[In the following essay, Michaels uses the example of the treatment of the Holocaust by American academics as an example of the importance of upholding cultural myths.]

DO THE AMERICANS BELIEVE THEIR MYTHS? OR, BELOVED

The title of this section is derived from Paul Veyne's Did the Greeks Believe in Their Myths?—a book that I read several years ago, first with great eagerness and then with a certain disappointment. The eagerness stemmed from my curiosity about whether the Greeks really thought, to take one of Veyne's examples, that events like “the amorous adventures of Aphrodite and Ares caught in bed by her husband” had actually happened;¹ the disappointment stemmed from Veyne's commitment to regarding such curiosity as naive. The book's subtitle is An Essay on the Constitutive Imagination; Veyne thinks that “it is we who fabricate our truths, and it is not ‘reality’ that makes us believe” (113); hence, truth is “plural” and myth, like “history,” “literature,” and “physics,” is “true in its way” (20). So, “of course,” the Greeks “believed in their myths,” and “what is true of ‘them’ is also true of ourselves” (129). The answer to the question, Do the Americans believe their myths? is yes.

At the same time, however, from Veyne's perspective, the answer to the question, Do the Americans believe their myths? must be no. Once we recognize that, as he puts it, “‘reality’ is the child of the constitutive imagination of our tribe” (113), we must also recognize that “truth,” as we ordinarily conceive it—in the sense, say, that we might think our myths true and the Greek myths false—“does not exist” (115). Indeed, “as long as we speak of the truth, we will understand nothing of culture and will never manage to attain the same perspective on our culture as we have on past centuries, when people spoke of gods and myths” (113). So if, on the one hand, we must believe our myths, on the other hand, we must not believe that they are true. That is the whole point of “culture” as Veyne understands it: “Culture, without being false, is not true either” (127). And while the Greeks, “of course,” believed their myths, insofar as in believing their myths they believed them to be true, they were, of course, mistaken. We, who know that our culture is neither true nor false, also believe our myths, but we believe them in the right way; in fact, insofar as “our perspective” on our own culture is “the same” as our perspective on the cultures of “past centuries, when people spoke of gods and myths,” we must not only believe our own myths, we must also believe the myths of the past. So not only must the Greeks have believed their myths and must the Americans believe theirs, the Americans must believe the Greek myths, too.

And, in fact, at least some Americans do. In his 1987 bestseller, Communion, Whitley Strieber argues that the alien “visitors” who on several occasions have made their presence known to him and who look, he thinks, like the ancient goddess Ishtar, are probably the originals for “the whole Greek pantheon.” His theory is that humans, unable to deal with “the stark reality of the visitor experience”—“the bad smells, the dreadful food, and the general sense of helplessness”—dress it up in what he calls “a very human mythology,” one that preserves the essential truth of “the visitor experience” while at the same time making it more palatable. But Communion is subtitled A True Story rather than An Essay on the Constitutive Imagination; whether or not the Greeks, in believing their myths, believed them to be true, Whitley Strieber does. His memories of his own experience count as testimony to their truth not only because they provide modern analogies for ancient myths but also because they may be understood to provide more direct evidence: “Do my memories come from my own life,” he wonders, “or from other lives lived long ago, in the shadowy temples where the grey goddess reigned?”² If they come from his own life, they provide evidence that godlike creatures are currently interacting with humans, and they provoke the reflection that such interactions may have taken place also in the past; if they come from lives lived long ago, they provide evidence that godlike creatures have always interacted with humans and so that the old mythologies are not only compatible with recent experience but true.

But how does the fact that some Americans believe the Greek myths shed any light on the question of whether Americans believe their own myths? It might, of course, be argued that the belief in “visitors” is an American myth and so that, for people like Whitley Strieber, believing in the Greek myths is a way of believing American myths. In my view, however, the fact that Whitley Strieber believes in the Greek pantheon is less relevant to American mythology than the question he raises in the course of stating that belief: “Do my memories come from my own life or from other lives lived long ago?” For it is this question, I want to suggest, that lies at the heart of the myths Americans believe insofar as it is in attempting to answer this question—Do our memories come from our own lives or from other lives lived long ago?—that Americans can come to think of themselves as distinctively American.

“History is to the nation,” Arthur Schlesinger Jr. has recently written in The Disuniting of America, “rather as memory is to the individual. As an individual deprived of memory becomes disoriented and lost … so a nation denied a conception of its past will be disabled in dealing with its present … As the means of defining national identity, history becomes a means of shaping history.”³ Memory is here said to constitute the core of individual identity; national memory is understood to constitute the core of national identity. Insofar, then, as individuals have a national as well as an individual identity, they must have access not only to their own memories but to the national memory; they must be able to remember not only the things that happened to them as individuals but the things that happened to them as Americans. The way they can do this, Schlesinger says, is through history. History, in other words, can give us memories not only of what Strieber calls our “own” lives but of “other lives lived long ago.” And it is in giving us these memories that history gives us our “identity.” Indeed, it is because our relation to things that happened to and were done by Americans long ago is the relation of memory that we know we are Americans. We learn about other people's history; we remember our own.

So American mythology has less (although, as we shall see, not nothing) to do with the belief in aliens (space aliens, anyway) than with the belief that we can remember “other lives lived long ago”—or perhaps, to put the point more neutrally, than with our ways of talking as if we remembered “other lives lived long ago.” For whether or not the belief that we can remember such lives is widespread, talk about remembering such lives is extremely widespread. To stick for another moment to texts that may, to an academic audience, seem marginal, Greg Bear's science fiction novel Blood Music (1985) imagines the restructuring of blood cells so as to enable them to perform a kind of memory transfer, first from father to son (“The memory was there and he hadn't even been born, and he was seeing it, and then seeing their wedding night”) and then more generally (“And his father went off to war … and his son watched what he could not possibly have seen. And then he watched what his father could not possibly have seen”). “Where did they come from?” he asks about these memories, and when he is told, “Not all memory comes from an individual's life,” he realizes that what he is encountering is “the transfer of racial memory” and that now, in “his blood, his flesh, he carried … part of his father and mother, parts of people he had never known, people perhaps thousands of years dead.”⁴Blood Music imagines as science what Communion—identifying its “visitors” with the “Greek pantheon” and speculating that they are the “gods” who created us—imagines as religion.

But both Blood Music and Communion should probably, as I suggested above, be considered marginal texts, not because they haven't been read by many—Communion, at least, has been read by hundreds of thousands—but because their account of what Blood Music calls “racial memory” is, in a certain sense, significantly anachronistic. By “racial,” Greg Bear means “human”; it's the human race, not the white or the black or the red race, that his transfusions of blood unite. And while it is true that, in an amazing moment, Whitley Strieber speaks of “visitor culture” and imagines our encounter with it along vaguely multicultural lines—it may be only “apparently superior”; we will come to understand “its truth” by understanding its “weaknesses” as well as its “strengths”—it is essential to remember that the “visitors” he has in mind are not merely foreigners.⁵ Strieber does produce the familiar nativist gesture of imagining himself a Native American, the “flower” of his “culture” crushed by “Cortez”-like invaders, but the vanishing race for which he is proleptically nostalgic is, like Greg Bear's, human rather than American. It would only make sense to understand Communion's aliens as relevant to the question of American identity if we were to understand them as allegories of the aliens threatening American identity. Insofar, however, as the apparatus of the allegory requires the redescription of differences between humans as differences between humans and others, it has the effect of establishing the human as an internally undifferentiated category and thus of making the designation of some humans as American irrelevant. In Communion and Blood Music, the emergence of “racial memory,” of a history made almost literally universal, unites us all.

So the technologies of memory imagined in Blood Music and Communion provide an image, but only a partial image, of what is required by Schlesinger's invocation of history as memory. If the obvious objection to thinking of history as a kind of memory is that things we are said to remember are things that we did or experienced, whereas things that are said to belong to our history tend to be things that were neither done nor experienced by us, then Blood Music and Communion imagine ways in which history can be turned into memory. But they don't meet Schlesinger's requirement that this history be national, which is to say that they don't deploy the transformation of history into memory on behalf of the constitution of identity; in Communion, the remembered past is merely a testament to the visitors' persistence; in Blood Music, the moment in which the past can be remembered actually marks the disappearance of nationality. It is instead in a much more important and influential text of 1987, Toni Morrison's Beloved, that Schlesinger's identification of memory, history, and national identity is given a definitive articulation.

And this is true even though Beloved, according to Morrison, is a story about something no one wants to remember: “The characters don't want to remember, I don't want to remember, black people don't want to remember, white people don't want to remember.”⁶ What no one wants to remember, she thinks, is slavery, and whether or not this characterization is accurate, it succeeds in establishing remembering or forgetting as the relevant alternatives. It establishes, in other words, that although no white people or black people now living ever experienced it, slavery can be and must be either remembered or forgotten. Thus, although Sethe's daughter Denver thinks early on that “only those who lived in Sweet Home [where Sethe was a slave] could remember it,” it quickly turns out that memories of “places” like Sweet Home can in fact be made available to people who never lived there.⁷ A “house can burn down,” Sethe tells Denver, “but the place—the picture of it—stays, and not just in my rememory, but out there, in the world” (36). Thus, people always run the risk of bumping into “a rememory that belongs to somebody else,” and thus, especially, Denver runs the risk of a return to slavery: “The picture is still there and what's more, if you go there—you who never was there—if you go there and stand in the place where it was, it will happen again; it will be there for you, waiting for you.” Because Denver might bump into Sethe's rememory, Sethe's memory can become Denver's; because what once happened is still happening—because, as Denver says, “nothing ever dies”—slavery needn't be part of your memory in order to be remembered by you.

From Sethe's standpoint this is, of course, a kind of threat; she and her contemporaries are, as one critic has put it, “haunted by memories of slavery that they seek to avoid.”⁸ But if Beloved's characters want to forget something that happened to them, its readers—“black people,” “white people,” Morrison herself—are supposed to remember something that didn't happen to them. And in insisting on slavery as the thing they are supposed to remember, Morrison not only gives Blood Music's “racial memory” what counts in the late twentieth century as its proper meaning, but she also establishes what we might call, in contrast to the marginality of Blood Music, the centrality of Beloved—or at least its discursive distance from the genres of science fiction and New Age space invasion. This distance involves, as we have already noted, the political difference between a certain universalism and a certain nationalism, but in fact it's much greater than that. Morrison's race, like Schlesinger's nation, provides the mechanism for as well as the meaning of the conversion of history into memory. Blood Music requires weird science to explain how people can “remember stuff” they haven't “even lived through”;⁹The Disuniting of America needs only the nation, Beloved needs only race. And while probably almost no Americans now believe that blood transfusions can make us remember things that did not happen to us, and probably only some Americans now believe that “visitors” can help us remember the lives we lived long ago, probably a great many Americans believe that nationality—understood by Schlesinger as citizenship in a state, transformed by Morrison and by multiculturalism more generally into membership in a race or culture—can do what blood transfusions and visitors cannot. It is racial identity that makes the experience of enslavement part of the history of African Americans today.¹⁰

So if some Americans believe the Greek myths—believe, that is, in the non-human creatures who used to be called “gods” and are now called “visitors”—most Americans need only believe in our own myths: race and nation. The supernatural presence that haunts 124 Bluestone Road outside of Cincinnati, Ohio, should not be understood as a version of the supernatural presence that haunts Whitley Strieber's cabin in upstate New York. At the same time, however, it is a striking fact about Beloved that it presents itself as a ghost story, that its account of the past takes the form of an encounter with a ghost, a ghost who is, as Valerie Smith has said, “the story of the past embodied.”¹¹ And if one way to regard this ghost is (along the lines I've just suggested) as a figure for the way in which race can make the past present, another way to regard the ghost is as the figure for a certain anxiety about the very idea of race that is being called upon to perform this function. For while races are, no doubt, more real than “visitors,” it isn't quite clear how much more or in what ways they are more real. To what extent, for example, are the races we believe in biological entities? Nothing is more common in American intellectual life today than the denial that racial identity is a biological phenomenon and the denunciation of such a biologism as racial essentialism. The race that antiessentialists believe in is a historical entity, not a biological one. In racial antiessentialism, the effort to imagine an identity that will connect people through history is replaced by the effort to imagine a history that will give people an identity.

If, then, we must not see the ghost in Beloved as a real (albeit biologically exotic) entity (like a visitor), we should not see her either as a figure for a real (and also biologically exotic) entity (like a race). She is a figure instead for a process, for history itself; Beloved is, in this respect, not only a historical but a historicist novel. It is historical in that it's about the historical past; it is historicist in that—setting out to remember “the disremembered”—it redescribes something we have never known as something we have forgotten and thus makes the historical past a part of our own experience. It's no accident that the year in which Communion and Beloved were published (1987) is also the year in which the University of California Press New Historicism series was inaugurated, or that the year in which Beloved won the Pulitzer Prize (1988) was the year in which Shakespearean Negotiations—written by the editor of the New Historicism series, Stephen Greenblatt, and beginning with the author's announcement of his “desire to speak with the dead”—was also published. The ghost story, the story in which the dead speak, either like Beloved in a voice that's “low and rough” (52) or like Shakespeare through “textual traces,” is the privileged form of the New Historicism.¹²

If, in other words, the minimal condition of the historian's activity is an interest in the past as an object of study, Greenblatt's account of the origins of his vocation (“I began with the desire to speak with the dead”) and his account of the nature of that vocation (“literature professors are salaried, middle-class shamans”) both insist on a relation to the past (he calls it a “link”) that goes beyond that minimal condition, and beyond also (it's this going beyond that the model of the shaman is meant to indicate) various standard accounts of the continuity between past and present. Greenblatt is not, that is, interested in the kind of continuity offered by the claim that events in the past have caused conditions in the present, or in the kind of continuity imagined in the idea that the past is enough like the present that we might learn from the past things that are useful in the present.¹³ Indeed, the interest proclaimed here has almost nothing to do with taking the past as an object of knowledge; what he wants is to speak with the dead, “to re-create a conversation with them,” not to find out or explain what they did. And although he himself proclaims this ambition a failed one, from the standpoint of the heightened continuity that the New Historicism requires, the terms of failure are even more satisfying than success would be: “Even when I came to understand that in my most intense moments of straining to listen all I could hear was my own voice, even then I did not abandon my desire. It was true that I could hear only my own voice, but my own voice was the voice of the dead.”¹⁴ If what you want is a “link” with the dead that is better achieved by speaking with them than by studying them—that is achieved, that is to say, by understanding studying them as a way of speaking with them—then it can't really count as a disappointment to discover that what one hears when one hears the dead speak is actually the sound of one's “own voice.” “My own voice was the voice of the dead”: the link envisioned in conversation is only made stronger by the discovery that the conversation is with oneself. Continuity is turned into identity.

For both Morrison and Greenblatt, then, history involves the effort to make the past present, and the ghosts of Beloved and Shakespearean Negotiations are the figures for this effort, the transformation of history into memory, the deployment of history in the constitution of identity. If, then, we ask a slightly revised version of the question whether the Americans believe their myths—Which myths do the Americans believe?—the answer turns out to be not visitors, not blood transfusions, not biological races, not even exactly history as such, but history as memory. To put the point in this way is no longer to say with Veyne that the difference between myth and history is erased insofar as the truths of both myth and history are revealed as truths constituted by the imagination. For although this idealism is, as we all know, widespread today, and although it does succeed in establishing, at least by the back door—we don't get our identity from history, history gets its identity from us—the desired link between past and present, the fact that that link must be imposed on the past before it can be derived from it makes it less promising as a ground of identity; if we create our history, then any history might be made ours. So what makes our commitment to history a commitment to myth is not our sense that the history we learn is true in (and only in) the same way that the Greeks thought their myths were true; what makes our history mythological is not our sense that it is constituted but our sense that it is remembered and, when it is not remembered, forgotten.¹⁵

Without the idea of a history that is remembered or forgotten—not merely learned or unlearned—the events of the past can have only a limited relevance to the present, providing us at most with causal accounts of how things have come to be the way they are, at least with objects of antiquarian interest. It is only when it's reimagined as the fabric of our own experience that the past can become the key to our own identity. A history that is learned can be learned by anyone, and it can belong to anyone who learns it; a history that is remembered can only be remembered by those who first experienced it, and it must belong to them. So if history were learned, not remembered, then no history could be more truly ours than any other. Indeed, no history, except the things that had actually happened to us, would be truly ours at all.

This is why the ghosts of the New Historicism are not simply figures for history, they are figures for a remembered history. But this is also why there is a problem in thinking about these ghosts as figures. For without the ghosts to function as partners in conversation rather than as objects of study, without rememories that allow “you who never was there” access to experiences otherwise available to “only those who” were there, history can no more be remembered than it can be forgotten. The ghosts cannot, in other words, be explained as metaphoric representations of the importance to us of our history because the history cannot count as ours and thus can have no particular importance to us without the ghosts. It is only when the events of the past can be imagined not only to have consequences for the present but to live on in the present that they can become part of our experience and can testify to who we are. So the ghosts are not merely the figures for history as memory, they are the technology for history as memory: to have the history, we have to have the ghosts. Remembered history is not merely described or represented by the ghosts who make the past ours, it is made possible by them. Beloved's ghosts are thus as essential to its historicism as Communion's visitors are to its New Age mysticism; indeed, Beloved's historicism is nothing but the racialized and, hence, authorized version of Communion's mysticism. Without the visitors, the remains of UFOs are just fragments of old weather balloons; without the ghosts, history is just a subject we study.¹⁶ It is only accounts like Sethe's of how other people's memories can become our own that provide the apparatus through which our history can, as Arthur Schlesinger puts it, define our identity.

HOLOCAUST REMEMBRANCE; OR, MARION

“As the eyewitnesses pass from the scene and even the most faithful memories fade,” Geoffrey Hartman has recently written, “the question of what sustains Jewish identity is raised with a new urgency.”¹⁷ What the eyewitnesses witnessed, and what they have begun to forget, is, of course, the Holocaust; and although the dependence of Jewish identity on the Holocaust is a topic that will require further discussion, the dependence of identity on memory is, as we have already seen, undeniable. So if the idea here is that memories of the Holocaust have sustained Jewish identity, and thus that the imminent disappearance of those memories as even the survivors die poses a threat to Jewish identity, then the task becomes to keep those memories alive if only in order to keep Jewish identity alive. Because some survivors are still alive, the issue here is not yet the transformation of history into memory that we saw at work in Beloved's representation of American slavery; it is instead an effort to forestall what Pierre Vidal-Nacquet has called the “transformation of memory into history.”¹⁸ Once memory is transformed into history, it can no longer be relevant to the project of sustaining Jewish identity. Hence, the outbreak, over the last fifteen years, of interest in the Holocaust must be understood less as a response to the idea that people will cease to know about the Holocaust than as a response to the idea that they will cease to remember it.

Indeed, there is, precisely from this perspective, a certain hostility to the idea that the Holocaust is the sort of thing that can be known. Claude Lanzmann, the maker of Shoah, has insisted that “the purpose of Shoah is not to transmit knowledge” and has instead characterized the film as “an incarnation, a resurrection,”¹⁹ thus identifying the ambitions of Shoah in terms that we may understand as characteristically New Historicist: the incarnated dead are the ones with whom Stephen Greenblatt wishes to speak. But whereas in the New Historicism understanding the past is at worst an irrelevance and at best an aid to remembering it, understanding the Holocaust seems to Lanzmann an “absolute obscenity,” and to try to “learn the Holocaust” is in fact to “forget” it.²⁰ The representations and explanations of historians, he thinks, are “a way of escaping,” “a way not to face the horror”;²¹ what the Holocaust requires is a way of transmitting not the normalizing knowledge of the horror but the horror itself. And it is this “transmission”—what Shoshana Felman calls “testimony”—that the film Shoah strives for and that, according to Felman, is the project of the major literary and theoretical texts of the post-World War II period.

But how can texts transmit rather than merely represent “horror”? How, as Felman puts it, can “the act of reading literary texts” be “related to the act of facing horror?” (2). If it could, then, of course, reading would become a form of witnessing. But it is one thing, it seems, to experience horror and another thing to read about it; the person who reads about it is dealing not with the experience of horror but with a representation of that experience. And Felman has no wish to deny this difference; on the contrary, she wishes to insist upon it, and it is out of her insistence that she produces her contribution to the theory of testimony. For when testimony is “simply relayed, repeated or reported,” she argues, it “loses its function as a testimony” (3). So in order for testimony to avoid losing its proper function, it must be “performative” (5); it must “accomplish a speech act” rather than simply “formulate a statement.” Its subject matter must be “enacted” rather than reported or represented. The problem of testimony is thus fundamentally a problem about “the relation between language and events” (16). Language that represents or reports events will fail as testimony—will fail, that is, to be properly “performative” or “literary.” Language that is itself an “act” and that therefore can be said to “enact” rather than report events will succeed. The reader of the “performative” text will be in the position not of someone who reads about the “horror” and understands it; he or she will be in the position of “facing horror.”

But how can a text achieve the performative? How can a text cease merely to represent an act and instead become the act it no longer represents? The idea of the performative is, of course, drawn from Austin's speech-act theory, where it is famously instantiated in the marriage ceremony: “When I say, before the registrar or altar, etc., ‘I do,’ I am not reporting on a marriage: I am indulging in it.”²² Austin's opposition between reporting and indulging anticipates (in a different key) Felman's opposition between reporting and enacting. But in Felman's Testimony, the first exemplar of the performative will not exactly be, as it is in Austin, the act of “saying certain words” (13); it will instead be what Felman calls the “breakdown” or “breakage of the words” (39). Citing these lines from Celan,

Your question—your answer.
Your song, what does it know?
Deepinsnow,
                                                            Eepinow,
                                                                                                                        Ee-i-o.

she argues that it is “by disrupting” “conscious meaning” that these “sounds testify” (37). It is, in other words, at the moment when the words as words begin to “break down” that they become performative, that they begin to enact rather than report. And it is at this moment that the readers of those words are “ready to be solicited” not by the “meaning” those words convey (since, as they break down, it is precisely their meaning that is put in question) but by what Felman calls the “experience” of their author, Celan.

The genealogy of this version of the performative is, of course, as much de Manian as Austinian, and is articulated most explicitly in de Man's well-known reading of a passage from Rousseau's Confessions that describes Rousseau's effort to escape being blamed for stealing a ribbon from his employer by blaming the theft instead on a young servant-girl, Marion. Recognizing, with Rousseau, that the crime of having named Marion as the thief is a good deal more serious than the theft itself, de Man works through a series of accounts of what Rousseau might have meant when he said “Marion”—moving from his desire to blame Marion to his desire to possess Marion to his desire for a public scene in which these previous desires are shamefully displayed—but concluding that none of these interpretations is adequate. For ultimately, de Man argues, Rousseau never meant to and did not in fact name Marion. He “was making whatever noise happened to come into his head; he was saying nothing at all, least of all someone's name.” So no attempt to understand what “Marion” means can succeed because, properly understood, “‘Marion’ is meaningless.” And de Man goes on to assert that the “essential non-signification” of this text is exemplary of textuality as such: “It seems to be impossible to isolate the moment in which the fiction stands free of any signification. … Yet without this moment, never allowed to exist as such, no such thing as a text is conceivable.” On the one hand, it is the “arbitrariness” of the sign that makes meaning possible; on the other hand, it is the revelation of that same “arbitrariness”—“the complete disjunction between Rousseau's desires … and the selection of this particular name … any other name, any other word, any other sound or noise could have done just as well”—that “disrupts the meaning.” For de Man, the speech act becomes performative only in the moment that it becomes illegible.²³

In the wake of the discovery of de Man's wartime journalism, some critics have read “The Purloined Ribbon” and the theory it articulates as a kind of alibi for de Man's own disinclination to acknowledge whatever involvement he may have had in the apparatus that produced the Holocaust. According to Felman, however, the refusal to confess is a sign not of indifference to one's own morally scandalous behavior but of a heightened sensitivity to exactly what is scandalous about it. The “trouble” with confessions, she writes, “is that they are all too readable: partaking of the continuity of conscious meaning and of the illusion of the restoration of coherence, what de Man calls ‘the readability of … apologetic discourse’ … pretends to reduce historical scandals to mere sense and to eliminate the unassimilable shock of history, by leaving ‘the [very] assumption of intelligibility … unquestioned’” (151). What is scandalous in “historical scandals” is what, in Felman's view, marks the Holocaust above all: its resistance to intelligibility. Insofar as confession produces a “referential narrative,” it necessarily diminishes the crime it confesses to. Thus, following Lanzmann, who identifies the Holocaust as a “pure event” and who characterizes the effort to make sense of it as a “perverse form of revisionism” (YFS [Yale French Studies], 482), Felman insists on the “refusal of understanding” (YFS, 477) as (quoting Lanzmann) “the only possible ethical … attitude” (YFS, 478). The attempt to explain it can only be an attempt to reduce it.

Felman thus regards what some have thought of as de Man's worst sin—his failure to confess—as his greatest virtue, for confession, diminishing the crime, would excuse the criminal. But whatever we may think of de Man's personal morality, his real contribution here is not his (from Felman's standpoint, admirable) refusal to confess; it is his discovery of the linguistic form that, unlike the “referential narrative,” is adequate to the Holocaust, the “performative.” The essence of the performative is, as we have seen, its irreducibility to “mere sense,” and it is precisely this irreducibility that makes it appropriate as a technology for what Lanzmann calls the “transmission” (YFS, 486), rather than the representation, of the Holocaust. Felman thus focuses intensely on the moment in Shoah when Lanzmann, listening to some Polish peasants describe the efforts of a Ukrainian guard to keep his Jewish prisoners quiet, hears sounds that he recognizes right away are “no longer simply Polish” (230). The Poles are saying, “So the Jews shut up and the guard moved off. Then the Jews started talking again in their language … : ra-ra-ra and so on” (230). The “ra-ra-ra” here is the aural equivalent of “eepinow, / ee-i-o,” and both “ee-i-o” and “ra-ra-ra” are occurrences of the performativity theorized in “The Purloined Ribbon” and embodied in Rousseau's “Marion.” Rousseau “was making whatever noise happened to come into his head,” de Man writes; testimony from “inside” the “horror” can only be heard as “pure noise,” according to Felman (232). If to understand is, inevitably, to misunderstand, to bear false witness, it is only the “mere noise” one “does not understand” (231) that makes it possible to bear true witness.

The point of the performative, then, is that, itself an event, it “transmits” rather than represents the events to which it testifies. This is what Felman means when she says that Shoah “makes the testimony happen” (267), and even that it happens “as a second Holocaust.” So the Holocaust, like slavery, is never over—it is “an event that … does not end” (67). And just as the transformation of history into memory made it possible for people who did not live through slavery to remember it, so the transformation of texts that “make sense” of the Holocaust into events that “enact” it makes it possible for people who did not live through the Holocaust to survive it. “The listener to the narrative of extreme human pain, of massive psychic trauma,” says Felman's collaborator, the analyst Dori Laub, “comes to be a participant and a co-owner of the traumatic event: through his very listening, he comes to partially experience trauma in himself” (57). “Is the act of reading literary texts itself inherently related to the act of facing horror?” asks Felman (2). De Man's account of the performative, of the replacement of “meaning” by “event,” makes the answer yes.

But what de Man characterized as the failure of reference—in order for a text “to come into being as text,” he says, its “referential function” has to be “radically suspended”²⁴—Felman cannily characterizes as the return of reference, “like a ghost” (267). Reference has returned because the text, insofar as it ceases to refer to things, has become a thing that can be referred to; it has returned “like a ghost” because the thing it is a kind of absence, “the very object—and the very content—of historical erasure” (267). When he said “Marion,” Rousseau “was making whatever noise happened to come into his head; he was saying nothing at all, least of all someone's name.” Both erased and embodied by performativity itself, Marion, like Beloved, walks. But this turn to the ghost makes clear not only an important point of resemblance between deconstruction and the New Historicism but also an important point of difference. The ghosts of New Historicism are, as we have seen, essential to its functioning, but as the simile—reference returns “like a ghost”—suggests, in deconstruction they are essentially supererogatory. In deconstruction the texts do what in New Historicism the ghosts must do. Indeed, if we take the ghosts of New Historicism as a figure for its ambition to turn history into memory, we can understand the “mere noise” of Felman's deployment of de Man as an effort to provide the thematics of historicism with its formal ground. Deconstruction requires no ghosts because the emergence of a meaningless and untranslatable signifier in the poem of Celan or in the film of Lanzmann can actually produce what Lanzmann calls the “resurrection” that a text like Beloved only narrates. Understood in these terms, deconstruction is the theory of, rather than the alternative to, the New Historicism; deconstruction explains how texts can not only thematize the transformation of the historical past into the remembered past but, by way of the performative, can actually produce that transformation.

So if the passing of eyewitnesses and the fading of memories do indeed give the question of what sustains Jewish identity a new urgency, this new deployment of deconstruction helps make the Holocaust available as a continuing source of identitarian sustenance. But in seeking to ensure that the Holocaust is not forgotten, deconstruction contributes not only to the maintenance of but also to a change in Jewish identity. For insofar as Jewish identity is understood to depend on what Michael Krausz has recently called “identification” with the “narrative” of the Holocaust as “the most salient episode in contemporary Jewish history,” it is significantly detached from the racial base that was definitive for the perpetrators of the Holocaust.²⁵ The primacy of the Holocaust as a guarantor of Jewish identity marks, in other words, the emergence of an explicitly antiessentialist Jewishness.

This antiessentialist Jewishness is disarticulated from the idea of a Jewish race and also, albeit less sharply, from the idea of a Jewish religion. Many of those who think of themselves as Jews do not think that they are Jews because they have Jewish blood and are in fact skeptical of the very idea of Jewish blood. For them, as for many members of other races (so-called), cultural inheritance takes the place of biological inheritance. And many of those who think of themselves as Jews do not think that they are Jews because they believe in Judaism. But by redescribing certain practices that might be called religious—circumcision, for example—as cultural, Jewishness can sever their connection to Judaism. Thus, Jews can give up the belief in Jewish blood and give up the belief in a Jewish God; what they can't give up is Jewish culture. Hence the significance of the Holocaust and of the widespread insistence that Jews remember it, and hence the importance of the idea that “understanding” the Holocaust is a kind of “obscenity.” The prohibition against understanding the Holocaust is at the same time formulated as the requirement that it be experienced instead of understood, and this requirement—fulfillable through technologies like the deconstructive performative—makes it possible to define the Jew not as someone who has Jewish blood or who believes in Judaism but as someone who, having experienced the Holocaust, can, even if he or she was never there, acknowledge it as part of his or her history.

And just as remembering the Holocaust is now understood as the key to preserving Jewish cultural identity, the Holocaust itself is now retrospectively reconfigured as an assault on Jewish cultural identity. “The commanding voice at Auschwitz,” Lionel Rubinoff writes, “decrees that Jews may not respond to Hitler's attempt to destroy totally Judaism by themselves co-operating in that destruction. In ancient times, the unthinkable Jewish sin was idolatry. Today, it is to respond to Hitler by doing his work.” Jews who might today be understood to be doing Hitler's work are not, of course, murdering other Jews, which is to say that Hitler's work, the destruction of Judaism, is understood here as only incidentally the murder of Jews. Rather, the Jews who today do Hitler's work are Jews who “survive” as people but not “as Jews.” They stop thinking of themselves as Jews; they refuse the “stubborn persistence” in their “Jewishness” that is required by Rubinoff as the mark of resistance to Hitler.²⁶ What this means is that the concept of “cultural genocide,” introduced in analogy to the genocide of the Holocaust, now begins to replace that genocide and to become the Holocaust. “A culture is the most valuable thing we have,” says the philosopher Eddy M. Zemach, and this commitment to the value of culture requires that the Holocaust be rewritten as an attack on culture.²⁷ Thus, the “Judaism” that Hitler wanted to destroy ceases to be a group of people who had what he thought of as “Jewish blood” and becomes instead a set of beliefs and practices; and the Hitler who in fact “opened almost every discussion on Jewish matters with the assertion that the Jews are not primarily a religious community but a race” is now reimagined as a Hitler who wished above all to destroy Jewish religion and culture.²⁸ From this standpoint Hitler becomes an opponent of cultural diversity, and those Jews who have, as Zemach puts it, “lost the will to retain their culture” become not only his victims but his collaborators. They do his work by assimilating, and insofar as, according to Zemach, American Jews in particular are abandoning their culture, what Jews now confront is the threat of a second Holocaust: if American Jews give up their Jewishness, Jews “will have lost the greatest and most advanced part of their people” “for the second time this century.”²⁹

This revaluation of assimilation as Holocaust marks the complete triumph of the notion of culture, which now emerges not merely as the defining characteristic of persons (“the most important thing we have”) but as itself a kind of person, whose death has a pathos entirely independent of the death of those persons whose culture it was. The Jew is here subsumed by his Jewishness.³⁰ The person is transformed into an identity, and the identity is treated as a person.³¹ And if we return to the revised version of the question with which we began—Which myths do Americans believe?—we can see that culture, not visitors, races, or even history, is the correct answer. Americans, especially American academics, believe in the myth of culture. Indeed, with respect to American academics, the point could be put more strongly: we do not simply believe in the myth of culture; many of us have accepted as our primary professional responsibility the elaboration and promulgation of the myth, and our theoretical disputes (between, say, deconstruction and the New Historicism) have become only local disagreements about how best to defend it (which are more real, ghosts or performatives?). In American academic life today, the resolution of these disagreements doesn't matter; it doesn't matter whether we say “Beloved” or “Marion.” What matters is only that we say one or the other, and that in saying whichever we say, we testify to our belief in entities—that is to say, in cultural identities—exactly as real as Ares and Aphrodite.

Notes

1. Paul Veyne, Did the Greeks Believe in Their Myths? An Essay on the Constitutive Imagination (Chicago: University of Chicago Press, 1988), 20. Hereafter cited parenthetically.

2. Whitley Strieber, Communion: A True Story (New York: Beech Tree Books, 1987), 123.

3. Arthur Schlesinger Jr., The Disuniting of America (New York: W. W. Norton, 1991), 20.

4. Greg Bear, Blood Music (New York: Arbor House, 1985), 111-12, 217. I am grateful to Joanne Wood for bringing this book to my attention.

5. Strieber, Communion, 297.

6. Toni Morrison, “The Pain of Being Black,” Time, May 22, 1989, quoted in Mae G. Henderson, “Toni Morrison's Beloved: Re-membering the Body as Historical Text,” in Comparative American Identities: Race, Sex, and Nationality in the Modern Text,” ed. Hortense J. Spillers (New York: Routledge, 1991), 83.

7. Toni Morrison, Beloved (New York: Knopf, 1987), 13. Hereafter cited parenthetically.

8. Valerie Smith, “‘Circling the Subject’: History and Narrative in Beloved,” in Toni Morrison: Critical Perspectives Past and Present, ed. Henry Louis Gates Jr. and K. A. Appiah (New York: Amistad, 1993), 345.

9. Bear, Blood Music, 197.

10. Which is not, of course, to say that enslavement is the only or necessarily the defining racial experience. Indeed, Paul Gilroy follows Morrison in claiming that too often slavery “gets forgotten,” and he explicitly opposes Morrison's memory of slavery to the memories of Kemet, the “black civilization anterior to modernity” that Afrocentrists sometimes invoke “in its place” (Gilroy, The Black Atlantic: Modernity and Double Consciousness [Cambridge: Harvard University Press, 1993], 190). The difference matters to Gilroy because he associates the appeal to Kemet with the attempt to “recover hermetically sealed and culturally absolute racial traditions,” and he thinks of the appeal to slavery “as a means to figure the inescapability and legitimate value of mutation, hybridity and intermixture” (223). Gilroy prefers hybridity to purity, and so for the purposes of “identity construction” he would rather remember slavery than Egypt; but from the standpoint of the argument developed in this essay, the questions of which past you choose to remember and what kind of identity you choose to construct obviously matter less than the commitment to constructing identity by remembering the past in the first place.

11. Smith, “‘Circling the Subject,’” 350.

12. Stephen Greenblatt, Shakespearean Negotiations: The Circulation of Social Energy in Renaissance England (Berkeley: University of California Press, 1988), 1. Candor requires the acknowledgment that my own book, The Gold Standard and the Logic of Naturalism, was published in 1987 as the second volume in the New Historicism series. In a recent essay (“The Ironic Romance of New Historicism,” Arizona Quarterly 51 [1995]: 33-60), Charles Lewis has also pointed to a relation between Beloved and the New Historicism, identifying them both as forms of “historical romance” (54) and arguing that Morrison's “appropriation of the conventions of romance” should be understood to pose a critical challenge to New Historicism's assertion of “a stable connection between those narrative techniques and a particular ideology” (51). Lewis is, in my view, right to see the relation but mistaken in his account of what's being related, missing what is distinctive about both the New Historicism and Beloved. What's new about the New Historicism is not that it seeks to establish connections between literary works and social history but that it seeks to reconfigure the relation between the historical past and the present. And what's important about Beloved is not that it challenges the connection between “narrative techniques” and “ideology” but that it produces its narrative techniques as the technology for a politics committed to replacing ideology with identity.

13. For a critical discussion of these two forms of interest in the past (Greenblatt calls them the interest in “continuity” and in “analogy”) and especially of the effort to imagine the historical past as simultaneously connected to and like the present, see Steven Knapp, Literary Interest (Cambridge: Harvard University Press, 1993), 106-36.

14. Greenblatt, Shakespearean Negotiations, 1.

15. In an appendix to the revised edition of his influential history of the origins of nationalism, Imagined Communities (London: Verso, 1991), Benedict Anderson distinguishes between “real” memory and “mythic” memory and suggests the problematic relation of the “mythic” to the “real” by putting the mythic in scare quotes: “memory.” In these terms, the question addressed here is, What does it mean to believe in this myth, in “memory”?

16. Caroline Rody accurately describes the appeal of Beloved's historicism when she observes that “writing that bears witness to an inherited tragedy approaches the past with an interest much more urgent than historical curiosity or even political revisionism” and goes on to contrast what she calls an “objective ‘prehistory of the present’” to “the subjective, ethnic possession of history understood as the prehistory of the self” (Rody, “Toni Morrison's Beloved: History, ‘Rememory,’ and a ‘Clamor for a Kiss,’” American Literary History 7 [1995]: 97). Insofar as to inherit a tragedy involves something more than living with its consequences—as, of course, it must, since everybody is already living with the consequences of past events—it is only through some mechanism of “possession” that any tragedy can count as an inherited one. The sense of urgency, in other words, is entirely dependent on the claim to possession. So one implication of my suggestion in this essay (and elsewhere) that no history can, in the required sense, be possessed by us is that there can be no real urgency to the study of history and no coherent motive beyond curiosity. And one by-product of the replacement of an indefensibly usable past with a defensibly useless one would presumably be a diminished interest in history.

17. Geoffrey Hartman, “Introduction: Darkness Visible,” in Holocaust Remembrance: The Shapes of Memory, ed. Geoffrey Hartman (Oxford: Blackwell, 1994), 7.

18. Pierre Vidal-Naquet, Assassins of Memory: Essays on the Denial of the Holocaust, trans. Jeffrey Mehlman (New York: Columbia University Press, 1992), 57.

19. Claude Lanzmann, “An Evening with Claude Lanzmann,” May 4, 1986, quoted in Shoshana Felman and Dori Laub, Testimony: Crises of Witnessing in Literature, Psychoanalysis, and History (New York: Routledge, 1992), 213-14. Testimony hereafter cited parenthetically.

20. Claude Lanzmann, “Seminar on Shoah,” Yale French Studies 79 (1991): 85. Hereafter cited parenthetically as YFS.

21. Claude Lanzmann, “The Obscenity of Understanding: An Evening with Claude Lanzmann,” American Imago 48, no. 4 (1991): 481.

22. J. L. Austin, How to Do Things with Words (New York: Oxford University Press, 1965), 6.

23. Paul de Man, Allegories of Reading (New Haven: Yale University Press, 1979), 39, 40, 36-37.

24. Ibid., 44.

25. Michael Krausz, “On Being Jewish,” in Jewish Identity, ed. David Theo Goldberg and Michael Krausz (Philadelphia: Temple University Press, 1993), 272.

26. Lionel Rubinoff, “Jewish Identity and the Challenge of Auschwitz,” in Jewish Identity, ed. Goldberg and Krausz, 150, 136.

27. Eddy M. Zemach, “Custodians,” in Jewish Identity, ed. Goldberg and Krausz, 122.

28. Yisrael Gutman, “On the Character of Nazi Antisemitism,” in Antisemitism through the Ages, ed. Shmuel Almog (Oxford: Published for the Vidal Sassoon International Center for the Study of Antisemitism, the Hebrew University of Jerusalem, by Pergamon, 1988), 359.

29. Zemach, “Custodians,” 129.

30. This is, of course, different from saying that the individual Jew is subsumed by the group of Jews. The issue in cultural (as in racial) identity, despite the assertions of polemicists on both sides, has nothing to do with the relative priority of the group over the individual; it has to do instead with the identification of a certain set of beliefs and practices as appropriate for a person or persons in virtue of the fact that those beliefs and practices are his, hers, or theirs. What's wrong with cultural identity, in other words, is not that it privileges the group over the individual but that it (incoherently) derives what you do from what you are.

31. This is, to some extent, implicit in the very idea of genocide, inasmuch as genocide is understood as the extermination of a people rather than as mass murder. In genocide, it is what makes the people a people that is the ultimate object of destruction, so the murder of persons is in a strict sense only incidental to the elimination of the people. If, of course, the people are understood as a race, then genocide will require that they be killed or sterilized; if the people are understood as a culture, then genocide will require only that they be forced to assimilate. From this standpoint, even writers who have not lost sight of the fact that Hitler's goal was physical extermination rather than cultural assimilation may find themselves subordinating the death of persons to the destruction of a people. Thus, Berel Lang describes Nazi genocide as worse than cultural genocide because “where life remains, as in cultural genocide or ethnocide, the possibility also remains of group revival; but this is not the case where genocide involves physical annihilation” (Lang, Act and Idea in the Nazi Genocide [Chicago: University of Chicago Press, 1990], 13). The point here is that physical destruction is the worst kind of genocide because, unlike cultural genocide, it is in principle irreversible. The relevant difference between physical and cultural genocide is not, in other words, the fact that in cultural genocide no persons may be killed, which is to say that what's worse about physical genocide is not, on this account, the fact that so many persons must die. For genocide involves the extermination not of persons but of a people. So cultural genocide is less bad than physical genocide not because no persons have been killed but because the people (“the genos”) may still be revived. It is not less murderous (in both cases the group dies); it is less irreversibly murderous (in the second case, the group may live again).

This essay is a revised version of my essay in Narrative, vol. 4, no. 1 (January 1996). Used by permission of Ohio State University Press.