Autobiography reveals gaps, and not only gaps in time and space or between the individual or the social, but also a widening divergence between the manner and matter of its discourse. That is, autobiography reveals the impossibility of its own dream: what begins on the presumption of self-knowledge ends in the creation of a fiction that covers over the premises of its construction.1
One of the most important questions haunting the writing of history in the wake of poststructuralism is that of identity and the definition of subjectivity.2 Poststructuralist authors as various as Roland Barthes, Jacques Lacan, Michel Foucault, and Jacques Derrida argued, not so long ago, that the autonomous subject of the humanist tradition, a subject capable of knowing both the world and itself, was a utopian dream of the European Enlightenment. This view of human subjectivity had to be abandoned in a period that recognized the existence of an unconscious mind, the opacity of language, and the role of discursive practices in the dissemination of social power.
This revision of the idea of subjectivity has had important reverberations for our conception of knowledge generally and our notion of history in particular. If subjectivity is conceived of as something unstable and changing rather than transcendental and constant, then human knowledge can no longer be viewed as something fixed and permanent. Instead of regarding knowledge as an edifice to which positivistic scholarship can continue to contribute so that the scope of its insights might continue to expand and evolve according to generally accepted universal principles, we live in an age that questions the very basis on which that structure was erected. It is these doubts about the traditional premises on which the knowledge-producing activities of the humanist disciplines were once based that has provided the justification for the introduction of a variety of politically-inspired forms of interpretation, such as gender studies, ethnic studies, cultural studies and postcolonialism. The new approaches to historical interpretation no longer claim the epistemological status traditionally associated with positivistic scholarship. Their findings and conclusions are specifically defined as forms of local knowledge rather than as pretensions to universality. These new perspectives subvert previously established knowledge claims by characterizing them as unavoidably tainted or colored by the values that inform the circumstances of their production. The "view from nowhere," the objectivity claimed by foundational epistemology, has come to be seen as suspect because of its identification with Western culture, with the dominance of white races, with masculinist bias, and with middle-class prejudice. The knowledge produced during the nineteenth and twentieth centuries, on which the disciplines of the humanities were founded, is now conceived as one way of understanding the world rather than the way in which the world can be understood.
This new conception of the status and function of knowledge has had a dramatic impact on the history of art. If history is not regarded as the interpretation of the past produced from a universal perspective, but rather as an interpretation of the past produced from a particular perspective, then it cannot be pursued for its own sake. The cultural function of historical interpretation can be openly acknowledged rather than masked behind an ideal of objectivity. As a consequence, the shape of the discipline has been decisively altered. Rather than operate according to an ideology of neutrality and disinterest that insists that the author repress his or her subjectivity in the pursuit of the "facts," rather than fetishize empirical data by suggesting that they might be relied upon to provide the interpretations that are actually forced on them by particular historians, scholars have begun to foreground their commitment to a specific form of understanding. In substituting an interpretive agenda for the allegedly neutral dedication to description, many art historians now offer us access to the methodological procedures and political goals that inform their particular views. What was once hidden in the interest of providing a common front, one which suggested that human subjectivity was universal in nature, is now placed in the open so as to assert the differing interests of divers interpretive communities.
The consequences of these changes have been profound, if not always beneficial. Art history is now characterized by a variety of voices, each seeking to represent the interests of different sectors of the discipline's population. Disciplinary conferences offer a variety of alternative points of view, all of which are engaged in a struggle to obtain the attention of the professionals in the field. In this new situation, issues of identity and subjectivity take on new meaning. It is not sufficient to destabilize humanist notions of subjectivity as something essential and autonomous, without reflecting upon the concept of subjectivity that replaces them. The problem is effectively stated by Ernesto Laclau.
Thus once objectivism disappeared as an "epistemological obstacle," it became possible to develop the full implications of the "death of the subject." At this point, the latter showed the secret poison that inhabited it, the possibility of a second death, "the death of the death of the subject," the reemergence of the subject as a result of its own death; the proliferation of concrete finitudes whose limitations are the source of their strength; the realization that there can be "subjects" because the gap that "the subject" was supposed to bridge is actually unbridgeable."3
How is this subject with a small "s" to be theorized if it is not simply to be an epigone of its ancestor? How is one, for example, to theorize the historian's relation to his or her text? Is there a correspondence between the historian's subjectivity on the one hand and the text on the other? Do the class, gender, or ethnic identities of the historian determine the nature of his or her intervention in the writing of history? How are the politics of identity inscribed in history writing?
In the course of writing the essays that constitute my next book, tentatively titled Motivating History, I have had to reflect upon my own relation to its argument. The circumstances in which this enterprise was undertaken are very different from those that reigned just a few years ago. At the time I wrote, The Practice of Theory (published in 1994), I wrote as if the history of art still had a disciplinary center, and the voice I articulated was deliberately located in the margins.4 I characterized myself as a historian interested in theoretical initiatives that had affected the structure of neighboring disciplines such as literary studies, anthropology, and history, initiatives that seemed to have had little impact on art history. The point of steeping myself in these theories was to try to make them register in art historical interpretations, to use theory to destabilize the master narratives that constituted the discipline. I was openly engaged in a polemic, one that championed change and transformation, the articulation of multiple discourses and perspectives.
In the years that have elapsed since then, art history has changed substantially. It is not as though the disciplinary establishment suddenly saw the light, abandoning a positivistic scholarship informed by a notion of objectivity for one that recognized the impossibility of keeping subjectivity and objectivity apart. Rather, the establishment accommodated itself to the theoretical and political interests of many of its members. Indeed, there are relatively few institutions, with the exception of some departments and an occasional fellowship-granting foundation, that have not responded in some way to the changes wrought in the discipline's way of doing business. Using pluralism as an ideology of the status quo, representatives of established methodologies now characterize themselves as part of what they regard as a range of incompatible and incommensurable forms of interpretation that clamor for the discipline's attention.5 Under the aegis of "let many flowers bloom," it is argued that since there is no objective way of determining the value of one form of interpretation over another, then they must all be equally viable. Pluralism masks the fact that different forms of interpretation exist in necessary tension to one another. In the absence of a value neutral position from which to argue the cultural and historical relevance of one perspective over another, in a situation that acknowledges that discourse cannot be disassociated from power, discussions between differing approaches become pointed and sometimes acrimonious as they seek to register their significance within art history's disciplinary fabric. The rhetorical warfare that marks, say, the making of academic appointments or the distribution of fellowships exposes pluralism as a myth imposed on fragmented and divided circumstances by those interested in maintaining the status quo.
Before attempting to characterize my own position and my own voice within the new context of art history's multivocal discourse, I want to return to a consideration of theories of identity and subjectivity. How does a writer's personal identity register in a historical text? What is the relation between authorial subjectivity and textual product? Theories of subjectivity and the nature of human agency have been much discussed in the context of feminist theory.
Following Foucault's suggestion that subjectivity is defined by the conventional systems responsible for making cultural meaning, systems he terms discursive practices, Judith Butler has argued that subjectivity is both constituted by those practices and empowered by them to act upon the processes that gave them shape. Butler theorizes the instantiation of subjectivity by means of the concept of performance. Exploiting the ambivalence inherent in this notion, she invokes its significance both as an art of repetition and as an act of personal agency. Subjectivity (in Butler's case, gendered subjectivity) is thus a process whose script has been prescribed but whose enactment is necessarily varied. The performance of subjectivity is thus a form of repetition without duplication, and it is this simultaneous production of sameness and difference, or difference within sameness, that allows for the possibility of a concept of agency:
Paradoxically, the reconceptualization of identity as an effect, that is as produced or generated, opens up possibilities of "agency" that are insidiously foreclosed by positions that take identity categories as foundational and fixed. For an identity to be an effect means that it is neither totally determined nor fully artificial and arbitrary.... Construction is not opposed to agency; it is the necessary scene of agency, the very terms in which agency is articulated and becomes culturally intelligible.6
If, as Butler suggests, the subjectivity of the historian is conceived of as something both constructed and constructing, as an effect of discursive processes as much as their author, the link between an author and his or her text is relational rather than determined. This point becomes important in attempting to understand the way in which identity might be inscribed in a historical text. The plethora of voices that currently characterizes the history of art cannot be viewed as incommensurable with one another. Rather than fixed and permanent, the identities that manifest themselves in politically-inspired forms of interpretation are part of a process of change and transformation.
Joan Scott, however, has indicated the persistence of the rhetoric of the humanist subject, a rhetoric of omniscience and finality, in forms of interpretation that respond to notions of identity in the production of situated knowledge. Such a rhetoric could not be more opposed to the idea of subjectivity as process. Arguing that the attempt to assert local interests by positing minority identities has often been subverted by a tendency to conceive of them in terms once used to ensure the dominance of the transcendental subject, Scott claims:
The logic of individualism has structured the approach to multiculturalism in many ways. The call for tolerance is framed in terms of respect for individual characteristics and attitudes; group differences are conceived categorically and not relationally, as distinct entities rather than interconnected structures or systems created through repeated processes of the enunciation of difference.7
The temptation to view the contestatory subjectivities that have arisen in the wake of the demise of the humanist subject as radically incommensurable depends upon a survival of the notion of individualism associated with the ancien regime. This tendency, subscribed to both by those whose political agendas have depended upon the assertion of differences and by those who have sought to discredit the politics of difference, is a travesty of the conception of subjectivity proposed by Butler. Scott writes:
[I]t makes more sense to teach our students and tell ourselves that identities are historically conferred, that this conferral is ambiguous (though it works precisely and necessarily by imposing a false clarity), that subjects are produced through multiple identifications, some of which become politically salient for a time in certain contexts, and that the project of history is not to reify identity but to understand its production as an ongoing process of differentiation, relentless in its repetition but also and this seems to me the important political point subject to redefinition, resistance, and change.8
Assuming, then, that identity is an "ongoing process of differentiation," what conclusions can we draw from the theoretical work of Butler and Scott for the work of the historian? How do we conceive of the discursive practices of which the historian is an "effect," and what is the nature of the "agency" he or she possesses in the production of an historical interpretation? First, in order to be able to intervene in the literary genre known as art history, the historian must have acquired a high degree of general and professional education. The historian is not only constituted by the discursive practices associated with educational institutions, but he or she must also absorb the reigning paradigms of knowledge production that characterize the historiographic moment.9 The discursive practices of educational and professional formation are inevitably class-inflected. The art historian for example, is necessarily implicated in the transmission of "cultural capital" from one generation to another.10 A knowledge of the visual arts has traditionally been associated with the social elite, and, since the late eighteenth century, works of visual art have been identified with a form of spiritual value, known as aesthetic value, which has been an integral part of the cultural life of the bourgeoisie. The art historian is thus inextricably involved in both the creation and support of class distinctions. The art historical canon, that collection of works of art to which the history of art has traditionally dedicated its attentionÛa canon established on princely and aristocratic tastes that was nationalized to become state property during the course of the eighteenth and nineteenth centuries was both ennobled and democratized by means of the notion of aesthetic value so as to become an essential aspect of bourgeois education.
Does the art historian's formation by and participation in the processes by which class distinctions are perpetuated necessarily mean that elitist values are embedded in the histories they produce? Only the most reductive account of identity politics would argue that this is necessarily the case. If the scholar is both an effect of cultural formations and an agent of their construction, then his or her text may either transmit the class ideology of art history's academic discourse relatively unchanged, or it may bear only the most tangential relation to it. Given the importance of class identification for art historical discourse, however, I would argue that the historian's relation to the discipline's social function must always be significant. In evaluating the nature of a scholar's intervention in the historiography of the discipline, his or her understanding of art history's role in maintaining class distinctions will be of interest for an interpretation of an author's work. If the connection between the historian and his or her text is relational rather than determined, as I have been arguing, then the inscription of class attitudes can take a multitude of different forms. A fairly common one, for example, is the scholar who assumes the notion of aesthetic value, the idea that there is some spiritual sustenance to be found in works of art that sets them apart from the rest of the paraphernalia of everyday life, without recognizing that such an understanding of aesthetic value is a characteristic of a social elite with the cultural capital to appreciate it.
Secondly, quite apart from the ideological processes in which the historian is either wittingly or unwittingly involved, we must consider the unconscious or psychoanalytic mechanisms that characterize the historian's work both as a scholar of the past and as a pedagogue of future generations. In other words, how rationally are disciplinary paradigms of knowledge production transmitted and received? To what extent is the absorption of the discursive practices to which art history's methodological alternatives belong unconsciously determined rather than consciously chosen? Dominick LaCapra has pointed out that the situation in which a graduate student acquires the interpretive models of a discipline, that is, his or her relation to a professor, is analogous to the relation that exists between an analyst and an analysand in psychoanalysis.11 The student is thus bound to the historiography of the discipline in a highly personal manner, one in which an unconscious bond may well be as important as a conscious one. Just as the analysand adopts certain attitudes of the analyst in the attempt to restructure his or her past experience in relation to the present, a process known as transference, so the student will adopt some of the characteristics of the professor in order to transform him or herself into a figure of equivalent cultural authority. This kind of identification often results in the perpetuation of accepted forms of meaning production at the expense of more innovative alternatives. Even if the student consciously repudiates the models absorbed during the training period, that rejection will itself be relevant to an evaluation of that professional's eventual historiographic contribution. The rejection might, for example, represent an "anxiety of influence," a fear of imitating a respected authority and a desire to break out of a professional mold in order to claim an authority the equivalent of or superior to that of the original mentor.12
And third, the discursive practices of professional formation demand that the art historian put his or her personal stamp on literary production. This requirement is made regardless of whether the scholar is simply duplicating or creatively extending and manipulating an established disciplinary paradigm. Paradoxical as it appears in a positivistic tradition that insists that the historian's task is to afford the public access to the truthÛa process that might be undertaken, presumably, by anyone with the time, training, and inclination to do soÛthe conventions of professional life insist that each scholar distinguish his or her contribution from those of peers. The call for the construction of a unique subjectivity, the cult of the exceptional individual, is, of course, also the heritage of a culture deeply invested in the ideology of the humanist subject. The degree to which a historian is susceptible to these ideological demands will also register significantly in any account of the discipline's history.13
I hope that this sketch of some of the discursive practices that constitute, enable and empower historical writing helps us to think about the relation of the historian to his or her interpretive text. Because the drives and neuroses that determine the historian's psychological formation and the nature of the discursive practices that have shaped his or her professional subjectivity are not necessarily accessible to that subjectivity, it is inevitable that the character of the individual historical narrative, its full implications for the historical moment in which it is composed, cannot be fully recognized. It is often, for example, not until the processes of research and writing have ended that the scholar can perceive the pattern that informs the work. As LaCapra points out, in the case of the historian there may well be a psychological mechanism that serves to make the work opaque to its creator. Just as the graduate student may have a transferential relation to certain professors, as well as to the historiographic traditions of the discipline, so he or she may have an analogous relation to the historical horizon under investigation.14 The encounter with the past may constitute a retreat from the cultural values of the present, one that valorizes the qualities that distinguish the culture of another moment in time. Just as the student identified with a mentor, so he or she may identify with a historical period, in which case the interpretive construct that arises from this encounter of the present and the past may well be marked by transference. The values of another horizon may be adopted, and used in the construction of a historical narrative. While the historian will inevitably project into the past the values of the contemporary world, as well as those of the selected historiographic paradigms, a historical interpretation may in turn be unconsciously shaped by the values of another horizon that have been incorporated into the construction of the past in the present.
A historical interpretation, therefore, bears only an oblique relation to the subjectivity of its author. The text is a kind of extended metaphor of the historian's psychological, educational, and professional formation, as well as of his or her intervention in the discursive practices that result in the creation of historical meaning. If the significance of a historical text is not fully accessible to the subjectivity responsible for its composition, and if it is impossible for the historian to discern the past because access is mediated by historically determined and psychologically-inflected paradigms that frame that encounter, then it seems possible to conclude that there is no obvious relation between a historical text and the subjectivity that created it. Nevertheless, the lack of a one-to-one correspondence between the discursive practices that shaped the historian and the texts he or she produces does not mean that an appreciation of the processes that constitute subjectivity as well as the production of texts is irrelevant. It is vital that the historian maintain a high degree of self-reflexivity about his or her intervention in the genre of historical interpretation so that some awareness of the social and cultural implications of this activity inform the pursuit of disciplinary and paradigmatic "objectivity."
Given the complexity of the operation outlined above, it may seem contradictory to reintroduce a reference to my own role in inserting the essays that constitute my book into contemporary debates about the nature of art history as a discipline. Some of the essays constitute a plea for greater theoretical and methodological diversity in the discipline's interpretive procedures and thus plainly belong to an earlier moment in its history. Others seek to articulate the ways in which the discursive practices that constitute the discipline's historiography manifest the attitudes that characterize their cultural location, as well as the way in which particular historians have either reiterated those ideologies or called them into question in the process of constructing their own interventions. My purpose in theorizing the role of identity and subjectivity in the production of historical narratives is to denaturalize disciplinary traditions that seek to maintain the idea that the historical voice is at best always a disembodied voice.
In calling attention to my own investment in these meta-historical narratives, I am fully aware of the debate that swirls around the role of autobiography in scholarly writing.15 The introduction of the personal into a discursive practice such as historical writing can often constitute a form of essentialism, a way to posit a direct connection between an author and his or her text. In this scenario, the introduction of the personal serves to ground the narrative in the author's experience, in such a way as to make the intimate bond between subjectivity and memory serve as an unassailable foundation for the views being presented. On this view, for example, only African-Americans can represent the views of African-Americans, and only women can articulate feminist agendas.
The concept of experience, that allegedly unmediated foundation on which claims to situated knowledge are sometimes based, has been usefully theorized by Joan Scott. Arguing that there is nothing transparent or immediate about appeals to biography, she suggests that those events we consider crucial to our definition of subjectivity are always decided in retrospect. Indeed, the Freudian concept of Nachträglichkeit suggests that what our memories call experience is subject to a continual process of change, as those memories are recalled in the ever-changing circumstances of the present.16 The quotidian flow of events makes no distinction between those experiences we deem formative and those we do not; indeed the process of transforming an event into an experience suggests the thought process involved in that metamorphosis. In addition, some of the most important events that have affected us, those of a traumatic nature, cannot be recalled in their original form, but only through the filter of constructed memories:
It is not individuals who have experience, but subjects who are constituted through experience. Experience in this definition then becomes not the origin of our explanation, not the authoritative (because seen or felt) evidence that grounds what is known, but rather that which we seek to explain, that about which knowledge is produced. To think about experience in this way is to historicize it as well as to historicize the identities it produces.17
If, as I have been arguing, there is no direct correspondence between an author and his or her text, then what is the point of introducing the so-called "personal" at all? What purpose can reference to biography serve in understanding a text, if it is impossible to demonstrate a connection between them? Returning to Butler's notion of performance allows us to conceive of a particular subject's acts of agency as both 1) prescribed, in the sense of having been installed in that subjectivity by means of the discursive practices that brought it into being, and 2) instantiated, in the sense in which those discursive practices must be enacted by the subjectivity in question in ever-changing circumstances that necessarily endow them with new meaning. Just as it is necessary for an appreciation of a scholar's historiographic location to acknowledge the discursive practices in which he or she was formed, so it is appropriate to understand the autobiographical account of the author responsible for the production of a specific text. The function of autobiography has been extensively theorized in the wake of Roland Barthes's remarkable autobiographical sketch, Roland Barthes par Roland Barthes. Barthes insists on textualizing his life, rigorously refusing to see through the web of language to some underlying reality, arguing that our notions of subjectivity are the product of language itself:
This book consists of what I do not know: the unconscious and ideology, things which utter themselves only by the voices of others. I cannot put on stage (in the text), as such, the symbolic and the ideological which pass through me, since I am their blind spot.18
Barthes's insistence on the prescriptive power of language was reiterated by Paul de Man for whom language served to alienate subjectivity from experience in such a way as to cancel or abolish the possibility of meaning. By denying subjectivity its capacity to inflect and manipulate the process by which meaning is made, de Man suggests language inflicts a kind of metaphorical death on the notion of the subject as agent.19
The idea that the subject is the product of language rather than its creator, however, has been interpreted very differently by other theorists of autobiography. James Olney, Paul John Eakin, as well as Liz Stanley, Shari Benstock and other feminist authors, view the prescriptive power of language positively, regarding it as an empowering process that, in Butler's terms, enables particular subjectivities to play a performative and therefore an active role within the culture that shaped them.20 For Olney, autobiography is not involved in reference to some pre-established reality but a metaphor for the subject's attempt to make order of the universe and thus an attempt to construct reality:
A metaphor, then, through which we stamp our own image on the face of nature, allows us to connect the known of ourselves to the unknown of the world, and, making available new relational patterns it simultaneously organizes the self into a new and richer entity; so that the old known self is joined to and transformed into the new and heretofore unknown self.21
Pursuing this line of thought, Nancy Miller argues that autobiography is a "self-fiction," yet one that enables the historiographer to comprehend the purpose behind the author's writing. She maintains that the introduction of the personal into the discursive practice of writing is not necessarily a form of essentialism, not a way of suggesting that there is a correspondence between author and text, for autobiography must necessarily be a carefully edited version of personal experience that depends for its shape on the deferred action of memory. Autobiography informs us which of the events in the author's life have been dignified with the status of experiences, which of those experiences the author identifies with, and which he or she does not. The insertion of autobiographical myth is thought by Miller to be a form of "personal materialism," one that calls attention to who is speaking:
By the risks of its writing, personal criticism embodies a pact, like the "autobiographical pact" binding writer to reader in the fabrication of self-truth, that what is at stake matters also to others: somewhere in the self-fiction of the personal voice is a belief that the writing is worth the risk. In this sense, by turning its authorial voice into spectacle, personal writing theorizes the stakes of its own performance: a personal materialism.22
The value of Miller's conception of the personal as autobiographical myth rather than autobiographical fact allows us to consider the function of anecdote in a new light. Joel Fineman, for example, has theorized the anecdote as the creation of a "reality effect," a way in which say, a historian, can nest one narrative within another so that they mutually reinforce each other's claims to the "real."23 Within the text, an anecdote opens a window onto context, so that the latter can substantiate and support the former. Anecdote steps outside the primary narrative so as to gesture more persuasively towards the "real." By contrast, Miller's view of anecdote as fabrication allows us to appreciate the role of autobiography not as an attempt to create a "reality effect," but as an effort to draw attention to the author's self-fiction. The point of making reference to myself and my own intentions is not to persuade you, my listener, of the "reality" of my argument, but rather to indicate the perspective from which my narrative is being written. Needless to say, neither the discursive practices that have formed me nor the nature of my own intervention in those practices of history writing is transparently available to me. Nevertheless, I am assuming that my own interpretation of these cultural processes is relevant to an appreciation of the argument I have placed before you.
Returning, one last time then, to my own investment in writing this book, it is difficult for me to know what aspect of an autobiographical myth may be most useful to understanding the perspective that informs my writing. Born to English parents in Buenos Aires, I spent my school years following both Argentine and British primary school curricula. My "experience" in school and elsewhere was complicated by the knowledge that I operated in two different cultural systems, systems that had different attitudes to just about every aspect of everyday life. Part way through my secondary education, I had to decide whether to study for entry into a British or an Argentine University. My choice of a British curriculum enabled me to appreciate the extent to which national identities are fabricated constructs depending on processes of acculturation and education. Upon completing secondary education in Argentina, I went to Britain to study at the University of Edinburgh. Having thought that part of me was "British," it was a nasty shock to discover that the Britain I had absorbed from my parents and their friends was the Britain of the 1940'sÛa very different place from that which I encountered in the 1960Ìs. Instead of cricket and crumpets, I discovered sex, drugs, and rock and roll! Graduate school and professional life in the United States confirmed what I had already suspected, that each national culture fabricates its own version of reality and that foreigners must negotiate an accommodation with the identifications required by each national myth.
I would argue that this self-fiction, one which I have characterized as both determined and empowered by conflicting national identities, can be used as a metaphor of my claim that all identities are constructed and that the scope of knowledge production must necessarily be limited and local. This self-fiction is clearly a heuristic device, a means of extracting from the complexity of my experience, some of the factors I believe have relevance for my thesis. The fact that in Argentina I was a participant in two cultures at the same time is a way of suggesting that subjectivities are both called into being by pre-established discursive practices and empowered by them to challenge as critical identities because their validity cannot be substantiated in the "real." The few facts I have retrospectively culled from my experience are clearly chosen for their application to the purpose of this essay, which is to call into question history's "voice from nowhere" and to insist that the subjectivity of the historian matters. If history writing is to be genuinely historical then it must be capable of acknowledging the cultural agenda that informs its approach to the past.
Contemporary theories of subjectivity offer the historian a paradox. They suggest that personal experience, in the form of autobiography, both matters and does not matter to an understanding of a historical text. On the one hand, autobiography cannot ever give us access to the relation between the historian and the text because it depends upon the fabrication of a self-fiction based on the deferred action of memory. Psychological and ideological forces also intervene to ensure that the text is forever opaque to its author. On the other hand, autobiography or self-fiction offers us insight into the type of self-awareness that informs the agency of a particular subjectivity. It affords the historiographer and the philosopher of history a means of comprehending some of the multitude of cultural practices that inform the writing. If it does not make them available, then at least it suggests the complexity of the processes involved in the writing of history.
In conclusion: The demise of a notion of a wholly rational, autonomous subject led to the proliferation of new voices based on assertions of specific identities that had previously been repressed or occluded by the dominant paradigm. In these circumstances, it has become necessary to theorize a new concept of subjectivity, one whose status as process ensures that it cannot be given stable definition. The idea of subjectivity has been rethought in terms that would have been unrecognizable to its late lamented ancestor. In this new guise, reference to subjectivity as agency serves also, and at the same time, as a reference to its status as the product of those unconscious and ideological forces that haunt the production of meaning. It serves as a means of suggesting that the writing of history is never an entirely rational process, and that, insofar as the cultural stakes that inform its production are capable of being metaphorized or invoked by means of self-fictions, they constitute as important a dimension of its interpretation as any other.
I thank Michael Holly for reading this text and for her help in clarifying some of the points it seeks to make. Janet Wolff read a draft and suggested several crucial readings. Cathy Sousloff also read the text and made a number of useful suggestions. The essay was first presented at the "Recycling Culture" conference organized by Mieke Bal at the Society for the Humanities, Cornell University in 1997. I benefited from her response as well as from the comments of Tim Murray. Versions were also read at the Power Institute in Sydney (courtesy of Terry Smith), the University of Otago (thanks to Peter Stupples), the State University of New York at Stonybrook, and finally at the Getty Institute for Visual and Cultural Studies held at the University of Rochester in the summer 1998.