warburg for the historical concept of the impurity of time but which has ended up reproaching him for it.


The high priest who exorcized our dibbuk is none other than Erwin Panofsky—but couldn’t we have expected it? Gombrich himself reluctantly admitted as much: it was primarily due to Panofsky that, for generations of art historians, Warburg’s work was “put in perspective” in such a way as to invalidate the Nachleben, this being the theoretical means by which the notion was exorcized. As early as 1921—just fifteen years after Warburg’s lecture on “Dürer and Italian Antiquity”—Panofsky published an article with a title too similar not be secretly rivaling the earlier publication: “Durer and Classical Antiquity.” In it, despite the requisite expressions of respect, the problematic of survival already has given way to a problematic of influence; and the question of the pathetic, which in Warburg’s work could be linked to Nietzsche’s Dionysian, has given way to a problematic of typification and of the “juste milieu,” supported by several references to Kant’s “ideal beauty” [beau idéal] and to classical rhetoric.

In the obituary Panofsky wrote in 1929, the crucial expression of Warburg’s Hauptproblem, the expression Nachleben der Antike, does not appear even once; instead of any mention of “survival,” the only issues we find discussed are the “heritage” of Antiquity (Erbteil des Altertums) and the “history of reception” of Antiquity (Rezeptionsgeschichte der Antike). Then, in 1933, joining his efforts to those of Fritz Saxl, who was already attempting to historicize Warburg’s conceptual schemas as much as possible —in itself a legitimate undertaking—Panofsky published a long article on “Classical Mythology in Medieval Art” in the Metropolitan Museum of Art of New York’s Bulletin. This was his first important publication in English —his entry visa into a new intellectual and institutional context that was going to transform his exile (his flight from Nazi Germany) into an empire (his undisputed domination of academic art history).

It is possible—and, up to a certain point, justified—to read this article as an extension of Warburg’s writings on the “survival of the ancient gods”; for Panofsky and Saxl appear to be satisfied to apply the notion of Nachleben to a chronological domain on which Warburg himself had not directly worked. At the start, therefore, they make a place for survival, a place showing that the Vasarian historical point of view is “wrong,” though only in part:

The earliest Italian writers about the history of art, such for instance as Ghiberti, Alberti, and especially Giorgio Vasari, thought that classical art was overthrown at the beginning of the Christian era and that it did not revive until, during the fourteenth and fifteenth centuries in Italy, it served as the foundation of what is usually called the Renaissance.... They were wrong in so far as the Renaissance was connected with the Middle Ages by innumerable links. Classical conceptions survived throughout the Middle Ages—literary, philosophical, scientific, and artistic—and they were especially strong after the time of Charlemagne, under whose reign there had been a deliberate classical revival in almost every cultural field. The early writers were right in so far as the artistic forms under which the classical conceptions persisted during the Middle Ages were utterly different from our present ideas of antiquity, which did not come into existence until the "Renaissance" in its true sense of the "rebirth" of antiquity as a well-defined historical phenomenon.

One already senses that this way of approaching the subject implies not only an extension but also a bifurcation, or possibly even a reversal of Warburg’s position, of which Panofsky and Saxl, however, claim to be “followers.” What, then, is extended here? The general idea of a polarization between survival and renaissance. What is reversed, or abandoned? The structural or synchronic aspect, the non-chronological aspect—in short, the anachronistic aspect of this double rhythm. Henceforth, things become more neatly separated in value and in time: they become hierarchized and periodized. Survival becomes a lower category of art history, making the Middle Ages into a period of artistic “conventions,” of “gradual degeneration” of the classical norms, and, finally, of the unfortunate “dissociation” of form and content: “the medieval mind [is] incapable of realizing … the unity of classical form and classical subject matter.”

The Renaissance, for its part, will become—or become again—that higher category of art history which makes the Quattrocento and the Cinquecento into summits of artistic activity, of archaeological authenticity, and, therefore, of stylistic purity. Reading Panofsky and Saxl, one would almost think that the Renaissance “in its true sense,” i.e., the Renaissance as a “well defined historical phenomenon,” was the only period ever to witness the birth of a genuine and “free” human being. Free, notably, from symbolic burdens and figurative conventions: “the reintegration of classical mythological subjects, realized in the Renaissance, was the motor as well as a characteristic of the general evolution that culminated in the rediscovery of man as a natural being stripped of his protecting cover of symbolism and conventionality.” Perhaps not all the tensions have been eliminated (and in this regard Panofsky and Saxl evoke the Counter-Reformation, that is to say, the end of the Renaissance). But it is only the “classical harmony” of the time of the Renaissance in its true sense that receives the accolade for surmounting the artistic and cultural crises that had characterized the periods of survivals, crises attested to, if only in a negative fashion, by what these periods lacked.

Only one conceptual difficulty remained to be resolved: the notion of a renaissance contrasts with that of a survival with regard to two aspects that are not easily coordinated. The hierarchical opposition does not automatically coincide with chronological succession. Panofsky found an effective solution by distinguishing two different conceptual orders within the word renaissance: a synchronic order, that he here calls “renovation,” and the “well defined historical phenomenon” that is the Renaissance. What has been called the Carolingian Renaissance is, for Panofsky, only a “renovation.” The only Renaissance, taking the word “in its true sense,” is the that of the 15th and 16th century. As for the notion of survival, it remains in the shadow of its relative indetermination.

Beginning in 1944 Panofsky used the term renascence to refer to what he formerly called a renovation. The system attained its final state in 1960 with Renaissance and Renascences in Western Art, a work that emerged from lectures he gave in 1952, and thus one benefiting from eight long years of reflection. Panofsky forcefully reiterated that the Carolingian “renovation” or renewal and, in general, all the “proto-humanist” moments experienced by the Middle Ages, are in no way “renaissances” in the strict sense; they are only renascences, partial moments of a “return to Antiquity.”

We can now see that in order to resolve the basic problem announced at the start—namely the relationship between continuity and change in history—Panofsky created a conceptual framework similar, in its ternary structure, to the famous “semiological” distinction between “primary subject,” “conventional subject,” and “intrinsic meaning” set forth in the introduction to his Studies in Iconology. According to Panofsky, therefore, the entire “theory of historical time” could be organized by a three-term hierarchy, with the Renaissance at the summit, its initial capital letter indicating both chronological centrality and atemporal dignity. A dignity that Panofsky highlighted by the use of virtually Hegelian expressions like “self-realization,” “becoming aware,” “becoming real,” and “total phenomenon.” For Panofsky, the Renaissance was the awakening of art to a consciousness of itself, that is to say, to its own history and to its own “realization” or ideal meaning, so that in the end, Vasari, who said the same thing, turned out to be right.

Anticipating this stage were the various partial “renewals” or renascences that, in the long course of the Middle Ages, stirred up the history of forms in those moments which experienced an awakening of classicism. Finally, there is the background of sleep from which all these movements arose. Panofsky hesitated to name it, to give it a theoretical status; he just barely managed to mention, in a one page digression, a “period of incubation.” But it is clear that what is involved here is none other than the Warburgian notion of survival. Significantly, the final sentences of Renaissance and Renascences oppose the “unredeemed phantom” of this survival to the soul of classicism all’antica, now finally resuscitated—a soul that is ideal, intangible, pure, immortal, and omnipresent:

The Middle Ages had left antiquity unburied and alternately galvanized and exorcised its corpse. The Renaissance stood weeping at its grave and tried to resurrect its soul. And in one fatally auspicious moment it succeeded. This is why the medieval concept of the Antique was so concrete and at the same time so incomplete and distorted; whereas the modern one, gradually developed during the last three or four hundred years, is comprehensive and consistent but, if I may say so, abstract. And this is why the medieval renascences were transitory; whereas the Renaissance was permanent. Resurrected souls are intangible but have the advantage of immortality and omnipresence.

In these sentences one seems to hear the echo of two symmetric exaltations—both of them idealist—the one stemming from Vasari, the other from Winckelmann. Death to errant phantoms and to survivors! Long live resuscitated and immortal souls! What all this expresses, of course, is simply an aesthetic choice. One might even say a phantasmal choice. In that respect, it is a legitimate one. But it appears here in a discourse purporting to present what is true and claiming to establish art history as an objective science. Its effect has been to orient the latter toward the study of “well-defined historical phenomena” rather than toward the uncertain time of the various survivals. It has preserved the immortal ideas and sent all the phantom images far away. In looking at the Renaissance, this approach wanted to see only a time without impurities, a period that could serve as a “standard,” in which the homogeneity, the “reintegration” of forms and contents, would be legible. It has, therefore, rejected Warburg’s fundamental intuition.

Veritas filia temporis [truth is the daughter of time], so the ancient adage tells us. But, for the historian, the question is how to know exactly of what time—or of what times, in the plural—truth is the “daughter.” As a student of Warburg’s, Panofsky began by recognizing the complexity and anachronism of the time involved in discussing images. Thus, in a text from his German period on “The Problem of Historical Time,” he used, and not by chance, a medieval example in order to introduce the theoretical difficulty inherent in any model of evolution that might be used in art history:

The sculptures at Rheims [in particular] engender ... an image of an unending, polychrome web, within which the most diverging threads become intertwined, running now beside each other and now in opposite directions. These individual stylistic directions (their marked differences in quality notwithstanding, which would seem to prohibit proposing a coherent, linear evolution) do not merely progress in parallel, indifferent to any interconnections; rather they penetrate one another and, not only that, they return again and again.... Thus this endless multiplicity of frames of reference, which seems to primarily constitute the world of the art historian, amounts to a confusing and unformalizable chaos.... Are we not then faced with a completely inhomogeneous contiguity of such frames of reference, which, to use Simmel's terminology, remain frozen in self-sufficient isolation and irrational specificity?

Panofsky indeed began—with Warburg—by recognizing the impurity of time. But he ended by extirpating it, dissolving it, subsuming it in an ordered framework that hearkened back to the aesthetic ambition of the golden ages (of which the Renaissance is one) and to the historical ambition of “reference periods.” Thus, his 1931 text concludes with the hope that a “chronology” of the sculptures of Rheims Cathedral might one day clarify and hierarchize their multiple stylistic reference systems. This expresses the desire of any idealist or positivist historian: that the times involved, once analyzed, become “pure” again. That survivals become logically eliminated from history the way the lee would be eliminated from a fine wine. But is that really possible? It is only ideal wines—wines without any taste—that can exist without any lee at all, without that impurity which, in a certain sense, gives them style and life.

<1>Historical Life: Forms, Forces and Time’s Unconscious

From Warburg to Panofsky, therefore, a word falls out of use and is forgotten: the word Nachleben, “survival.” And with it—with its fundamental impurity—went a second word contained within it: Leben, “life.” Panofsky, it is clear, sought to understand only the “meaning” of images, whereas Warburg also sought to understand their “life,” that impersonal “force” or “power” (Kraft, Macht) that he occasionally speaks of but regularly declines to define. Where did he get this vocabulary, which is so lacking in rigorous conceptual analysis and yet is so important? Above all from Burckhardt, about whom he liked to say—referring to the role of ephemeral spectacles in the visual culture of the Renaissance—that he attempted to find “a true transition from life into art” (ein wahrer Übergang aus dem Leben in die Kunst). Just as for Burckhardt, art, for Warburg, was not a simple question of taste, but rather a vital question. Similarly, history was not for him a simple chronological question, but rather a stirring up [remous] of the past, a debate in which “life” is at stake and which continues throughout the long span of a culture’s existence.

The history of images was thus for Warburg what it had already been for Burckhardt (but which it no longer has been since Panofsky): a question of “life” and—since in this “life” death is omnipresent—of “survivals.” The biomorphism expressed here has nothing in common with that of a Vasari, or even of a Winckelmann; for the “life” in question here does not exist without the element of the non-natural, which, in the view of Burckhardt and of Warburg, is required by the notion of culture. Nor does it exist without the element of impurity, which, again for each of them, is required by the very notion of historical time. Let us try to briefly characterize this enigmatic “life.” It seems to me that it can be understood as being, simultaneously, a play of functions (requiring an anthropological approach), a play of forms (requiring a morphological approach), and, finally, a play of forces (requiring a dynamic or energetic approach).

“Life” is a play of functions inasmuch as it the life of a culture. This did not escape Burckhardt’s first readers, who read his philosophical anthropology in the still vague terms of the “soul” or of culture understood as the “intimate state of the consciousness of a people.” Thus, in 1887, Emile Gebhart wrote that it was “to the Italian soul” that he posed the question of the secret of the Renaissance; and, for him, “all the great facts of this history: the politics, the erudition, the art, morality, pleasure, religion, [and] superstition manifest the action of certain forces vives.” We know that Burckhardt’s Kulturgeschichte has been looked at anew by social history, just as Warburg’s Kulturwissenschaft has been revisited by Panofskian iconology and the social history of art. Certain of Burckhardt’s ambiguities have been left aside in the process (and that is as it should be), but along with them, so have certain of his major theoretical hypotheses and certain of his most pertinent critical articulations. Let us mention several that Warburg was to incorporate, more or less explicitly, into his own thinking.

For Burckhardt, “life,” viewed as a play of functions is, in the first place, neither the life of facts, nor that of systems. One must speak of “life” and its concrete movement in culture because positivist history, in its rush to establish chronological facts, tends to blot out everything else, while idealist history—that of Hegel above all—tends to enlist everything in its effort to announce grand, overly abstract truths. In both cases, it is time itself that is disincarnated as a result of the desire to simplify, that is to say, to deny its complexity. Considering “life as culture,” in contrast, leads to a critical formulation designed to get beyond a dilemma that is really only schematic, and thus trivial, namely history-as-nature versus history-as-idea:

Yet history is not the same thing as nature (Die Geschichte ist aber etwas anderes als die Natur), and it creates, brings to birth and abandons to decay in a different way.... By a primordial instinct, nature creates in consistently organic fashion with an infinite variety of species and a great similarity of individuals. In history, the variety (within the one species homo, of course) is far from being so great. There are no clear lines of demarcation, but individuals feel the incentive inequality—inciting to development. While nature works on a few primeval models (vertebrates and invertebrates, phanerogams and cryptogams), in the people, the body social is not so much a type as a gradual product.... We shall, further, make no attempt at system (wir verzichten ferner auf alles Systematische), nor lay any claim to “historical principles.” On the contrary, we shall confine ourselves to observation, taking transverse sections of history in as many directions as possible. Above all, we have nothing to do with the philosophy of history.... Hegel.... speaks of “the purpose of eternal wisdom,” and calls his study a theodicy by virtue of its recognition of the affirmative in which the negative (in popular parlance, evil) vanishes, subjected and overcome….We are not, however, privy to the purposes of eternal wisdom: they are beyond our ken. This bold assumption of a world plan leads to fallacies because it starts out from false premises.

One could say that with this twofold refusal Burckhardt inaugurated a new manner of writing history, a “third way.” And Warburg later adopted the basic choices Burckhardt made: to be a philologist who goes beyond the facts (for the facts are important primarily for the basic questions they give rise to), and to be a philosopher who goes beyond the systems (for the basic questions are important primarily for the singular ways they are actually employed in history). Such, then, is what the “third way” demands: the refusal to accept either teleology or absolute pessimism, and the recognition of the historical “existence” (Dasein, Leben) of every culture, that it to say, of its complexity. Burckhardt went as far as to assert that authentic history is distorted as much by the “ideas” deriving from “preconceived theories” as by “chronology” itself. For history, he thought, was that aspect of our intellectual effort which rescues us from our basic incapacity “to understand what is varied, accidental” (unsere Unfähigkeit des Verständnisses für das Bunte, Zufällige) [Translation modified—Trans.].

With this approach Burckhardt thus established a strange dialectic of times, one which needed neither “good” nor “evil,” neither “beginnings” (origins or sources from which everything supposedly derived) nor “ends” (a direction towards which all history is heading). It needed none of all that to express the complexity –the impurity—of its “life.” It is composed of rhizomes, of repetitions, of symptoms. Local history—along with patriotic or racial history—is not its concern; for such history lacks a way of conceiving relationships and differences. Neither is universal history its subject; for Burckhardt renounced in advance any attempt to look for a general formula for the “system” of all these rhizomes.

The philosophers, encumbered with speculations on origins, ought by rights to speak of the future. We can dispense with theories of origins, and no one can expect from us a theory of the end.... Questions such as the influence of soil and climate...are introductory questions for the philosophers of history, but not for us, and hence quite outside our scope. The same holds good for all cosmologies, theories of race, the geography of the ancient continents and so on. The study of any other branch of knowledge may begin with origins, but not that of history. After all, our historical pictures are, for the most part, pure constructions, as we shall see more particularly when we come to speak of the State. Indeed, they are mere reflections of ourselves. There is little value in conclusions drawn from people to people or from race to race. The origins we imagine we can demonstrate are in any case quite late stages.... Its greater intelligibility is merely apparent, and arises in part from an optical illusion, namely our own much livelier readiness to understand, which may go hand in hand with great blindness.

In reflecting on the relationships between the local and the global, Burckhardt did not fail to reflect, as well, on the relationships between change and stability. For him, the “life” of history is not only a spatial play of individual and contextual events; it is also, of course, a play of time, the dialectic of what changes and of what resists change. To be an historian, for Burckhardt, does not mean just composing a narrative of things that change and succeed each other; it is necessary, above all, to “deal first with their continuous and gradual interaction and in particular with the influence of the one variable (Bewegtes), culture, on the two constants (Stabiles)” [Burckhardt is referring here to the state and to religion—Trans.]. In this regard, the “life” of history falls within the domain of morphology: it is a play of forms, if one understands by “forms” the tangible crystallization of such a dialectic or “reciprocal influence.”

Since ... time bear[s] away ceaselessly the forms (die Formen) which are the vesture of material as of spiritual life (das geistige Leben), the task of history as a whole is to show its twin aspects, distinct yet identical, proceeding from the fact that, firstly, the spiritual, in whatever domain it is perceived, has a historical aspect (eine geschichtliche Seite) under which it appears as change, as the contingent, as a passing moment which forms part of a vast whole beyond our power to divine, and that, secondly, every event has a spiritual aspect (eine geistige Seite) by which it partakes of immortality. For the spirit knows change, but not mortality.


In considering the word spirit, it was to the domain of culture that Burckhardt directed his attention, and he did so as an historian and an anthropologist, not as a philosopher. Thus, even before Warburg claimed the status of “psycho-historian,” Burckhardt had already thought of Kulturgeschichte in terms of a morphology, or even an aesthetic of the “psychic forms” of culture. He recognized that this issue was central to any historical project, but not if it was conceived in the “romantic-fantastic” mode (nicht etwa romantisch-phantastisch). It should be treated in the manner in which one would observe the “marvelous process of the metamorphosis of a chrysalis” (als einen wundersamen Prozess von Verpuppungen). That is why Burckhardt was able to cover his notebooks with all those visual notations (Fig. 8). The culture of an epoch, he held, could be detected in its written sources and in the events of its history, but equally well in its paintings, in its architectural ornaments, in the details of its clothing, in the landscapes that its people refashioned, in its heraldic imagination, and in its most marginal figures, in grotesques, for example.

<insert fig. 8 about here>

It has wrongly been said that Burckhardt’s aesthetization of history was due to epistemological weakness, to an art lover’s failing, or to a disciplinary indiscretion of an historian stricto sensu. Burckhardt, however, did not aestheticize history in the way one let’s oneself be drawn into intoxication in order to forget something. He simply recognized—in itself an important finding—that the temporal hinge between change and stability, between Geschichte and Typus, is a formal hinge, which involves the workings of something like the “process of the metamorphosis of a chrysalis.” It is thus necessary to “aestheticize” history; for Kultur, according to Burckhardt, assumes the place, in a certain way, of Hegel’s “reason in history.” No history is possible without a history of culture, and there can be no history of culture without an art history open to the anthropological and morphological resonances of images. This is a task that Burckhardt, of course, left in its initial stages—a task that Warburg and Wölfflin, each in his own fashion, sought to take up, even if they could not complete it.

Burckhardt considered the establishment of such a morphology to lie at the heart of the historian’s task—and someday one should critically examine the history of this morphological theme, from Goethe to, say, Carlo Ginzburg. This also explains the pronounced visual tenor of his theoretical vocabulary, which displays a violent refusal of, a shrinking back from Kant’s a priori and Hegel’s “speculation,” and, symmetrically, an insistence on the historian’s right to “look” and to employ “contemplation” (Anschauung) and even “imagination” (Phantasie). History, for Burckhardt, is constructed less like a story and more like a “picture” (Bild): “Pictures, paintings, that’s what I want” (Bilder, Tableaux, das ist’s was ich möchte), he wrote as early as 1844—a formulation that Warburg made his own even before putting it into practice with the collection of plates that forms his atlas Mnemosyne Atlas. How can one fail to see here, among other possible examples, that the very choice of grisaille as a color expresses a form of time in which the present time of a given historical moment (that of Mantegna, for example) asserts its own archaeological distance, its own anachronism, its own task of making possible the survival—like phantoms—of the figures of Antiquity.

Thus, no history is possible without a morphology of the “forms of time.” But the reasoning involved here would be incomplete without an essential clarification: there can be no morphology, or analysis, of forms without a dynamic, or analysis, of forces. To omit that is to reduce morphology—and this is often what happens—to the establishment of sterile typologies. It amounts to assuming that the forms are the reflections of a time, whereas they are really the casualties or fallen elements [les chutes], whether ridiculous or sublime, of a conflict taking place within time—that is to say, of a play of forces. This, then, is the third characteristic of “life,” according to Burckhardt. The dynamic of the “type” (Typus) and “development” (Entwicklung) constitutes the “main problem” (Hauptproblem) of history. The phenomenon in question is a tense and oscillatory one, and it generates formidable complexities:

What issues from this main phenomenon (die Wirkung des Hauptphänomenes) is historical life (das geschichtliche Leben), rolling on in a thousand forms, complex, in all manner of disguises, bound and free, speaking now through the masses, now through individuals, now in hopeful, now in hopeless mood, setting up and destroying states, religions, civilizations, now a dark enigma to itself, moved by inchoate feelings born of imagination rather than thought, now companioned only by thought, or again filled with isolated premonitions of what is fulfilled long afterwards. 6

To speak of “historical life” (geschichtliches Leben) is, therefore, to seek to understand time as a play of “forces” (Kräfte, Mächte) or of “powers” (Potenzen), out of which, Burckhardt states, “all kinds of forms of life” (Lebensformen) arise. Elsewhere he writes: “it is our task simply to observe and describe objectively the various forces (Potenzen) as they appeared side by side or one after another. But the task is very difficult, because a power [puissance] always tends to evade our notice: it is difficult to observe when it is too violent and omnipresent, and difficult to observe when it is too virtual (a “potential” force [en puissance]) and invisible. This double meaning of the word power—manifest force and latent force—is not at all simply an anecdotal matter; it gives rise to at least two important consequences, two bifurcations that profoundly alter our way of conceiving historicity.

The first yields a dialectic of time—the very one we are trying to grasp in the notion of the symptom. In reading Burckhardt we find that this dialectic functions in the manner of a continually renewed debate between “latencies” (Latenzen) and “crises” (Krisen). There is no historical time, in fact, without some play of latencies: “[We are] ignorant ... of everything which we call latent forces (latente Kräfte), physical or mental, and [of] the incalcuable factor of mental contagions, which can suddenly transform the world.” This historical and collective condition has its psychological and individual counterpart in the circumstance that “in man, no one side is ever active to the exclusion of the rest; the whole is always at work, even though some elements may function in a weaker, unconscious (im Unbewussten) fashion.”

Now, every latency seeks to work its way towards the surface of events. In Burckhardt, the term “crisis” (Krisis) designates that particularly effective way that time has of making its own power spring forth—through a contretemps or through a symptom. At least two chapters of the Reflections on History are entirely devoted to this question. And every other part of the book is concerned in some fashion with the observation of the dialectical relationship, which is such a difficult one to analyze, between the fixed forms and the forces which cause them to vacillate, or between the dominant forces and the forms which cause them to fail: “in history, the way of annihilation is invariably prepared by inward degeneration, by decrease of life. Only then can a shock from outside put an end to the whole.... The crisis which has one specific cause is borne along on the storm-wind of many other things, yet not a man involved in it but is absolutely blind as to the force which will finally win the day.”


We see that for Burckhardt the practice of history amounted to the analysis not of facts succeeding each other over time but, rather, of something like an unconscious of time, with all its latencies and its catastrophes. Warburg developed his history of images, it seems to me, in accord with the consequences of this methodological decision. History, then, is to be a symptomatology or even a pathology of time, which it would be wrong, however, to reduce to a simple moral pessimism, even though an element of tragedy is everywhere visible in it. It is first of all in morphological and dynamic terms that Burckhardt wanted to speak of the “catastrophes,” indeed of the “illnesses” of time:

[The historian must analyze each force,] State, Religion and Culture, dealing first with their continuous and gradual interaction and in particular with the influence of the one variable, Culture, on the two constants. We shall then discuss the accelerated movements of the whole process of history, the theory of crises and revolutions, as also of the occasional abrupt absorption of all other movements, the general ferment of all the rest of life, the ruptures and reactions—in short, everything that might be called the theory of storms (Sturmlehre).... We, however, shall start out from the one point accessible to us, the one eternal center of things—man, suffering, striving, doing, as he is and was and ever shall be. Hence our study will, in a certain sense, be pathological (pathologisch) in kind” [Translation modified by the author, and, accordingly by the translator—Trans.]

Must one still speak of a dialectic of time? Yes, if by this term one understands a process that is filled with tensions rather than resolutions, one that is obsidian and sedimentary rather than linear and oriented in a certain direction. The dialectic of the “stable powers” (Stabiles) and of the “mobile element” (Bewegtes) produced a far-reaching critique of historicism, one that only complexifies, multiplies, and even disorients the models of time that Burckhardt in this passage calls “crises,” “revolutions,” “ruptures,” “reactions,” “occasional absorptions,” “ferment,” “perturbations”—a list that could go on indefinitely. To speak of an “unconscious” (Unbewusstes) or of a pathology, is to affirm, moreover, that the dialectic at work here demonstrates only the impurity and anachronism of time. This, then, may be considered the second lesson, the second consequence of a morphological and dynamic approach to history: time liberates symptoms, and with them it causes the phantoms to act. Time, for Burckhardt, is already a time of obsessions, of hybridization, of anachronism; in this respect, it directly anticipates Warburg’s notion of “survivals.”

Thus Burckhardt speaks of Western culture as an unlimited sphere of influence, “impregnated with the traditions of all times, of all peoples, and of all civilizations.” He also states that “there are no clear limits” to be found within it, and that the “body social” of any culture is nothing but a perpetual “gradual product,” a “process” marked by “the effect of the contrasts and affinities.” The conclusion being that: “in history, everything is fully bastardized (Bastardtum), as if [that was] an essential element of fecundation (Befruchtung) of great spiritual events” [English translation modified—Trans.].

Now, this impurity is not only synchronic: it affects time itself, its rhythm and its development. One must not, Burckhardt asserts, rely on the use of periods to separate history into “ages of the world”; rather, one should note the existence of “countless incarnations,” which presuppose “mutations” and, therefore, “human inadequacy.” The whole, then, is a difficult-to-analyze mixture of “destructions” and of something that must be called survivals. It is here that Burckhardt comes closest to the notion of Nachleben, in rejecting any attempt to set up a hierarchical periodization of history that would separate barbarism from civilization—just as later Warburg would refuse to sharply separate the Middle Ages from the Renaissance.

We can no more begin our presentation of history with the earliest state formations than with the transition from barbarism to civilization. Here, also, the concepts are much too vague….In the final analysis, the use or non-use of this word becomes a matter of temperament. I consider it barbarism to keep birds in cages. First one ought to eliminate those elements which have lived on from the infant days of mankind in petrified form in the most advanced civilization, perhaps for sacral or political reasons, such as individual human sacrifice.... Countless elements also subsist in the unconscious (lebt auch unbewusst weiter) as an acquisition bequeathed to mankind perhaps by some forgotten people. An unconscious accumulation of vestiges of culture (unbewusstes Aufsummieren von Kulturresultaten) in peoples and individuals should always be taken into account. This growth and decay (Wachsen und Vergehen) follows higher, inscrutable laws of life (höhere, unergründliche Lebensgesetze).

On the same page, Burckhardt uses the word Weiterleben, which means “subsistence” and, already here, “survival.” The way was open for understanding what Nachleben means. And with this “survival,” the way was also open for understanding time as that impure game, full of tensions, that debate between latent powers and violently acting ones we might call, with Warburg, the “life” (Leben) of images.