CHAPTER 2

The Exercise of Memory: Uses and Abuses

READING GUIDELINES

The cognitive approach to memory developed in the preceding chapter from the “objective” angle does not exhaust the description of memory. A pragmatic approach is also required. This new consideration is joined to the earlier one in the following manner: remembering is not only welcoming, receiving an image of the past, it is also searching for it, “doing” something. The verb “to remember” stands in for the substantive “memory.” What the verb designates is the fact that memory is “exercised.” Now the notion of exercise, applied to memory, is no less ancient than that of eikōn, of representation. Joined to that of “searching” (zētēsis), it shines in the firmament of Socratic concepts. Following Socrates, Plato does not hesitate to shift his discourse on the eikōn to the arena of “imitative techniques” and to distinguish a “fantastic” mimetics, fated to be deceitful, and an “iconic” mimetics, reputed to be “upright” (orthos) and “truthful” (alēthinos). Aristotle, in his turn, in the chapter “Anamnēsis” in his short treatise bearing a double title, describes recollection as a “search,” whereas mnēmē was characterized in the first chapter as an “affection” (pathos). Both of our Greek masters thus anticipated what will be called the effort of memory by Bergson and the work of remembering by Freud, as we shall soon see.

The remarkable fact is that these cognitive and practical approaches overlap in the operation of recollection; recognition, which crowns the successful search, designates the cognitive side of the recollection, while effort and work are inscribed in the practical field. Henceforth, we will reserve the term “remembering” (remémoration) to signify this superposition of the two problematics—cognitive and pragmatic—in one and the same operation of anamnēsis, of recollection, of recall.

This split into cognitive and pragmatic dimensions accentuates the specificity of memory among the phenomena classified under the heading of the psychical. In this regard, the act of exercising memory comes to be inscribed within the list of powers, capacities, belonging to the category “I can,” to take up the expression dear to Merleau-Ponty.¹ But it seems that this act of exercising memory itself presents the most complete superimposition in a single act, with respect to its description, of the cognitive aim and the practical operation—the act of remembering, direct heir to the Aristotelian anamnēsis and indirect heir to Platonic anamnēsis.

This originality of the mnemonic phenomenon is of considerable importance for all that follows in our investigations. Indeed, it also defines the historiographical operation as a theoretical practice. The historian undertakes to “do history” (faire de l’histoire) just as each of us attempts to “remember” (faire mémoire). For the most part, the confrontation between memory and history will play itself out on the level of these two, inseparably cognitive and practical, operations.

The ultimate stakes of the investigation that follows concern the fate of the desire for faithfulness that we have seen linked to the intention of memory as the guardian of the depth of time and of temporal distance. In what way, with respect to these stakes, are the vicissitudes of the exercise of memory likely to affect memory’s ambition to be truthful? In a word, the exercise of memory is its use; yet use includes the possibility of abuse. Between use and abuse slips the specter of the bad “mimetics.” It is from the angle of abuse that memory’s aim of truthfulness is seriously threatened.

The pages that follow aim at sketching out a loosely knit typology of these abuses of memory. They are in each case correlated with an aspect of the exercise of memory.

The feats of ars memoriae, the art celebrated by Frances Yates, will be set out separately;² the excesses it has occasioned are those of an artificial memory that methodically exploits the resources of the operation of memorization, which we want carefully to distinguish, already on the level of natural memory, from remembering in the limited sense of the evocation of singular facts, of events. The longest section of this chapter will, thus, be devoted to the abuses of natural memory. These will be divided into three levels: on the pathological, therapeutic level, the disturbances of blocked memory will emerge; on the properly practical level, those of manipulated memory; and on the ethico-political level, those of a memory abusively summoned, when commemoration rhymes with rememoration. These multiple forms of abuse expose the fundamental vulnerability of memory, which results from the relation between the absence of the thing remembered and its presence in the mode of representation. The highly problematical character of this representative relation to the past is laid bare in its essence by all the abuses of memory.

§

THE ABUSES OF ARTIFICIAL MEMORY: THE FEATS OF MEMORIZATION

There is one modality of the act of exercising memory that presents itself as practice par excellence, namely, the memorization that has to be rigorously distinguished from remembering.

With remembering, the emphasis is placed on the return to awakened consciousness of an event recognized as having occurred before the moment when consciousness declares having experienced, perceived, learned it. The temporal mark of the before thus constitutes the distinctive feature of remembering, under the double form of simple evocation and of the recognition that concludes the process of recall. Memorization, on the other hand, consists in the ways of learning relating to forms of knowledge, know-how, capacities marked from a phenomenological point of view by a feeling of facility, ease, spontaneity, in such a way that these are fixed and remain available for activation. This feature constitutes the pragmatic counterpoint to the recognition in which recall terminates on the epistemological plane. In negative terms, this is an economy of effort, as the subject is dispensed from learning all over again in order to perform a task appropriate to specific circumstances. The feeling of ease then represents the positive side of this successful actualization of a memory, which Bergson would say is “acted” rather than “represented.” In this regard, memorization can be held to be a form of habit-memory. But the process of memorization is specified by the methodical character of the ways of learning aiming at an easy actualization, the privileged form of happy memory.

It thus becomes a legitimate project to describe the methods of learning directed to this easy actualization from the perspective of the techniques of acquisition and to attempt to spot the weaknesses that allow abuse to infect use. We will follow an order of increasing complexity in which the opportunities for misuse grow along with the ambition of mastery exerted over the entire process of memorization. For it is indeed in this ambition of mastery that the possibility to slip from use to abuse resides.

At the lowest level, we find the techniques belonging to what is called learning (apprentissage) in experimental psychology. I am speaking in general and broad terms of “ways of learning” in order carefully to mark out the field. Learning, openly associated with memory in specialized works, belongs to a biology of memory.³ This learning, in fact, consists in the acquisition by a living being of new behaviors that are not part of inherited, genetically programmed repertoires of abilities or know-how, nor do they stem from cortical epigenesis. What is important for our investigation is that the control over the learning process belongs to the experimenter who directs the manipulation. He or she determines the task, defines the criteria of success, organizes punishments and rewards, and, in this way, “conditions” the learning. This situation constitutes the form most in opposition to that of the ars memoriae, which we will discover at the end of our survey and which will be the fruit of discipline, of an “ascetic”—the askēsis of the Socratics, meaning “exercise”—of which the apprentice will himself be the master. Speaking of manipulation, we are certainly not denouncing an abuse but only intend to characterize the type of mastery that presides over experimentation. Only manipulation in the human milieu, as will be discussed below in connection with ideology, will deserve to be marked with the seal of infamy. Nevertheless, already at this level, and without leaving the psycho-biological plane on which these experiments are conducted, one can level an appropriate critique at the conditions controlling the manipulation of the living beings undergoing these tests. During the period of behaviorism, such tests were supposed to provide an experimental basis for verification using “models” based on stimulus-response type hypotheses. Criticisms raised by authors such as Kurt Goldstein, echoed by Merleau-Ponty in The Structure of Behaviour⁴ and Canguilhem in La Connaissance de la vie,⁵ relate essentially to the artificial character of the situations in which an animal, even a human subject, is placed under the control of the experimenter, in contrast to the spontaneous relations of the living being with its environment, as these are apprehended by ethological science in an open setting. The conditions of experimentation are not neutral with respect to the meaning of observed behavior. They contribute to masking the living being’s resources of exploration, anticipation, and negotiation, through which this being is engaged in an Umwelt that belongs to it in its own right and that it helps to construct.

This discussion matters to us inasmuch as the forms of learning that we are now going to consider have the capacity, in their turn, to oscillate between manipulation, the mastery of the teacher, and the discipline expected of the disciple.

And so, it is to the dialectic of the teacher and the disciple that the exercises of memorization as part of a program of education, of paideia, belong. The classical model is well known: it consists in the recitation of the lesson learned by heart. Augustine, the rhetorician, likes to derive his analysis of the threefold present—present of the past or memory, present of the future or expectation, present of the present or intuition—from an examination of the act of reciting a poem or a biblical verse. Reciting from memory, we say, without hesitation or mistake, constitutes a small feat that prefigures larger ones, as we will say later. Now, before unleashing our critique of the abuses of learning “by heart,” we must first recall the justification for its beneficial use. Within the framework of teaching, which, as we will soon see, is only one part of paideia, recitation has long constituted the preferred mode of transmission, under the direction of educators, of texts considered, if not as founding works of the culture of instruction, at least as prestigious, in the sense of texts that are authoritative. For it is indeed authority that is at issue in the final analysis, more precisely enunciative authority, to distinguish it from institutional authority.⁶ Here, we reach a political concept in the most fundamental sense, concerning the establishment of the social bond. We can hardly conceive of a society in which the horizontal bond of living together would not intersect with the vertical bond of the authority of the Ancients, following an old adage cited by Hannah Arendt: potestas in populo, auctoritas in senatu. The eminently political question is knowing what the “Senate” is, who the “Ancients” are, and where their authority comes from. Education takes place short of this problem and as though sheltered from the questioning of its own legitimacy. Whatever the effect of this enigma of authority—the heart of what Rousseau called the “labyrinth of politics”—every society has the burden of transmitting from one generation to the next what it holds to be its cultural acquisitions. For each generation, the learning process, as we suggested above, can dispense with the exhausting effort to reacquire everything each time all over again. In this way, people in Christian communities have long learned to recite the catechism. But it is also in this way that the rules of correct spelling were learned—Oh, those dictations!—then the rules of grammar and of mathematics. And it is yet again in the same manner that we learn the rudiments of a dead language or a foreign language—Oh, those Greek and Latin declensions and conjugations! As young children, we learned rhymes and rounds, then fables and poems; in this regard, have we not gone too far in the war against learning “by heart?” Happy, indeed, are those who, like Jorge Semprun, can whisper in the ear of a dying man—Maurice Halbwachs, helas!—Baudelaire’s verses: “Ô mort, vieux capitaine, il est temps, levons l’ancre . . . nos coeurs que tu connais sont remplis de rayons . . .” But learning “by heart” is not the perquisite only of the schooling of the past. Many professionals—doctors, lawyers, scientists, engineers, teachers—have recourse over the entire course of their lives to the copious memorization of information resting on repertories, lists of items, protocols that are kept available for actualization at the appropriate time. All must have at their disposal a well-trained memory.

This is not all: neither the pedagogical nor the professional use of memorization exhausts the treasury of the ways of learning countenanced by a faultless recitation without hesitation. We must mention in this connection all the arts that Henri Gouhier places under the generic title of the arts in two measures—dance, theater, music—in which the execution is distinct from the composition of the work, found in a libretto, a score, or some other form of inscription.⁷ These arts require of their practitioners a laborious training of the memory, based upon a stubborn and patient repetition, until an execution, at once faithful and innovative, is obtained, one in which the prior labor is forgotten under the appearance of a happy improvisation. How can one fail to admire the dancers, actors, musicians who have recorded extensive repertories, which they “execute” for our pleasure. They are the true athletes of memory. Perhaps, in this, they represent the only indisputable witnesses to a use without abuse of memory, obedience to the injunctions of the work inspiring in them the humility capable of tempering their legitimate pride in the exploit accomplished.

At the third stage of our journey through the forms of learning, I would like to mention the long-standing tradition that raised memorization to the level of an ars memoriae, worthy indeed of the name of art, of technique. Under the title The Art of Memory, Frances A. Yates has devoted a work to this tradition, one that remains a classic on this subject.⁸ The Latin name is not a mere convention: originally, it refers to the mnemotechnic procedures recommended and practiced by the Latin rhetoricians—the unknown author of the Ad Herennium (mistakenly attributed to Cicero by the Medieval tradition), Cicero himself—frequently called Tullius—and Quintilian. The founding myth, however, is not Roman but Greek. It refers to a famous episode, situated around the year 500 B.C. at the fatal end of a banquet given by a rich benefactor in honor of a well-known athlete. The poet, Simonides of Ceos—who, moreover, is mentioned favorably by Plato—had been hired to give a panegyric in honor of the victorious athlete. Opportunely summoned outside the banquet hall to meet two benevolent demi-gods, Castor and Pollux, he escapes the catastrophe that befalls the athlete and the other guests, who are buried under the ruins of the collapsed roof. This happy fate suffices for the Greek myth in which the poet is held to be blessed by the gods. But the Latins are acquainted with a sequel better suited to their culture of eloquence. The poet is supposed to have been able to indicate from memory the place of each guest and, so, according to Weinrich, “to identify the dead according to their location in space.” A fabulous victory over forgetfulness—the catastrophe symbolized by sudden death—is the meaning of the exploit. But this is at the price of a hard schooling that annexes the art of memory to rhetoric. This art consists essentially in associating images with places (topoi, loci) organized in rigorous systems corresponding to a house, a public place, an architectural setting. The rules of this art are of two sorts: the first govern the selection of the places, the second govern the mental images of the things one wishes to remember and which the art assigns to the places selected. The images stored in this way are supposed to be easy to recall at the appropriate moment, the order of the places preserving the order of the things. From the treatise Ad Herennium—the earlier Greek treatises having been lost—comes the lapidary definition that will be repeated from age to age: “The artificial (artificiosa) memory is established from places and images.” As for the “things” that are depicted by the images and the places, these are objects, persons, events, facts relating to a cause to be argued. What matters is that the ideas be attached to images and that the items be stored in places. We thus meet up again here with the old metaphor of inscription, with places now in the role of the wax tablet and images in that of the letters inscribed on it. And, from beneath this metaphor, reemerges the genuinely founding metaphor, coming from the Theaetetus, of the wax, the seal, and the imprint. But the novelty consists in the fact that the body—eventually, the brain—or the soul joined to the body is no longer the support for this imprint, but rather the imagination considered as a spiritual power. The mnemotechnics applied to it are all to the glory of the imagination, of which memory becomes an annex. In the same stroke, spatialization obliterates temporalization. Not the spatiality of the lived body and its environing world, but that of the mind. The notion of place has chased away the mark of the past which had characterized memory since Aristotle’s De memoria et reminiscentia. Memory no longer consists in recalling the past but in actualizing what has been learned and stored in a mental space. In Bergsonian terms, we have crossed over to the side of habit-memory. But this habit-memory is a memory that, according to certain texts, is exercised, cultivated, trained, sculpted. These are genuine exploits performed by the fabulous memory of true athletes of memorization. Cicero calls such performances “almost divine.”

The tradition that stems from this “oratorical institution,” to borrow the title of Quintilian’s treatise, is so rich that our contemporary discussion concerning the places of memory—real places inscribed in geography—can be considered the most recent heir to this art of artificial memory coming from the Greeks and the Latins, for whom the places were the sites of a mental script. If, before Ad Herennium, the tradition was most certainly a long and varied one, leading back not only to the Theaetetus and its apologue of the seal in the wax, but also to the Phaedrus and its famous condemnation of memory bound by external “marks,” how much richer it has proven from “Tullius” to Giordano Bruno, in whom Frances Yates sees the culmination of the ars memoriae! What a journey from one end to the other and what twists and turns! At least three turning points have punctuated this strange epic of memorizing memory.

First comes the reinscription by Augustine of Latin rhetoric in a decidedly Platonic interpretation of a memory more closely related to the essential than to the order of events. From the beginning of the present work, we have alluded to the De memoria of book 10 of the Confessions: in addition to the famous exordium on the “palaces” and “storehouses” of memory, we find there the apologue of the stamp in the wax, extended by the theme of the “effigies.” What is more, the act of reciting is taken as the basis for the analysis of recall. We retain in particular, though, the exclamation: “How great is the power of memory!” For it is indeed the power exerted in the act of remembering that is at issue in the entire tradition of the ars memoriae. But Augustine still fears forgetfulness, which will, clearly, be forgotten at the apogee of the ars memoriae.

With the second turning point, the ars memoriae undergoes a thorough moralization on the part of the Medieval Scholastics; this happens through a surprising union of the already moralized rhetoric of Cicero—“Tullius”⁹—and the Aristotelian psychology of De anima and De memoria et reminiscentia.¹⁰ The latter text, in particular, considered an appendix to De anima, was highly esteemed by medieval thinkers; Saint Thomas wrote a detailed commentary on it. Memory, then, is found inscribed on several lists: it is one of the five parts of rhetoric, alongside intelligensia and providentia, where rhetoric itself is one of the seven liberal arts (grammar, rhetoric, dialectic, arithmetic, geometry, music, astronomy); but memory is also one part of the virtue of prudence, which figures among the major virtues, along with courage, justice, and temperance. Framed in all these different ways, and as such submitted to a second-order memorization, the medieval memory is the object of praise and of particular attentions, as we would expect from a culture that knows writing, of course, but not printing, and that, in addition, carried enunciative and scriptural authority to its pinnacle: Greek and Latin masters figure as auctoritates, alongside Holy Scripture, the texts of the councils, and the works of the doctors of the church. At the dawn of the Middle Ages, Alcuin, whom Charlemagne entrusted with restoring the educational system of antiquity in the Carolingian empire, declared to his emperor that memory is the “treasure-house of all things”; all things: articles of faith, paths of virtue leading to heaven, paths of iniquity leading to hell. Through memorization, on the basis of “memory notes,” are inculcated all systems of knowledge, know-how, belief, ways of living that mark the progress toward beatitude. The Secunda Secundae of Saint Thomas’s Summa Theologica constitutes, in this respect, the major document of this instruction of reason and faith for which the ars memoriae became the repository and the organon. Along with reason and faith, devotion receives its share with eloquent images of Hell, Purgatory, and Paradise, themselves considered as the places in which vices and virtues are inscribed, memory places, in the strong sense of the word. It is then not surprising that this path of memorization leads far beyond the feats of individual memory to Dante’s Divine Comedy. The places visited under Virgil’s and then Beatrice’s guidance form so many way-stations for a meditating memory, which unites the recollection of exemplary figures, the memorization of the major teachings of the tradition, and the commemoration of the founding events of Christian culture.¹¹ Compared to this superb metaphor of spiritual places, the exploits of artificial memory prove to be paltry indeed. In fact, a poetic memory was required to transcend the opposition between natural memory and artificial memory, to grind to dust the opposition between use and abuse.¹² This will no longer be the case at the conclusion of the third turning point.

The third turning point affecting the fate of artificial memory is marked by the union of mnemotechnics and hermetic secrets. Giordano Bruno, toward whom all of Frances Yates’s analyses converge, is the emblematic figure of this new and almost final phase of the incredible progress of the ars memoriae. The art in question has become a magic, an occult art. Presiding over this metamorphosis is the conception of a system of correspondences between the stars and the lower world, presented as a revelation, as a secret that has been pierced. The art consists in placing on concentric circles of a “wheel”—the “wheel of memory”—following the principle of a one-to-one correspondence, the position of the stars, the table of virtues, the collection of expressive images of life, lists of concepts, the series of heroic or saintly human figures, all the conceivable archetypal images, in short, everything that can be enumerated and put into systematic order. What is entrusted to memory in this way is a divine power, conferring the absolute mastery of a combinatory art that links the astral order to the earthly one. It is still a matter of “placing” images onto places, but these places are stars and these images, “shadows” (the first book on memory published by Bruno in 1582 was called De umbris idearum), in which the objects and events of the lower world consist. This true “alchemy of the imagination,” as Frances Yates calls it (224), presides over a magical mnemotechnics which gives limitless power to the one who possesses it. The revenge of Platonic, and especially Neoplatonic, reminiscence over the Aristotelian psychology of memory and recollection is complete, but at the price of the transformation of reasoned speculation into mystagogy. Yes, “great is the power of memory,” to borrow Augustine’s words; but the Christian rhetorician did not know to what eccentricity this praise of happy memory could lead. And Cicero might have called the exploits of trained memory “almost divine”; but he, too, could not have predicted the excesses produced by the occult memory of a man of the Renaissance, the one whom Yates calls the “Magus of Memory” (293).

To conclude this rapid overview of the ars memoriae, I would like to refer to the questions posed by Frances Yates at the end of her own study, before she writes the sort of post-scriptum that composes her final chapter, titled “The Art of Memory and the Growth of Scientific Method” (368ff.). I quote Yates: “The question to which I can give no clear or satisfactory answer is: What was the occult memory? Did the change from forming corporeal similitudes of the intelligible world to the effort to grasp the intelligible world through the tremendous imaginative exercises such as those to which Giordano Bruno devoted his life really stimulate the human psyche to a wider range of creative imaginative achievement than ever before? Was this the secret of the Renaissance and does the occult memory represent that secret? I bequeath this problem to others” (367).

How to answer Yates? We cannot be content to simply record the fact that the history of ideas has not produced a sequel to this extravagant cultivation of memory and that a new chapter was opened with the notion of method, with Francis Bacon’s Novum Organon and Descartes’s Discourse on Method. After all, the ars memoriae, with its cult of order on the levels of both images and places, was in its own way an exercise of method. It is at the heart of the enterprise that the reason for its eclipse must be sought. Francis Bacon goes straight to the critical point when he denounces the prodigious ostentation that lies at the base of the culture of artificial memory. From the start, this art was vaunted as an exploit, as a marvel. A sort of inebriation—Kant will speak of Schwärmerei in the sense at once of enthusiasm and drunkenness—insinuated itself at the point of intersection of natural memory and artificial memory. An inebriation that transformed into its opposite the modesty of a hard schooling begun within the limits of natural memory, with regard to which it had always been legitimate to try to reinforce its powers, that is to say, at once its scope and its exactness. For it is indeed the notion of limit that is at issue here. With Giordano Bruno the transgression of limits is carried to its furthest point. But what limits? Basically, this is the limit suggested by the relation of memory to forgetfulness.¹³ The ars memoriae is an outrageous denial of forgetfulness and, following this, of the weaknesses inherent in both the preservation of traces and their evocation. Correlatively, the ars memoriae is unaware of the constraints of traces. As was suggested above in the context of our discussion concerning the Platonic metaphor of the tupos, of the imprint, the phenomenological notion of trace, distinct from the material, corporeal, cortical condition of the imprint, is constructed on the basis of being-affected by an event, becoming its witness after-the-fact through narration. For the artificial memory, all is action, nothing is passion. The places are sovereignly chosen, their order hides the arbitrariness of their selection; and the images are no less manipulated than are the places to which they are assigned. A twofold denial is then posed: of forgetting and of being-affected. The seed of the deadly infatuation lies in this original denial. Great indeed is the power of memory, Augustine exclaims. But, as we have noted in the opening pages of this book, he was not unaware of forgetfulness; with fright, he took the measure of its danger and its ravages. Moreover, from this denial of forgetting and of being-affected results the preeminence accorded to memorization at the expense of remembering (remémoration). The overemphasis on images and places by the ars memoriae has as its price the neglect of events that astonish and surprise. By breaking the pact of memory with the past in this way to the benefit of private writing in an imaginary space, the ars memoriae passed from the athletic exploits of a trained memory to what Yates rightly calls an “alchemy of the imagination.” The imagination, freed from its service to the past, has taken the place of memory. The past, the absent with respect to the history that recounts it, constitutes the other limit of this ambitious mnemotechnics, along with forgetfulness, which we will discuss below in terms of its solidarity with the pastness of the past.¹⁴

There are two ways to follow up on these primary considerations that reintroduce the idea of limit into a project that excludes it. The first is to restore measure to a culture of memorization within the limits of natural memory; the second is to take into consideration the abuses that are grafted onto its use, once this use becomes a form of manipulation under the guise of artificial memory.

The final considerations of this section are devoted to the modalities of an art of memorization contained within the bounds of natural memory. We will, therefore, retreat from the magic of memory in the direction of a pedagogy of memory—framing the cultivation of memory within an educational project. In this way, we are led back to the discussion initiated above concerning the use and abuse of memory in education. We return to it, however, mindful of the principal episodes of the fabulous history of artificial memory. It is not, truly speaking, the power of an imagination carried to extreme that serves as the target in the process of reciting by heart, during the very period of the Renaissance that witnessed the feats of artificial memory, but rather the authority of the cultural heritage transmitted by texts. For these critics, the beast of burden is readily labeled the animal emblematic of silly memory plodding under the weight of imposed knowledge: “You simply,” says Montaigne, “produce donkeys laden with books” (Essays, I, 25).¹⁵ It is noteworthy that the critique of memorizing memory coincided with the praise of ingenium—genius, spirit—in the sense given to this word by Helvétius in De l’esprit.¹⁶ In this way, there was a fusion between the plea for method, harkening back to Ramus, and the plea for ingenium, which contained the seed of the cultivation of the creative imagination. This fusion occurs in the notion of judgment, dear to the champions of the Enlightenment. But at the very heart of judgment, the rational understanding does not succeed in restraining ingenium. Witness Rousseau’s revolt against the Enlightenment thinkers. It is in the name of an untamed ingenium that Rousseau attacks the cultivation of even natural memory, pounding it with the strongest blows: “Émile will never learn anything by heart, not even the fables, not even those of La Fontaine, as innocent, as charming as they are.”¹⁷

We might wonder if, at this point, the critique of memorizing memory has not outstripped its goal. To the abuse of excess in Giordano Bruno responds the abuse of deficiency in Jean-Jacques Rousseau. It is true that it is not the same memory celebrated by one and struck down by the other. The excess of the first affects the memoria artificiosa, the abuse by deficiency of the second wrongs natural memory, which also demands its due. We return, then, to the mind, beyond the scholastic use of memorization, to the admirable feats of professional memory, to the memories of doctors, judges, teachers, and so on, and to those of the artists of dance, theater, and music. We have, in truth, never finished with memorization.

Before turning the page on the ars memoriae, I would like to take a brief excursus with Harald Weinrich into the region of forgetting. We stated above that the ars memoriae was inspired by the exorbitant desire “not to forget anything.” Does not a measured use of memorization also imply a measured use of forgetting? Can we not, following Descartes, speak of a “methodical forgetting”? If, in fact, methodical doubt produces a reflective rejection of the entire memory-based pedagogy, and in this sense implies a certain strategy of forgetting, does not the rule of recapitulation of the Discourse on Method constitute a methodical use of memory, but of a natural memory freed from mnemotechnics? In the same way, can we not speak of “enlightened forgetting,” in the spirit of the Enlightenment? Enlightened forgetting, which would serve as a guardrail (garde-fou, in the strict sense of the term) to protect against the frenzied cultivation of memorizing memory? We will have to return to this at the appropriate time, when we will attempt to give to the ars memoriae the symmetry that would come from the ars oblivionis, as this is expressed by Weinrich in Lethe.¹⁸ In the meantime, these suggestions converge toward the plea for a measured use of remembering (rémemoration)—under the heading of a just memory, an idea that will take shape when the time comes through our reflection on the abuses of a memory manipulated by ideology. In a sense, the poetic surpassing of artificial memory by Dante and methodical forgetting after the manner of Descartes lead back, each in its own way, to the rich problematic of natural memory.

THE ABUSES OF NATURAL MEMORY: BLOCKED MEMORY, MANIPULATED MEMORY, ABUSIVELY CONTROLLED MEMORY

The present study will be devoted to a typology of the uses and abuses of natural memory. The path in this direction has been cleared by Nietzsche in his Unfashionable Observations, the second of which bears the eloquent title, “On the Utility and Liability of History for Life.”¹⁹ The manner of questioning, inaugurated by this text, unites in a complex semiology the medical treatment of symptoms and the philological treatment of texts. To be sure, the polemic raised here concerns history above all, more precisely, the philosophy of history and its place in culture. But the tone is set for a similar treatment of memory, in particular, collective memory, which, as I will repeat at the beginning of the next study, constitutes the soil in which historiography is rooted. As was stated at the beginning of the present study, it is as exercised that memory falls under this perspective.

I propose the following reading grid, in an effort to avoid a broad and indiscriminate use of the notion of the abuse of memory. I will first distinguish a clearly pathological approach, employing clinical, and eventually therapeutic, categories, borrowed principally from psychoanalysis. I will attempt to restore breadth and density to this pathology by relating it to some of the most basic human experiences. Then I will make room for concerted forms of the manipulation or instrumentalization of memory, within the framework of a critique of ideology. It is at this median level that notions of the abuse are the most relevant. Finally, I would like to reserve for a normative, explicitly ethico-political viewpoint the question of the duty of memory; this normative viewpoint must be carefully distinguished from the preceding viewpoint with which it is too often confused. In this way, the path from one level to the next will become a path from one figure to the next characterizing the uses and abuses of memory, from blocked memory to forced memory, passing through manipulated memory.

The Pathological-Therapeutic Level: Blocked Memory

It is on this level and from this viewpoint that we can legitimately speak of wounded, even of sick memory. Common expressions such as traumatism, wound, scar, and so forth, attest to this. The use of these words, themselves expressions of pathos, cannot but give rise to certain serious difficulties. Up to what point, we will first ask, are we authorized to apply to collective memory categories forged in the analytical colloquy, hence at the interpersonal level, marked principally by the mediation of transference? This first difficulty will be definitively resolved only at the end of the following chapter. We will acknowledge here, provisionally, the operative value of the concept of collective memory; moreover, the use it will be put to will later contribute to legitimizing this problematical concept. There is another difficulty to be resolved here: one can wonder to what extent a pathology of memory, and so the treatment of memory as pathos, fits into an inquiry into the exercise of memory, the memorial tekhnē. The difficulty is a new one: what is at issue are the individual and collective alterations due to the use, to the practice of memory.

To situate us with regard to this twofold difficulty, I thought it appropriate to turn to two remarkable essays by Freud and to compare them to each other, which their author seems not to have done. The first of these texts, dating from 1914, is titled, “Remembering, Repeating, and Working-Through.”²⁰ Note right away that the German title contains only verbs, stressing the fact that these are three processes belonging to the play of psychical forces with which the psychoanalyst “works.”

The starting point for Freud’s reflection lies in identifying the main obstacle encountered by the work of interpretation (Deutungsarbeit) along the path of recalling traumatic memories. This obstacle, attributed to “resistances due to repression” (Verdrängungswiderstände), is designated by the term “compulsion to repeat” (Wiederholungszwang); it is characterized among other things by a tendency to act out (Agieren), which Freud says is substituted for the memory. The patient “reproduces it not as a memory but as an action; he repeats it, without, of course, knowing that he is repeating it” (S.E., 12:150; G.W., 10:129). We are not far from the phenomenon of obsession mentioned above. Let us leave aside the implications of the phenomenon as they concern forgetting. We shall return to them in the chapter on forgetting in part 3. Then the emphasis is on acting out and on the place this takes, without the knowledge of the patient. What matters to us is the tie between the compulsion to repeat and resistance, as well as the substitution of this twofold phenomenon for the memory itself. This forms the obstacle to continuation of the analysis. Beyond this clinical aspect, Freud offers two therapeutic proposals that will be of the greatest importance for us, when we transpose the clinical analysis to the level of collective memory, as we consider ourselves authorized to do at that stage of the discussion. The first proposal concerns the analyst, the second, the analysand. The analyst is advised to be very patient with regard to the repetitions occurring under the cover of transference. Transference, Freud notes, creates something like an intermediary domain between illness and real life; one can speak of it as a “playground,” in which the compulsion is authorized to manifest itself in almost total freedom, offering an opportunity for the pathogenic background of the subject to manifest itself openly. But something is also asked of the patient: ceasing to lament or to hide his true state from himself, “he must find the courage to direct his attention to the phenomenon of his illness. His illness itself must no longer seem to him contemptible, but must become an enemy worthy of his mettle, a piece of his personality, which has solid ground for its existence and out of which things of value for his future life have to be derived” (S.E., 12:152; G.W., 10:132). Otherwise, there will be no “reconciliation” (Versöhnung) of the patient with the repressed material. Let us hold in reserve this term of reconciliation, which will return to the fore in our subsequent reflections on forgiveness. Let us stop for a moment with the twofold handling of resistances by the patient and the analyst, to which Freud gives the name of Durcharbeiten (S.E., 12:155; G.W., 10:136), “working through” in the English translation, perlaboration in the French, or, as I prefer to say, remaniement (reworking). The important word here is work—or rather, “working”—which underscores not only the dynamic character of the entire process, but the collaboration of the analysand in this work. It is in relation to this notion of work, stated in the form of a verb, that it becomes possible to speak of memory itself, freed in this way, as a work—the “work of remembering” (Erinnerungsarbeit) (S.E., 12:153; G.W., 10:133). Work is thus the word repeated several times and symmetrically opposed to compulsion: the work of remembering against the compulsion to repeat, thus could be summed up the theme of this precious little essay. Belonging to this work are both the patience of the analyst with respect to the repetition channeled by the transference and the courage required on the part of the analysand to recognize himself as ill, in search of a truthful relation to his past.

Before examining the transpositions that might be made between the analytic relation and the public plane of collective memory and of history, and taking into consideration the reservation in principle mentioned above, let us turn to the second essay, titled “Mourning and Melancholia” (“Trauer und Melancholie”) (S.E., 14; G.W., 10).²¹ This essay, no doubt, offers greater resistance than the preceding one to a transposition to the plane of collective memory, insofar as mourning is treated less in its own right, precisely as work, than as a term of comparison in order better to pierce the enigmas of melancholia. It is the pairing with the preceding essay that can help to draw some positive information from the comparison itself concerning the work of mourning.²² But especially, this essay awakens resounding echoes of a millenary experience, which has had melancholia itself as a theme of meditation and as a source of torment.

These initial reservations do not prevent us from noting that it is mourning—the work of mourning—that is first taken as a term of comparison assumed to be directly accessible, at least initially. Furthermore, the pair mourning and melancholia is to be taken as one block, and it is the tendency of mourning to become melancholia and its difficulty in extracting itself from this terrible neurosis that will give rise to my subsequent reflections on the pathology of collective memory and the therapeutic perspectives this opens.

“Mourning,” it is stated, “is regularly the reaction to the loss of a loved person, or to the loss of some abstraction which has taken the place of one, such as one’s country, liberty, an ideal, and so on” (S.E., 14:243). An opening is thus made from the outset in the direction that we shall take below. And the first question that the analyst poses is to know why, in the case of certain patients, “the same influences produce melancholia instead of mourning” (my emphasis). The expression “instead of . . .” directly indicates the kinship, from the viewpoint of my strategy of argumentation, between the two essays we are comparing: instead of remembering, acting out; instead of mourning, melancholia. So in a certain sense it is the opposition between mourning and melancholia that is at issue here, the bifurcation on the “economic” level between different affective investments, and in this sense, of a bifurcation between two sorts of work. The first opposition noted by Freud is the dissimulation of “self-regard” (Selbstgefühl) in melancholia, whereas “the disturbance of self-regard is absent in mourning” (244). Whence the question: what is the work supplied in mourning? Answer: “Reality-testing has shown that the loved object no longer exists, and it proceeds to demand that all libido shall be withdrawn from its attachment to that object. This demand arouses understandable opposition” (244). There follows a careful description of the “great expense of time and cathetic energy” (245) required for the obedience of the libido to the commands of reality. Why is the cost so high? Because “the existence of the lost object is psychically prolonged.” The heavy price to pay for this liquidation is, therefore, due to the hypercathexis of the memories and expectations by which the libido remains attached to the lost object: “this compromise by which the command of reality is carried out piecemeal . . . [is] extraordinarily painful.”

But why then is mourning not melancholia? And what is it that makes mourning tend toward melancholia? What makes mourning a normal, albeit painful, phenomenon is that “when the work of mourning is completed the ego becomes free and uninhibited again” (245). It is from this angle that the work of mourning can be compared to the work of remembering. If the work of melancholia occupies a strategic position in the present essay parallel to that occupied by the compulsion to repeat in the previous one, this suggests that it is as a work of remembering that the work of mourning proves to be liberating, although at a certain cost, and that this relation is reciprocal. The work of mourning is the cost of the work of remembering, but the work of remembering is the benefit of the work of mourning.

Before drawing the consequences I have in mind, let us look at some complementary lessons that the work of melancholia contributes to the preceding picture of the work of mourning. Starting again from the opening remark concerning the diminution of Ichgefühl in melancholia, we must say that unlike mourning, in which it is the universe that seems impoverished and empty, in melancholia it is the ego that finds itself in desolation: it succumbs to the blows of its own devaluation, its own accusation, its own condemnation, its own abasement. But this is not all, not even the essential point: do not the reproaches addressed to the self serve to mask reproaches aimed at the love object? As Freud audaciously puts it: “Their complaints are really ‘plaints’ in the old sense of the word [Ihre Klagen sind Anklagen]” (248). These “plaints” or accusations can even defame the loved one, reaching into the deepest recesses of mourning. Freud proposes the hypothesis that the accusation, by weakening the object-cathexis, facilitates the retreat into the ego, while transforming the discord with the other into self-laceration. We will not follow Freud any further in his properly psychoanalytic investigations concerning the regression from object-love to original narcissism, even to the oral phase of the libido—including the element of sadism incorporated into narcissism—or the investigations concerning the tendency of melancholia to invert itself into the symptomatically inverse state of mania. Freud himself, moreover, is very circumspect in his explorations. We will limit our remarks to this citation: “Melancholia, therefore, borrows some of its features from mourning, and the others from the process of regression from narcissistic object-choice to narcissism” (250).

If we now ask what melancholia teaches about mourning, we must return to the Ichgefühl, considered well established and which Freud once characterized as “recognition of oneself.” To it belongs shame before others, something the melancholic is unaware of, so occupied is he with himself. Self-esteem and shame would then seem to be joint components of mourning. Freud remarks on this: “the censorship of consciousness”—the expression of the agency generally termed conscience—goes together with “reality-testing, among the major institutions of the ego” (247). This remark links up with what was said in the preceding essay concerning the responsibility of the analysand in forgoing acting out and in pursuing the work of memory. Another observation: if in melancholia complaints are accusations, mourning also contains the mark of this uncanny similarity, under the condition of a certain measure proper to mourning, a measure that limits the accusation as well as the self-reproach under which it is concealed. Finally—and this is perhaps the most important thing—does not the proximity between Klage and Anklage, between complaint and reproach, which is exhibited in melancholia, stem from the ambivalence of amorous relations in which love and hate are side-by-side, even in mourning?

However, it is on the positive outcome of mourning, in contrast to the disaster of melancholia, that I would like to conclude this brief incursion into one of Freud’s most famous essays: “Melancholia confronts us with yet other problems, the answer to which in part eludes us. The fact that it passes off after a certain time has elapsed without leaving traces of any gross changes is a feature it shares with mourning. We found by way of explanation that in mourning time is needed for the command of reality-testing to be carried out in detail, and that when this work has been accomplished the ego will have succeeded in freeing its libido from the lost object. We may imagine that the ego is occupied with analogous work during the course of melancholia; in neither case have we any insight into the economics of the course of events” (252–53). Let us set aside Freud’s admission concerning the explanation and retain only his clinical lesson: the time of mourning is not unrelated to the patience required by analysis in the passage from repetition to memory. Memory does not only bear on time: it also requires time—a time of mourning.

I do not want to end this confrontation between mourning and melancholia on Freud’s statement of perplexity: “in neither case have we any insight into the economics of the course of events.” If the last word on mourning and on the work of mourning in psychoanalysis has not been uttered, this is because it has also not been uttered on melancholia. Must melancholia be left to the physicians, psychiatrists, and psychoanalysts? Is it solely a mental illness? For anyone who has read Saturn and Melancholy by Raymond Klibansky, Erwin Panofsky, and Fritz Saxl,²³ the nosological reduction of melancholia, begun by E. Kraepelin and redirected by Ludwig Binswanger, is unacceptable. How indeed could we fail to mention the place held by melancholia in the ancient system of four humors in Greek medicine, in which the melancholic humor—black bile (altra bilis)—takes its place alongside the sanguine, choleric, and phlegmatic humors? Here is one more list to memorize, taking into account the network of correspondences with the cosmic elements, divisions of time, ages of life: Melancholia, state medieval texts of the twelfth century, imitates the earth, grows in autumn, reigns in maturity. Physiology, psychology, cosmology are thus found to be joined, following a threefold principle: the search for primary elements common to the microcosm and the macrocosm, the establishment of a numerical expression for these complex structures, and the law of harmony or proportionality among the elements. In this, we recognize the spirit of Pythagoras, followed by Empedocles. What is important in the sort of excursus that I am making, beyond—or perhaps, short of—Freud, is that the concept of humor has continued to oscillate between the idea of illness and the idea of character or temperament, the scale showing the degree of harmony or disharmony among the elements. It is precisely in melancholia that the ambivalence culminates, and this becomes the critical point of the entire system. This privilege, so to speak, accorded to melancholia has become more pointed as the theory of the four humors has been transformed into a theory of temperaments and mental types. Depression and anxiety (or fear) become the characteristic symptoms of melancholia. Melancholia then becomes synonymous with insanity, madness. The intersection between melancholia and humoreal theory and the madness of tragic heroes—Ajax, Hercules, Bellerophone—raised by Plato to the level of a philosopheme, is completed with the most famous of the “Problems” attributed to Aristotle, book 30, problem 1—a monograph on black bile, according to our sources. “Why,” asks the author of problem 1, “is it that all those who have become eminent in philosophy or politics or poetry or the arts are clearly melancholics?” And the text adds the names of Empedocles, Plato, and Socrates to the list of troubled spirits. How, then, can we not call to mind the theory of the multiple figures of mania in Plato himself and the comparison made in so many dialogues between exaltation, ecstacy, inebriation, and other “divine” states? For all these states are the work of black bile! Here the normal and the pathological rub shoulders, melancholia being passed from the physician to the pedagogue and vice versa. The melancholic is the “exceptional” individual. The Romantic theory of “genius” is contained in germ in this ambiguous description of “fury” (to borrow Cicero’s translation of the Greek mania). The Stoics alone resist, clearly opting before the fact for the psychoanalytic reading.

It is the thinkers of the Renaissance, who, beyond the medieval transmission of the divided heritage received from the Greek physicians and philosophers of nature, redirected the meditation on melancholia toward the modern doctrine of genius.²⁴ The astral theme, which scholars trace back to Arab astrology, stands ready to emerge in our impassioned Renaissance figures.²⁵ The Renaissance man—represented by an Erasmus, a Marsilio Ficino, a Pico della Mirandola, a Nicholas of Cusa, a Dürer—is in pursuit less of individual salvation than of the unfettered development of individual spontaneity; it is in this élan, which announces the fire of Romantic genius, that the troubling contrast between exaltation and depression is to be found. The negative pole is nothing other than what Lessing will call “voluptuous melancholia,” heir to medieval acedia, that perfidious temptation hesitating between sin and illness. The Renaissance man, however, also makes the wager that melancholia can be melancholia generosa (Klibansky et al., 241).²⁶

It is, however, in Dürer’s engraving, Melencolia I, that all the attempts at rehabilitating Saturn and melancholia are crystallized. And it is over this engraving that the commentaries of Klibansky, Panowsky, and Saxl linger. Let us “read” this engraving. A woman is seated, her gaze plunging into empty space, her face indistinct, her chin resting on a clenched fist; at her waist hang some keys, symbols of power, and a purse, symbol of wealth—two forms of vanity, in short. Melancholia remains forever this hunched, pensive figure. Fatigue? Grief? Sorrow? Meditation? The question returns: is this the posture of declining health or of reflective genius? The response is not to be sought in the human figure alone; the environment is also tacitly eloquent: unused instruments lie scattered across the immobile scene—a three-dimensional geometrical figure depicting geometry, the fifth of the “liberal arts.” The vanity of knowledge is thus incorporated in the idle figure. This fusion of geometry giving way to melancholia and melancholia lost in a dreamlike geometry gives to Melencolia I its enigmatic power:²⁷ Might not truth itself be gloomy, following the adage of Ecclesiastes?

The question then arises: what shadowy light is cast on Freud’s text by this backward turn? It seems to me that to make sense of it we must extend the inquiry into melancholia to one of the sources of the theme buried under medicine, psychology, literature, and iconography: behind the lament of an Alain Chartier invoking “Dame Merencolye” or of Roi René celebrating “Dame Tristesse” is silhouetted acedia, named already above, in which the spiritual leaders of the Middle Ages saw the worst of temptations, beyond even sanguine “luxury,” choleric “discord,” namely, complaisance toward sadness. Acedia is this sort of laziness, lassitude, disgust, to which the member of a religious order who is not praying or working is in danger of succumbing. Do we not touch here upon the moral ground of melancholia, barely alluded to by Freud in the vocable, Selbstgefühl? That toward which acedia is complaisant, is it not the sadness of meditative memory, the specific “mood” of finitude rendered conscious of itself? This sadness without a cause, is it not akin to Kierkegaard’s sickness-unto-death, that relative of despair, or rather, following Gabriel Marcel’s suggestion, of no hope (inespoir)?²⁸ By moving back in this way to the acedia of the religious, have we not provided a worthy vis-à-vis for the work of mourning? Someone may object that the work of mourning has no antecedent in the literature of melancholia. In this sense, it would indeed be a creation of Freud. But the work of mourning also has antecedents in the antidotes by which the medical, psychological, moral, literary, and spiritual tradition has attacked melancholia. Among these remedies, we find gaiety, humor, hope, trust, and also . . . work. The authors of Saturn and Melancholy are not wrong to seek in lyrical poetry dating from the end of the Middle Ages and from the Renaissance, in particular, the English Renaissance, from Milton and the Shakespeare of the Sonnets to Keats, the praise of a contrasting and, if I may call it so, a dialectical humor, in which Delight responds to Melancholia under the auspices of beauty. This review of the poeticized figures of melancholia would have to be followed as far as Baudelaire, to restore to melancholia its enigmatic profoundness, which no nosology could never exhaust. This is the direction in which Jean Starobinsky pulls us in La Mélancolie au miroir: Trois lectures de Baudelaire.²⁹ Does not the opening poem, “To the Reader,” in The Flowers of Evil term the “saturnine book” the book of Ennui? The lost gaze of Melancholia is reflected in the mirror of reflective consciousness, whose reflections are modulated by poetry. A path of memory is opened in this way by “Spleen”: “Je suis le sinistre miroir.” “J’ai plus de souvenirs que si j’avais mille ans. . . .” It is, in fact, figures of the historical past that haunt the famous poem “Le Cygne” (“The Swan”) which we will approach from a different angle, at the point where the memorization of history intersects with the historization of memory:³⁰

Andromaque—my thoughts are turned to you! . . .

This lying Simois swelled by your tears,

Has suddenly enriched my fertile memory . . .

Thus in the forest where my mind is exiled

An old Memory blows mightily into its horn!³¹

And why should we not refer in fine to Beethoven’s last quartets and sonatas and to their powerful evocation of a sublime sadness? There, the word has been uttered: sublimation. This missing piece in the panoply of Freud’s metapsychology might perhaps have provided him with the secret of the reversal from the complaisance toward sadness to sadness sublimated—into joy.³² Yes, grief is that sadness that has not completed the work of mourning. Yes, joy is the reward for giving up the lost object and a token of the reconciliation with its internalized object. And, inasmuch as the work of mourning is the required path for the work of remembering (souvenir), joy can also crown with its grace the work of memory (mémoire). On the horizon of this work: a “happy” memory, when the poetic image completes the work of mourning. But this horizon recedes behind the work of history, the theory of which has yet to be established beyond the phenomenology of memory.

Having said this, I return to the question left in abeyance, namely, to what extent it is legitimate to transpose to the plane of collective memory and to history the psychological categories proposed by Freud in the two essays we have just read. A provisional justification can be found on both sides—on the side of Freud and on the side of the phenomenology of wounded memory.

On the side of Freud, one will have noted the various allusions to situations that go far beyond the psychoanalytic scene, in terms of both the work of remembering and the work of mourning. This extension is all the more anticipated as all of the situations referred to in the psychoanalytic treatment have to do with the other, not only the other of the “familial novel,” but the psychosocial other and the other, as it were, of the historical situation. Furthermore, Freud did not hesitate to make similar extrapolations in Totem and Taboo, in Moses and Monotheism, in The Future of an Illusion, or in Civilization and Its Discontents. And even certain of his private psychoanalyses, we may venture to say, were psychoanalyses in absentia, the most famous of these being the case of Dr. Schreber. And what are we to say of “Michelangelo’s Moses” or of A Childhood Memory of Leonardo da Vinci? No scruples should hamper us, then, on this score. The transposition has been rendered easier by certain reinterpretations of psychoanalysis close to hermeneutics, as we see in some of the earlier works of Jürgen Habermas, in which psychoanalysis is reformulated in terms of desymbolization and resymbolization, and in which the emphasis is placed on the role of systematic distortions of communication on the plane of the social sciences. The sole objection that has not been answered in the hermeneutical interpretations of psychoanalysis concerns the absence of recognized therapists in interhuman relations. But could we not say that, in this case, the public space of discussion constitutes the equivalent of what above was called the “playground” as the intermediary region between the therapist and the analysand?

Regardless of this genuinely formidable difficulty, it is more important for our purpose to look to collective memory and to discover there the equivalent of the pathological situations with which psychoanalysis is concerned. It is the bipolar constitution of personal and community identity that, ultimately, justifies extending the Freudian analysis of mourning to the traumatism of collective identity. We can speak not only in an analogical sense but in terms of a direct analysis of collective traumatisms, of wounds to collective memory. The notion of the lost object finds a direct application in the “losses” that affect the power, territory, and populations that constitute the substance of a state. Mourning behaviors, from the expression of affliction to complete reconciliation with the lost object, are directly illustrated by the great funeral celebrations around which an entire people is assembled. In this way, we can say that such mourning behaviors constitute a privileged example of the intersecting relations between private and public expression. It is in this way that our concept of a sick historical memory finds justification a posteriori in this bipolar structure of mourning behaviors.

The transposition of pathological categories to the historical plane would be more completely justified if we were able to show that it applies not only to the exceptional situations just mentioned, but that these depend on a fundamental structure of collective existence. What I refer to here is the fundamental relation of history to violence. Hobbes was not wrong in making political philosophy arise out of an original situation in which the fear of violent death pushes man out of the “state of nature” into the bonds of a contractual pact that, first of all, guarantees him security; moreover, there exists no historical community that has not been born out of a relation that can, without hesitation, best be likened to war. What we celebrate under the title of founding events are, essentially, acts of violence legitimated after the fact by a precarious state of right. What was glory for some was humiliation for others. To celebration on one side corresponds execration on the other. In this way, symbolic wounds calling for healing are stored in the archives of the collective memory. More precisely, what, in historical experience, takes the form of paradox—namely, too much memory here, not enough memory there—can be reinterpreted in terms of the categories of resistance and compulsion to repeat, and, finally, can be found to undergo the ordeal of the difficult work of remembering. Too much memory recalls especially the compulsion to repeat, which, Freud said, leads us to substitute acting out for the true recollection by which the present would be reconciled with the past: how much violence in the world stands as acting out “in place of” remembering! We can, if we like, speak of repetition-memory for these funeral celebrations. But then it must immediately be added that this repetition-memory resists criticism and that recollection-memory is fundamentally a critical memory.

If this be the case, then too little memory belongs to the same reinterpretation. What some cultivate with morose delectation, and what others flee with bad conscience, is the same repetition-memory. The former love to lose themselves in it, the latter are afraid of being engulfed by it. But both suffer from the same lack of criticism. They do not attain what Freud termed the work of remembering.

We can take one further step and suggest that it is on the level of collective memory, even more perhaps than on that of individual memory, that the overlapping of the work of mourning and the work of recollection acquires its full meaning. When it is a matter of national self-love, we can properly speak of a lost love-object. It is always in terms of its losses that the wounded memory is forced to confront itself. What it does not know how to do is the work imposed on it by reality-testing: giving up the investments (cathexes) by which the libido continues to be bound to the lost object, as long as the loss has not been definitively internalized. But this is also the place to underscore that this submission to reality-testing, constituting the true work of mourning, is also an integral part of the work of recollection. The suggestion made above concerning the exchanges of meaning between the work of recollection and the work of mourning finds its full justification here.

A transition from the pathological level to the properly practical level is provided by notations concerning the appropriate therapy for these troubles. Freud continuously calls upon the cooperation of the analysand, placing the analytical experience, in this way, entirely at the point of intersection between the passive, pathic side of memory and the active side of the exercise of memory. In this respect, the notion of work—the work of remembering, the work of mourning—occupies a strategic position in reflection on the failures of memory. This notion supposes that the difficulties in question are not only undergone, but that we are responsible for them, as witnessed by the therapeutic advice that accompanies the working-through. In one sense, the abuses of memory, which we will now discuss, can appear as perverse diversions of this work, in which mourning is joined to remembering.

The Practical Level: Manipulated Memory

Whatever may be the validity of the pathological interpretations of the excesses and deficiencies of collective memory, I would not want them to occupy all of the territory. A distinct place must be set aside, next to the more or less passive modes of these “abuses,” undergone, suffered—even taking into account the correction made by Freud himself to this unilateral treatment of passivity—for abuses, in the strong sense of the term, resulting from a concerted manipulation of memory and of forgetting by those who hold power. Thus I shall be speaking here less of wounded memory than of instrumentalized memory (the Weberian category of rationality in accordance with an end—Zweckrationalität—in opposition to the category of rationality in accordance with a value—Wertrationalität—has a place here; as does the category employed by Habermas of “strategic reason” in opposition to “communicational reason”). It is on this plane that we can more legitimately speak of the abuses of memory, which are also abuses of forgetting.

The specificity of this second approach lies in the intersection of the problematics of memory and of identity, collective as well as personal.

We will linger over this problem of intersection in the next chapter, in the context of Locke’s theory in which memory is established as the criterion of identity. The heart of the problem is the mobilization of memory in the service of the quest, the appeal, the demand for identity. In what follows from this, we recognize some disturbing symptoms: too much memory, in a certain region of the world, hence an abuse of memory; not enough memory elsewhere, hence an abuse of forgetting. It is in the problematic of identity that we have to seek the cause of the fragility of memory manipulated in this way. This is in addition to the properly cognitive frailty resulting from the proximity between memory and imagination, which finds in the latter its spur and its helper.

What constitutes the fragility of identity? It is identity’s purely presumptive, alleged, reputed character. This “claim,” as one would say in English, this Anspruch in German, is lodged in the responses to the question, “Who?” “Who am I?” Responses in terms of “What?” of the form: this is what we are, we, ourselves. How we are, this way and not otherwise. The fragility of identity consists in the fragility of these responses in terms of what, claiming to give the recipe of the identity proclaimed and reclaimed. The problem is therefore carried back a step, from the fragility of memory to that of identity.

As the primary cause of the fragility of identity we must cite its difficult relation to time; this is a primary difficulty that, precisely, justifies the recourse to memory as the temporal component of identity, in conjunction with the evaluation of the present and the projection of the future. This relation to time is a problem by reason of the equivocal nature of the notion of the same, implicit in the notion of the identical. What, in fact, does it mean to remain the same over time? In the past, I took on the challenge of this enigma, with respect to which I suggested distinguishing two senses of identical: the same as idem, même, gleich, and the same as ipse, self, Selbst. It seemed to me that self-constancy over time rests on a complex interplay of sameness and ipseity, if I may venture these barbarisms; in this equivocal play, the practical and pathetic aspects are more formidable than the conceptual, epistemic ones. I will say that the temptation of identity, of déraison identitaire, as Jacques Le Goff calls it, consists in the retreat of ipse identity into idem identity or, if one prefers, in the slippage, the drift, from the flexibility, proper to self-constancy as manifested in the promise, to the inflexible rigidity of a character, in the quasi-typographical sense of the term.

The second cause of fragility lies in the confrontation with others, felt to be a threat. It is a fact that the other, because other, comes to be perceived as a danger for one’s own identity, our identity as well as my identity. To be sure, we may find this surprising: is our identity so fragile that we are unable to bear, unable to endure the fact that others have different ways than our own of leading their lives, of understanding themselves, of inscribing their own identity in the web of living together? This is so. There are indeed humiliations, real or imagined attacks on self-esteem, under the blows of poorly tolerated otherness, that turn a welcome into rejection, into exclusion—this is the relation that the same maintains with the other.

Third cause of fragility: the heritage of founding violence. It is a fact that there is no historical community that has not arisen out of what can be termed an original relation to war. What we celebrate under the heading of founding events are, essentially, violent acts legitimated after the fact by a precarious state of right, acts legitimated, at the limit, by their very antiquity, by their age. The same events are thus found to signify glory for some, humiliation for others. To their celebration, on the one hand, corresponds their execration, on the other. It is in this way that real and symbolic wounds are stored in the archives of collective memory. Here, the third cause of the fragility of identity merges with the second. It remains to be shown under what angle the forms of misuse of memory can be grafted onto the demand for identity, whose specific fragility we have just outlined.

The manipulations of memory we are going to discuss below result from the intervention of a disturbing and multiform factor that insinuates itself between the demand for identity and the public expressions of memory. This is the phenomenon of ideology, involving a mechanism I have attempted to demonstrate elsewhere.³³ The ideological process is opaque in two ways. First, it remains hidden; unlike utopia, it is unacknowledged; it masks itself by inverting itself, denouncing its adversaries in the field of competition between ideologies, for it is always the other who stoops to ideology. On the other hand, the process is extremely complex. I have suggested distinguishing three operative levels of the ideological phenomenon, in terms of the effects it exerts on an understanding of the human world of action. Running from top to bottom, from surface to depth, these effects are, in succession, distortions of reality, the legitimation of the system of power, and the integration of the common world by means of symbolic systems immanent in action. At the deepest level, that on which Clifford Geertz works, the ideological phenomenon indeed appears to constitute an unsurpassable structure of action, to the extent that symbolic mediation marks the difference between the motivations of human action and the hereditary structures of genetically programmed behaviors. A remarkable correlation is established at this fundamental level between a symbolic synthesis and a semiotic system, some of this belonging clearly to a system of rhetorical tropes.³⁴ Considered at this deep level, the analysis of the ideological phenomenon is obviously part of a “semiotics of culture.” It is in fact in this role that ideology, as a factor of integration, can be established as the guardian of identity, offering a symbolic response to the causes affecting the fragility of this identity. At this level of radicality, that of symbolically mediated action, there is as yet no manipulation, hence no abuse of memory. One can speak only of the silent constraint exerted on the mores of a traditional society. This is what makes the notion of ideology practically ineradicable. But it must be added straight away that this constitutive function of ideology can scarcely operate outside of the connection to its second function—the justification of a system of order or power—nor can it operate even potentially apart from the function of distortion that is grafted onto the preceding one. At the limit, it would only be in societies without a hierarchical political structure—and, in this sense, societies without power—that we might have a chance of encountering the naked phenomenon of ideology as an integrative structure in its, so to speak, innocent form. Ideology, when all is said and done, revolves around power.³⁵

In fact, what ideology aims to legitimize is the authority of order or power—order, in the sense of an organic relation between the whole and the part; power, in the sense of a hierarchical relation between governing and governed. In this regard, the analyses that Max Weber devotes to the notions of order (Ordnung) and domination (Herrschaft) are of considerable interest for our undertaking, even if the author of Economy and Society does not treat ideology and its relation to identity thematically. The entire Weberian analysis of power revolves around the claim to legitimacy raised by every form of power, be it charismatic traditional, or bureaucratic.³⁶ Everything then turns on the nature of the knot—the nexus—that binds the legitimacy claims raised by the governors to the belief in that authority on the part of the governed. The paradox of authority resides in this knot. Ideology, we may presume, arises precisely in the breach between the request for legitimacy emanating from a system of authority and our response in terms of belief. Ideology is supposed to add a sort of surplus value to our spontaneous belief, thanks to which the latter might satisfy the demands of the authority. At this stage, the function of ideology would be to fill the gap of credibility opened by all systems of authority, not only the charismatic system—because the chief is sent from above—and the system based on tradition—because things have always been done this way—but also the bureaucratic system—because the experts are supposed to know. Max Weber provides support for the present hypothesis by defining the types of legitimacy, their imperatives, and their requirements on the basis of the types of belief “by virtue of which” the order is legitimized, and power justified. These types of belief constitute, each in its own way, a reason for obeying. Moreover, it is in this way that authority is defined as the legitimate power to make (others) obey. Herrschaft, according to Weber, consists essentially in a hierarchical relation between commanding and obeying. It is expressly defined by the expectation of obedience and the probability—the likelihood—that this will be fulfilled. It is at this critical point that the symbolic systems and their rhetorical expressions discussed elsewhere by Clifford Geertz are seen to be mobilized. They supply the selling points that elevate ideology to the level of a surplus value added to the belief in the legitimacy of power.³⁷

This relation between ideology and the legitimation process of systems of authority seems to me to constitute the central axis in relation to which are organized, on the one hand, the more radical phenomenon of community integration on the basis of the symbolic—even rhetorical—mediations of action, and on the other hand, the more visible phenomenon, easier to deplore and to denounce, of the distortion-effect on which Marx focused his best analyses in The German Ideology.³⁸ We are familiar with the debatable metaphors of the inverted image or of man standing on his head. The mechanism of distortion, itself cast into images in this way, would be plausible only if it were joined to the phenomenon of legitimation, which I placed at the center of the ideological apparatus, and if this mechanism were, in the final analysis, to affect the unavoidable symbolic mediations of action. Brushing aside these mediations, the detractor of ideology assumes he is capable of giving a true, unwarped description of fundamental human reality—namely, praxis, transforming activity—hence a description free from any interpretation in terms of meaning, value, or norm. This realism, even ontology, of praxis³⁹ and, more precisely, of living labor⁴⁰ are at once the strength and the weakness of the Marxian theory of ideology. If, in fact, praxis does not, primordially, incorporate an ideological layer, in the primary sense of the word “ideology,” then I cannot see what in this praxis could provide the material for distortion. Separated from this original symbolic context, the denunciation of ideology is reduced to a pamphlet against propaganda. This enterprise of purification is not vain; it can possess circumstantial necessity, if it is conducted from the perspective of the reconstruction of a public space of discussion and not from the viewpoint of a merciless struggle, whose horizon would be circumscribed by civil war.⁴¹

If this analysis is plausible, or even accurate, we easily see which springs are working the various efforts to manipulate memory.

It is easy to trace them back to the respective operative levels of ideology. On the deepest level, that of the symbolic mediation of action, it is through the narrative function that memory is incorporated into the formation of identity. Memory can be ideologized through the resources of the variations offered by the work of narrative configuration. And, as the characters of the narrative are emplotted at the same time the story is told, the narrative configuration contributes to modeling the identity of the protagonists of the action as it molds the contours of the action itself. The narrative, Hannah Arendt reminds us, recounts the “who of action.” It is, more precisely, the selective function of the narrative that opens to manipulation the opportunity and the means of a clever strategy, consisting from the outset in a strategy of forgetting as much as in a strategy of remembering. We will account for this in the thematic study reserved for forgetting. However, it is on the level where ideology operates as a discourse justifying power, domination, that the resources of manipulation provided by narrative are mobilized. Domination, we have understood, is not limited to physical constraint. Even the tyrant needs a rhetorician, a sophist, to broadcast his enterprise of seduction and intimidation in the form of words. The narrative imposed in this way then becomes the privileged instrument of this twofold operation. Even the surplus value that ideology adds to the belief offered by the governed in responding to the claim of legitimacy made by the governing body presents a narrative texture: stories of founding events, of glory and humiliation, feed the discourse of flattery or of fear. It thus becomes possible to account for the express abuses of memory on the level of the effect of distortion belonging to the phenomenal level of ideology. At this level of appearance, imposed memory is armed with a history that is itself “authorized,” the official history, the history publicly learned and celebrated. A trained memory is, in fact, on the institutional plane an instructed memory; forced memorization is thus enlisted in the service of the remembrance of those events belonging to the common history that are held to be remarkable, even founding, with respect to the common identity. The circumscription of the narrative is thus placed in the service of the circumscription of the identity defining the community. A history taught, a history learned, but also a history celebrated. To this forced memorization are added the customary commemorations. A formidable pact is concluded in this way between remembrance, memorization, and commemoration.

We are touching here upon the precise abuses denounced by Tzvetan Todorov in his essay appropriately titled Les Abus de la mémoire,⁴² in which we can read a stern indictment of the contemporary frenzy of commemorations, with their parade of rites and myths, ordinarily tied to the founding events referred to above. This grip on memory, Todorov insists, is not a specialty of totalitarian regimes alone; it is the apanage of all those enamored of glory. This denunciation sounds a warning against what the author terms the “unconditional praise of memory” (13). “The stakes of memory,” he adds, “are too great to be left to enthusiasm or to anger” (14). I will not dwell upon a further aspect of the problem, namely, the claim of our contemporaries to place themselves in the position of victim, to assume the status of victim: “To have been a victim gives you the right to complain, to protest, and to make demands” (56). This position engenders an exorbitant privilege, which places everyone else in the position of owing a debt. I will retain instead a final remark by Todorov, that will bring us to the difficult question of the duty of memory: “The work of the historian, like every work on the past, never consists solely in establishing the facts but also in choosing certain among them as being more salient and more significant than others, then placing them in relation to one another; now this work of selecting and combining is necessarily guided by the search, not for truth, but for the good” (50). Whatever our reservations may be regarding the alternative suggested here between truth and goodness, we must hold in reserve, in light of our later discussion of the duty of memory, the reorientation of the whole issue of the abuse of memory under the auspices of the search for justice. This concern links up with the preceding discussion thanks to a most judicious piece of advice from Todorov, to extract from traumatic memories the exemplary value that can become pertinent only when memory has been turned into a project. If the trauma refers to the past, the exemplary value is directed toward the future. What the cult of memory for the sake of memory obliterates is, along with the aim of the future, the question of the end, of the moral issue. For the very notion of use, implicit in that of abuse, is unavoidably related to this question of the end. And with the question of the end we have already crossed over the threshold into the third level of our investigation.

The Ethico-Political Level: Obligated Memory

What about, I will ask in conclusion, the alleged duty of memory? In truth, the question is rather premature in view of the distance our thinking has yet to cover. It projects us well beyond a simple phenomenology of memory, and even beyond an epistemology of history, to the heart of the hermeneutics of the historical condition. We cannot, it is true, abstract from the historical conditions in which the duty of memory is required, namely, those of Western Europe and, in particular, France, several decades after the horrible events of the mid-twentieth century. The injunction is meaningful only in relation to the difficulty experienced by the national community or by the wounded parts of the body politic to remember (faire mémoire de) these events calmly. These difficulties cannot be discussed responsibly until we have crossed through the arid plains of the epistemology of historical knowledge and entered the region of conflicts among individual memory, collective memory, and historical memory, at the point where the living memory of survivors confronts the distantiated, critical gaze of the historian, to say nothing of the viewpoint of the judge.

It is at this point of friction that the duty of memory proves to be particularly equivocal. The injunction to remember risks being heard as an invitation addressed to memory to short-circuit the work of history. For my part, I am all the more attentive to this danger as this book is a plea on behalf of memory as the womb of history, inasmuch as memory remains the guardian of the entire problem of the representative relation of the present to the past. The temptation then is great to transform this plea into a claim on behalf of memory in opposition to history. Just as I shall resist, when the time comes, the inverse claim to reduce memory to a simple object of history among its “new objects,” at the risk of stripping it of its function of matrix, so too shall I refuse to allow myself to be enlisted into making the inverse plea. It is in this frame of mind that I have chosen to pose the question of the duty of memory for the first time in the context of the uses and abuses of memory, prepared to return to it at greater length in the context of forgetting. To say: you will remember, is also to say: you will not forget. It may even be that the duty of memory constitutes, at one and the same time, the epitome of good use and of abuse in the exercise of memory.

Let us wonder first at the grammatical paradox that is formed by the injunction to remember. How is it possible to say: “you will remember,” hence to employ the future tense to speak of this memory that is given as the guardian of the past? More seriously: how can it be permissible to say: “you must remember,” hence speak of memory in the imperative mood, although it is characteristic of memory to emerge as a spontaneous evocation, hence as pathos, according to Aristotle’s De memoria? How can this prospective movement of the mind, turned toward memory as a task to be accomplished, be joined to the two dispositions left as though in suspense—the work of memory and the work of mourning, taken in turn separately and as a pair? In a certain manner, it extends their prospective character. But what does it add to this?

It is true that within the precise framework of the therapeutic cure the duty of memory is formulated as a task: it marks the will of the analysand to contribute from then on to the joint undertaking of the analysis as it navigates through the pitfalls of transference. This will even adopts the form of the imperative, allowing the representatives of the unconscious to speak and, in this way, as far as is possible, to “tell all.” We should reread in this regard the advice that Freud gives to analyst and to analysand alike in the course of his essay “Remembering, Repeating, and Working-Through.” For its part, the work of mourning, since it requires time, projects the artisan of this work ahead of himself: he will have to continue, one by one, to cut the ties that hold him in the grip of the lost objects of his love and his hate; as for reconciliation with the loss itself, this will forever remain an unfinished task. This patience toward oneself even possesses the features of a virtue, if it is contrasted, as we have tried to do, to the vice of giving in to sadness, to the acedia of spiritual teachers, that hidden passion that drags melancholia ever downward.

Having said this, what is missing from the work of memory and from the work of mourning that would make them equivalent to the duty of memory? What is missing is the imperative element that is not expressly present in the notion of work: work of memory, work of mourning. More precisely, what is still absent is the twofold aspect of duty, as imposing itself on desire from outside and as exerting a constraint experienced subjectively as obligation. Where are these two features found together, in a form least subject to dispute, if not in the idea of justice, which we have already mentioned above in reply to the abuses of memory on the level of manipulation? Extracting the exemplary value from traumatic memories, it is justice that turns memory into a project; and it is this same project of justice that gives the form of the future and of the imperative to the duty of memory. We can then suggest that the duty of memory considered as the imperative of justice is projected as a third term onto the point of intersection of the work of mourning and the work of memory. In return, the imperative receives from the work of memory and the work of mourning the impetus that integrates it into an economy of drives. This united force of the duty of justice can then extend beyond the memory and mourning pair to the pair formed by the truthful and the pragmatic dimensions of memory; indeed, our own discourse on memory has been conducted up to now along two parallel lines, the line of memory’s concern for truth, under the aegis of the epistemic fidelity of memories with respect to what actually took place, and the line of memory use, considered as practice, even as the technique of memorization. This marks the return, therefore, of the past and the exercise of the past, this bi-partition repeating the division of the two chapters of Aristotle’s treatise. It is as though the duty of memory were projected ahead of consciousness as a point of convergence between the truth perspective and the pragmatic perspective on memory.

The question then arises as to what gives the idea of justice its federating force with regard to the truthful and pragmatic aims of memory as well as to the work of memory and the work of mourning. It is thus the relation of the duty of memory to the idea of justice that must be interrogated.

First element of a response: it must be recalled, first, that among all the virtues, the virtue of justice is the one that, par excellence and by its very constitution, is turned toward others. We can even say that justice is the component of otherness inherent in all the virtues that it wrests from the closed-circuit of the self with itself. The duty of memory is the duty to do justice, through memories, to an other than the self.⁴³

Second element of a response: the time has come to introduce a new concept—debt, which must not be limited to the concept of guilt. The idea of debt is inseparable from the notion of heritage. We are indebted to those who have gone before us for part of what we are. The duty of memory is not restricted to preserving the material trace, whether scriptural or other, of past events, but maintains the feeling of being obligated with respect to these others, of whom we shall later say, not that they are no more, but that they were. Pay the debt, I shall say, but also inventory the heritage.

Third element of a response: among those others to whom we are indebted, the moral priority belongs to the victims. Todorov cautioned above against the tendency to proclaim oneself a victim and endlessly to demand reparation. He was right. The victim at issue here is the other victim, other than ourselves.

This being the legitimation of the duty of memory as a duty of justice, how are abuses grafted upon its proper use? These abuses can occur only through the manner in which the idea of justice is handled. It is here that a certain demand raised by impassioned memories, wounded memories, against the vaster and more critical aim of history, lends a threatening tone to the proclamation of the duty of memory, which finds its most blatant expression in the exhortation to commemorate now and always.

Anticipating further developments that rely upon a more advanced state of the dialectic of memory and history, I want to indicate the existence of two clearly distinct, yet compatible, interpretations of this slippage from use to abuse.

On one side, the emphasis can be placed on the regressive character of the abuse, which carries us back to the first stage of our inquiry into the uses and abuses of memory under the sign of thwarted memory. This is the explanation proposed by Henry Rousso in The Vichy Syndrome: History and Memory in France since 1944.⁴⁴ This explanation is of value only within the limits of present-day history, so over a relatively short time-span. Rousso makes the best possible use of categories belonging to a pathology of memory—trauma, repression, return of the repressed, obsession, exorcism. Within this conceptual framework, which draws its legitimacy from its heuristic effectiveness, the duty of memory functions like an attempted exorcism in a historical situation marked by the obsession with traumas suffered by the French in the years 1940–45. The extent to which the proclamation of the duty of memory remains captive to the symptom of obsession makes it waver continually between use and abuse. Yes, the way in which the duty of memory is proclaimed can take the form of an abuse of memory in the manner of the abuses denounced earlier under the heading of manipulated memory. To be sure, these are no longer manipulations in the sense defined in terms of the ideological relation of the discourse of power, but in a more subtle manner in the sense of an appeal to conscience that proclaims itself to be speaking for the victims’ demand for justice. Inveigling the silent word of the victims in this way makes use turn to abuse. We should not be surprised to find again on this somewhat higher level of obligated memory the same signs of abuse recognized in the preceding section, principally in the form of the frenzy of commemoration. The concept of obsession will be treated in a thematic manner at a later stage of this work in the chapter on forgetting.

Pierre Nora proposes an explanation less centered on the recitative of the history of the present day in the article that concludes the third volume of Les Lieux de mémoire—Les France—under `the title: “L’Ère de la commémoration.”⁴⁵ The article is devoted to “obsession with commemoration” (609) and can be understood only in terms of the dialogue the author undertakes with the inaugural text of “the places of memory.” I shall, at the appropriate moment, devote a study to Nora’s dialogue with himself.⁴⁶ If I mention it now, it is to extract a warning from it concerning the assimilation of my own work to an attack on history in the name of memory. Nora himself complains of a similar assimilation of the theme of the “places of memory” by “commemorative bulimia [that] has all but consumed all efforts to control it” (609): “The destiny of these Lieux de mémoire has been a strange one. The work was intended, by virtue of its conception, method, and even title, to be a counter-commemorative type of history, but commemoration has overtaken it. . . . What was forged as a tool for maintaining critical distance became the instrument of commemoration par excellence” (609). There is a historical moment, our own, that now defines itself completely in terms of “the obsession with commemoration”: May 1968; the bicentennial of the French Revolution, and so on. The explanation proposed by Nora does not yet concern us, just his diagnosis: “Thus the very dynamics of commemoration have been turned around; the memorial model has triumphed over the historical model and ushered in a new, unpredictable, and capricious use of the past” (618). For what historical model has the memorial model been substituted? For the model of celebrations devoted to the impersonal sovereignty of the nation-state. This model deserved to be called historical because French self-understanding was identified with the history of the establishment of the nation-state. What has been substituted for it are particular, fragmented, local, and cultural memories.⁴⁷ What claim is attached to this inversion of the historical into the commemorative? What interests me here has to do with the transition from the phenomenology of memory to the epistemology of scientific history. The latter, Nora tells us, “consisted in the rectification and enrichment of the history of memory. Although it was intended to be ‘critical,’ it was in fact only a deepening of that tradition. Its ultimate goal was identification through filiation. It was in this sense that history and memory were identical: history was verified memory” (626). The inversion that is at the origin of commemorative obsession is said to consist in the assimilation of defunct traditions, slices of the past from which we have become separated. In short, “commemoration has freed itself from its traditionally assigned place, but the epoch as a whole has become commemorative” (627).

I would like to state at the close of this chapter devoted to the practice of memory that my undertaking is not a part of this “enthusiasm for memorial commemoration” (629). If it is true that “memory’s moment” (632) defines an era, our own, my work aspires to escape the criteria that define this era, whether this be in its phenomenological phase, its epistemological phase, or its hermeneutical phase. Rightly or wrongly. For this reason my work is not threatened but reassured by Pierre Nora’s conclusion, announcing a time when “the era of commemoration will be over for good” (637). For it is not to “the tyranny of memory” (637) that it will have desired to contribute. This abuse of abuses is among those it denounces with the same vigor with which it resists the substitution of the duty of memory for the work of mourning and the work of memory, and it limits itself to placing both of these labors under the sign of the idea of justice.

The question posed by the duty of memory, therefore, exceeds the limits of a simple phenomenology of memory. It even outstrips the resources of intelligibility of an epistemology of historical knowledge. Finally, as an imperative of justice, the duty of memory belongs to a moral problematic the present work just begins to approach. A second partial evocation of the duty of memory will be proposed within the framework of my meditation on forgetting, in relation to an eventual right of forgetting. We will then be confronted with the delicate connection between the discourse of memory and forgetting and the discourse of guilt and forgiveness.

In this state of suspense, we interrupt our examination of the exercised memory, its exploits, its uses and its abuses.