Philosophy and racial identity Linda Martin Alcoll In the 1993 film Map of the Human Heart an Inuit man asks a white engineer who has come to northern Canada to map the region, 'Why are you making maps?' Without hesitating, the white man responds 'They will be very accurate.' Map-making and race-making have a strong historical as well as conceptual relationship. The ordering and labelling of natural terrain, the classifying of natural types, and the typologies of 'natural races' emerged simultaneously in what Foucault called the Classical episteme. Arguing via Foucault, both Cornel West and David Theo Goldberg have attempted genealogies of modern racism - meaning here not contemporary racism so much as the racism of modernism - that link the Western fetishistic practices of classification, the forming of tables, and the consequent primacy of the visible with the creation of metaphysical and moral hierarchies between racialized categories of human beings.! Given this genesis, the concepts of race and racial difference emerge as that which is visible, classifiable and morally salient. West argues that the application of natural history techniques to the study of the human species yields a comparative analysis 'based on visible, especially physical, characteristics ... [which] permit one to discern identity and difference, equality and inequality, beauty and ugliness among animals and human bodies. '2 Goldberg argues that the universal sameness that was so important for the liberal self required a careful containment and taxonomy of difference. Where rights require sameness, difference must be either trivialized or contained in the Other across a firm and visible border. The result of these classification practices juxtaposed with liberal ideology is a paradox wherein 'Race is irrelevant, but all is race. '3 Visible difference is the route to classification and therefore know ledge, and yet visible difference threatens the security of claims to know by challenging universal applicability and invoking the spectre of relativism. Classification systems can contain this threat and impede relativism by enclosing the entirety of difference within a taxonomy organized by a single logic. In this way the continuing hegemony of liberal discourse is ensured. But the resultant juxtaposItIOn between universalist legitimation narratives that deny or trivialize difference and careful delineations of supposedly morally relevant phenotypic human difference is one of the greatest antinomies of modern discourse. We have finally come to recognize and acknowledge this paradox, but we have not yet solved or moved beyond it. Today the naturalistic classification systems which would reify human variability into moral categories, the Eurocentric teleologies which would excuse if not justify colonialism, and the phallogocentric binaries which would obscure relations of domination by presenting them as 'separate spheres', have been largely exposed as specious. And the realm of the visible, or what is taken as self-evidently visible (which is how the ideology of racism naturalizes racial designation), is recognized as the product of a specific form of perceptual practice, rather than the natural result of human sight. Thus Foucault claims that the object [of discourse] does not await in limbo the order that will free it and enable it to become embodied in a visible and prolix objectivity; it does not preexist itself, held back by some obstacle at the first edges of light. It exists under the positive conditions of a complex group of relations. 4 His central thesis in The Birth of the Clinic is that the gaze, though hailed as pure and preconceptual, can only function successfully when connected to a system of understanding which dictates its use and interprets its results. What defines the act of medical knowledge in its concrete form is not ... the encounter- between doctor and patient, nor is it the confrontation between a body of knowledge and a perception; it Radical Philosophy 75 (JanlFeb 1996) 5 is the systematic intersection of two series of information ... whose intersection reveals, in its isolable dependence, the individual fact. 5 On this account, which is largely unique to Foucault, visibility itself cannot serve as the explanatory cause of the development of racial taxonomies. The apparent obviousness of racial difference - the emphasis on hair type, nose shape and skin colour - is a produced obviousness. The visibility of racial identity is a peculiarly variegated phenomenon, with little acknowledgement of this by dominant discourses. Those of us with hybrid identities surely have a better sense of this, as our public identity is variously interpellated across geographical borders or even just neighbourhoods. When the mythic bloodlines which are thought to determine identity fail to match the visible markers used to identity discourse to signify race, one often encounters these odd responses by acquaintances announcing with arrogant certainty 'But you don't look like ... ' or then retreating to a measured acknowledgement 'Now that you mention it, I can sort of see ... ' To feel one's face studied with great seriousness, not for its (hoped-for) character lines, or its distinctiveness, but for its telltale racial trace, can be a peculiarly unsettling experience, fully justifying of all Sartre's horror of the Look. 6 Anti-essentialisms have corroded the sense of visible difference as the 'sign' of a deeper, more fundamental difference, a difference in behavioural disposition, in moral and rational capacity, or in cultural achievement. Moreover, there is a newly emerging scientific consensus that race is a myth, that the term corresponds to no significant biological category, and that no existing racial classifications correlate in useful ways to gene frequencies, clinal variations or any significant human biological difference. For semantic realists such as Anthony Appiah, the only philosophically respectable position one can take in the face of this evidence is that the concept of race cannot be used correctly, that there is no philosophically defensible way to realign the term 'race' with a referent, even one which would invoke historical experience or culture rather than biology.7 So today race has no semantic respectability, biological basis or philosophical legitimacy. However, at the same time, and in a striking parallel to the earlier liberal attitude toward the relevance and irrelevance of race, in the very midst of our contemporary scepticism toward race stands the compelling social reality that race, or racialized identities, have as much political, sociological and economic salience as they ever had. Race tends toward opening up or shutting down job prospects, career possibilities, available places to live, 6 potential friends and lovers, reactions from police, credence from jurors, and presumptions by one's students. Race may not correlate with clinal variations, but it persistently correlates with statistically overwhelming significance in wage levels, unemployment levels, poverty levels, and the likelihood of incarceration. As of 1992, black and Latino men working full time in the USA earned an average of 68 per cent of what white men earned, while black and Latina women earned 59 per cent. As of 1995, Latino and blac·k unemployment rates were more than double those of whites. But these sociological facts are not thought to be of philosophical significance. For those still working within a liberal framework, the devastating sociological reality of race is but an artificial overlay on more basic elements whose specificity can be legitimately set aside toward the aim of a general analysis. For postmodernists, race is a contingent construction, the epiphenomenon of essentialist discourses, and thus ultimately without any more explanatory power or epistemological relevance than the liberal view. Thus, for all our critical innovations in understanding the vagaries of racist domination and the conceptual apparatus that yields racism, we remain stuck in the modernist antinomy that race is (fundamentally) irrelevant, even though all is race. It will be my contention that we will not be able to progress beyond this unworkable dilemma until we acknowledge the philosophical salience of racial identity, a project that must begin with understanding what racial identity is~ Race as ontology Refusing the reality of racial categories as elements within our current social ontology only exacerbates racism, because it helps to conceal the myriad effects that racializing practices have had and continue to have on social life, including philosophy. In claiming that race is an ontological category, I don't mean to say that we should begin by treating it as such, but that we must begin acknowledging the fact that race has been 'real' for a long time. And I am not putting this forward as a strategic essentialism: the claim that race is philosophically salient is not merely a strategic claim, but a truth claim. There is a visual registry operating in social relations which is socially constructed, historically evolving and culturally variegated, but nonetheless powerfully determinant over individual experiences and choices. And, for that reason, it also powerfully mediates SUbjectivity. Consider the following passage from Richard Rodriguez: I used to stare at the Indian in the mirror. The wide nostrils, the thick lips. Starring Paul Muni as Benito Juarez. Such a long face - such a long nose - sculpted by indifferent, blunt thumbs, and of such common clay. No one in my family had a face as dark or as Indian as mine. My face could not portray the ambition I brought to it. s This mediation through the visible, working on both the inside and the outside, both on the way we read ourselves and the way others read us, is what is unique to racialized identities as against ethnic and cultural identities. The processes by which racial identities are produced work through the shapes and shades of human morphology, and subordinate other markers such as dress, customs and practices. And the visual registry thus produced has been correlated with rational capacity, epistemic reliability, moral condition and, of course, aesthetic status. Yet, as a result of the theoretical critique of race, this visual registry has largely not been brought into theoretical play, in either cultural studies or in philosophy.9 This visual registry cannot be fully or adequately described except in ontological terms, because the difference that racializing identities has made is an ontologizing difference, that is, a difference at the most basic level concerning knowledge and subjectivity, being and thinking. If we say that race is not an ontological category, and that it is a mere artificial overlay on top of more basic and more real categories, we risk losing sight of how significant the effects of racial identities have been, and how those effects have permeated every philosophical idea. Ontology itself might then be able to avoid a needed self-critique. Metaphysics and epistemology could proceed with their habitual disregard for issues of race, and political philosophy could continue to introduce racial topics only in the stages of applied theory. Obviously, when I say that race is an ontological category I am using ontology here to refer to basic categories of reality which are within history, at least partly produced by social practices, and which are culturally various. Race itself signifies differently and is lived differently between different discursive and cultural locations. This usage of ontology is controversial, and I cannot take the space here to justify it fully, but I will make one point. The problem with the social constructionist, anti-essentialist view that we should give up the language game of ontology altogether is that we are then left with a reduced ability to offer deep descriptions of reality, descriptions which can differentiate between more and less significant and persisting features of reality. The weakness of a strict social constructionist approach is that it tends toward flattening out all descriptive categories as having equal (non-)metaphysical status. Thus, for example, male/ female is put on the same plane as masculine/feminine, David Alfaro Siqueiros, Ethnography, 1939 and the importance of the biological division of reproduction is made analogous to gendered dress codes. In order to avoid this, without lapsing back into essentialism, the traditional ontological project or ascertaining certain basic categories can be reconfigured as the attempt to ascertain those elements of reality which, although mutable, currently intersect and determine a wide variety of discourses and practices, and thus are more fundamental not because of their ahistorical or transcendental status but because of their central intersectional position. The fact that race has lost its scientific credibility does not entail, then, that it has lost its ontological status, since on this usage ontology does not imply a reference in a transcendental reality. Race does not need to refer to a natural kind or a piece of reality in a metaphysical realist sense if it is to have any ontological meaning. What is race, then? Race is a particular, historically and culturally located form of human categorization involving visual determinants marked on the body through the interplay of perceptual practices and bodily appearance. Race has not had one meaning or a single essential criterion, but its meanings have always been mediated through visual appearance, however complicated. The criteria determining racial identity have included ancestry, experience, outside perception, internal perception, coded visibility, habits and practices - all 7 these and more are variously invoked for both individual and groups. The criteria which will be primarily operative vary by culture, neighbourhood, historical moment, so that some people place ancestry as all determining, while others make subjective identification the key. What is a philosopher to do in the face of this variation? We could take ordinary language, the way in which people speak of race, and use it to sift through these criteria to show which are most consistent with the way we speak. This approach could certainly be useful in pointing out contradictions between the way we speak and what we believe, and in showing the presuppositions we are implicitly committed to by the way we speak, but it cannot show us what the 'truth' of racial identity is. Phenomenological description of the experience of racial designations would also be useful in achieving a better understanding of the lived reality of race, but this again is not decisive in establishing the 'underlying essence' of race. My view is that the meaning of race will shift as one moves through the terrain and interplay of different discourses, where here discourses signify practices and institutions as well as systems of knowledge (a usage well exemplified in Wittgenstein' s concept of a language game, which involves linguistic practices connected with and embodied in actions). The 'answer' to the question of what racial identity really is will depend on what language game we are playing, although the relativism of this situation can be mitigated by showing overlaps between language games, and by offering immanent critiques that reveal internal contradictions, such as a language game that claims to be non-racist but actually is racist. Philosophy is a prime example of the latter. It has committed both crimes of omission - the neglect of race, and crimes of commission - correlating race with epistemic reliability (Kant) and potential for selfgovernment (Mill). But given this, we still have yet to understand either what racial identity is how we should articulate its relationship to philosophy. In order to answer these questions, we must first address several others, not only the scepticism toward race but also the postmodern critique of identity and of the visible, and the political debate over identity politics. It is these debates, more than the scientific status of race, that will determine the future of philosophical treatments of racial identity. I will summarize some of the relevant issues in these debates, and then try to address them in the context of race. The principal argument against identity politics has been that it assumes an essentialist, coherent identity that is efficacious over one's political orientation, epistemo- 8 logical standpoint and justificatory status. And so it might be thought that making racial identity epistemologically salient, for example, could lead to a reductionist form of evaluation that puts identity considerations over argument, that is, holding that that which is pious is so because the gods love it rather than the gods love it because it is pious. Also relevant is the critique of the tyranny of the visible, as in Rorty' s denigration of the visual metaphor for metaphysical realism, to Martin Jay's discussion of the anti -ocularcentric thrust of critical theory, to Jameson's blunt claim that 'the visual is essentially pornographic ... ' One thing that this view has in its favour is that it would make sense of the non-reciprocal visibility of dominant and non-dominant racial identities: where the invisibility of whiteness renders it an unassailable form, the visibility of non-whiteness marks it as a target and a denigrated particularity. And the concepts of identity and race are both often charged with assuming a unity and homogeneity that do not in fact obtain. Iris Young, building on Derrida and Adorno, criticizes the idea that identity is a coherent unity which can serve as the origin of thought and practice, and which can be neatly separated from external things such as others or discourses. lo This description is metaphysically incorrect, for reasons with which we are all familiar, having to do with the fundamental disunity of the self, its lack of complete self-knowledge, an~ its constitution by and through processes of narrativization which are only partially accessible to the subject Henri Matisse, The Hindu Pose, 1923 herself. lIOn the basis of these arguments, Young would have us make a parallel case against group identity, and reject identity concepts altogether. Racial identities are increasingly recognized as particularly disunified (since their group status is even more obviously arbitrary or conventional than nation or culture), split within by class, gender, sexuality, etc., and, as Engle and Danielson point out in their collection After Identity, without clear borders or a unifying internal essence. 12 Moreover, Freud argued that the effort to overcome disunity through collective identification or group solidarity may itself be the sign of a pathological condition caused by 'the inability of the ego to regain autonomy following the loss of an object of desire'. 13 Thus, the conclusion of these critiques is that racial identity is a dangerous illusion. Now, I take all of these worries about racial identity very seriously. My original entry into this area of work was motivated by a concern to understand and in some sense validate hybrid identity or hybrid positionality against purist, essentialist accounts. And the motivation for this was the felt alienation of having a mestizo identity (normative in Latin America and the Caribbean) but living in a purist culture (the USA), where racial categories are assumed to be mutually exclusive. In my nuclear family, which is anything but nuclear, I have a cholito Panamanian father (mixed Spanish, Indian, African), a white Anglo mother, and, through my father's multiple liaisons, a range of siblings from black to brown to tan to freckled, spanning five countries and three continents at the last count. This personal genealogy has not motivated me to try to repair dissonance into a coherent unity, but rather to understand the formation and position of the self precisely within an unresolvable heterogeneity. If I did not have any sympathy for the antiessentialist, my concern with the persistent paradox of the relevance of race would not be felt so strongly. It is because the arguments against racial identity have merit that the paradox is a paradox and not simply an error. But in the face ofthese anti-race arguments, we need a better position than one which merely relies on the withering away of racial categorization. And we need one that can do two things the anti-essentialist positions cannot do: (1) take into account the full force of race as a lived experience, understanding this not as mere epiphenomenon but as constitutive of reality; and (2) acknow ledge and account for the epistemological and theoretical importance racial perspective has had on, for example, the undermining of modernist teleologies (e.g. Du Bois' use of slavery to undermine US supremacist claims, and the Frankfurt School's critique of Western rationality from the perspective of the Holocaust). These facts suggest that we need to understand racial identity as having both metaphysical and epistemological implications. Race as identity Racial identifications have been causally associated with classical liberalism and philosophical modernism. Given the fact that the practices of racializing identity developed within the greatest period of colonialism and genocide the world has ever known, anti-racists have been understandably sceptical about the possibility of racial identity coexisting alongside equality and justice. Furthermore, the liberal conceptualizations of justice and enlightenment presupposed a decrease if not an end to the social relevance of racial particularity, and this can be traced out in the history of integrationist thought in the USA, as Gary Peller has so usefully shown: A commitment to a form of universalism, and an association of universalism with truth and particularism with ignorance, forms the infrastructure of American integrationist consciousness .... Integrationist beliefs are organized around the familiar enlightenment story of progress consisting of the movement from mere belief and superstition to knowledge and reason, from the particular and therefore parochial to the universal and therefore enlightened. 14 Where truth and justice require universalism, racial identity cannot be accorded salience without endangering progress. Racial identity threatens to return us to feudal hierarchy, a system in which identity determined one's life, which was precisely the system against which liberal enlightenment was organized and developed. As a result, anti-racism is assumed to require an anti-racial identity (at least in so far as race has political or non-trivial salience). Furthermore, an anti-racism that pursues universalism against particularism also 'confirms our sense of the possibility of true and authentic relations that transcend racial status and other forms of cultural distance and difference' . 15 It thus legitimates our perhaps natural hope for significant human relationships against ones that are necessarily deformed or atrophied by structurally produced separations. For whites and others who benefit in the present from a history of oppression, the appeal of universal racelessness may also lie in its ability to deface their/our race-based connections with that unpleasant past; in other words, it may entitle whites to believe they/we don't need to acknowledge the salience of white identity and thus to avoid the moral 9 discomfort that that identity cannot help but present. But there is an argumentative complicity - whether intentional or not doesn't matter of course - between the suspicion against the visible, against identity, and certainly against the intersection of these which would occur in a racialized conception of identity, and the continuing inattention to race matters in philosophy and political theory. As many people have pointed out, one of the persistent problems with the discourses in the USA around multiculturalism and cultural studies is that race, racism and racial hierarchies are relatively ignored. Explorations of culture and ethnicity can all too easily avoid any account of white supremacy and focus instead on the recognition of difference, flattening out differences in a way that makes them appear equal. Race, on the other hand, is difficult to focus on for very long without it working to discredit the imagined landscapes of pluralist difference that cultural studies so often presupposes. And mainstream political language in both Britain and the USA codes racial talk as cultural talk, so racist claims can be cloaked as claims about cultural difference. Interestingly in this context, Lewis Gordon's recent book, Bad Faith and Antiblack Racism, argues that, in an antiblack world, blackness signifies absence, the absence of identity in the full sense of a self, a perspective or a standpoint with its own self-referential point of view. 16 In other words, what is denied black people is the ability to wield the Look, to be a source of value and meaning. The infamous three-fifths formulation from the US Constitution might be explicated as a concept of black personhood as having a consciousness without judgement, or a limited capacity for affective sensibility and cognitive distinctions. Antiblack racism denied visible black people the standpoint of a subject as capable of judging and knowing and reciprocating in an intersubjective relationship between persons. Charles Mills argues in his essay 'Non-Cartesian Sums: Philosophy and the African American Experience' that the concept of 'sub-personhood', or Untermensch, is a central way to understand 'the defining feature of the African-American experience under conditions of white supremacy (both slavery and its aftermath)'.17 By this concept, which he develops through a contrast drawn between the Cartesian sum and Ralph Ellison's invisible man, Mills elucidates the comprehensive ramifications that white racism had on 'every sphere of black life - juridical standing, moral status, personal/racial identity, epistemic reliability, existential plight, political inclusion, social metaphysics, sexual relations, aesthetic worth' .18 To be a sub-person is not to be a non-person, or an 10 object without any moral status whatsoever. Rather, Mills explains, the peculiar status of a sub-person is that it is an entity which, because of phenotype, seems (from, of course, the perspective of the categorizer) human in some respects but not in others. It is a human (or, if this seems normatively loaded, a humanoid) who, though adult, is not fully a person ... [and] whose moral status was tugged in different directions by the dehumanizing requirements of slavery on the one hand and the (grudging and sporadic) white recognition of the objective properties blacks possessed on the other, generating an insidious array of cognitive and moral schizophrenias in both blacks and whites. 19 On the basis of this, Mills suggests that the racial identity of philosophers affects the 'array of concepts found useful, the set of paradigmatic dilemmas, the range of concerns' with which they each must grapple. He also suggests that the perspective one takes on specific theories and positions will be affected by one's identity, as in the following passage: The impatience, or indifference, that I have sometimes detected in black students [taking an ethics course] derives in part, I suggest, from their sense that there is something strange, for example, in spending a whole course describing the logic o~ different moral ideals without ever talking about how all of them were systematically violated for blacks.20 The result is an understanding that black lived experience 'is not subsumed under these philosophical abstractions, despite their putative generality' .21 As a further example of Mills' claim, consider the following passage from Hannah Arendt: In America, the student movement has been seriously radicalized wherever police and police brutality intervened in essentially nonviolent demonstrations: occupations of administration buildings, sit-ins, et cetera. Serious violence entered the scene only with the appearance of the Black Power movement on the campuses. Negro students, the majority of them admitted without academic qualifications, regarded and organized themselves as an interest group, the representatives of the black community. Their interest was to lower academic standards. They were more cautious than the white rebels, but it was clear from the beginning (even before the incidents at Cornell University and City College in New York) that violence with them was not a matter of theory and rhetoric. Moreover, while the student rebelling in Western countries can nowhere count on popular support outside the universities, and as a rule encounters open hostility the moment it uses violent means, there stands a large minority of the Negro community behind the verbal or actual violence of the black students. Black violence in America can indeed be understood in analogy to the labor violence in America a generation ago; and although ... only Staughton Lynd has drawn the analogy between labor riots and student rebellion explicitly, it seems that the academic establishment, in its curious tendency to yield more to Negro demands, even if they are clearly silly and outrageous, than to the disinterested and usually highly moral claims of the white rebels, also thinks in these terms and feels more comfortable when confronted with interests plus violence than when it is a matter of nonviolent 'participatory democracy'.22 The ambivalence Mills points to can be discerned in this account. On the one hand, black students are clearly persons, having a self-interested perspective which they pursue through collective action, and capable of greater collectivity across campus and community divisions than the white students. On the other hand, this perspective is less intelligent, hence its desire to lower standards; and (probably as a result) it is too self-interested, too Frida Kahlo, Self-Portrait with Cropped Hair 1940 particular, and thus unable to achieve the moral approbation of the purportedly disinterested white rebels. Arendt clearly pits morality against self-interestedness, the universal against the particular, once again. But the result is a curious replay of the liberal antinomy between having a racialized self and having a less developed self; between being a person with a perspective and being a non-person precisely because ofthat perspective: having 'demands' but demands which are silly, outrageous, and pursued through what she clearly considers unnecessary violence. There is no question that Arendt's white racial identity affected her ability to assess black student actions, or that her response to the possibility of blackorganized violence was affected by her identification with its targets. This is so obvious as to be uninteresting. But does this judgement entail the reductionist evaluations imputed to adherents of identity politics? Are we forced into holding that Arendt's views can be reduced to a consideration of Arendt's race? Or, if we want to avoid such a position, are we forced to conclude that her race was irrelevant to the above account? It seems clear to me that racial identity is a crucial category of analysis to have at our disposal in order to understand Arendt's reactions to and assessment of black students. Yet I believe we can retain this category without essentializing racial identity or reducing philosophical this case analysis to racial identification. I will devl~p through a reading of Paul Gilroy's The Black Atlantic: Modernity and Double Consciousness. 23 This book has a twin purpose. On the one hand, Gilroy's purpose is to reconfigure and reconceptualize the concept of black identity so important to cultural studies, black studies and Afrocentric theory, in such a way that he can avoid the metaphysical criticisms of prior concepts of identity and develop a more adequate accounting of the cultural formations and political practices created under diaspora conditions than Afrocentric theories can explain. As Gilroy tells the story, there is an identifiable cultural formation organized by the black diaspora and existing in multiple sites which he groups together under the term 'black Atlantic'. Given the internal cultural, linguistic and geographic heterogeneity of this group, to call it a 'culture' would be actually misleading and more evocative of homogenization than the term 'black'. Moreover, the racial designation more accurately signifies the principal organizing logic of this group, which was and is the historical experience of an institution of slavery that operated through phenotype. This experience has yielded an ongoing process of identity formation that cannot be traced back to an African essence or distilled into its pure 11 concepts of reason and liberation to critique its practices and its self-understandings will not go deeply enough. In his readings of Du Bois, Richard Wright and others, Gilroy claims to be able to trace the formations of a vernacular variety of unhappy consciousness which demands that we rethink the meanings of rationality, autonomy, reflection, SUbjectivity, and power in the light of an extended meditation both on the condition of the slaves and on the suggestion that racial terror is not merely compatible with occidental rationality but cheerfully complicit with it. 26 Cheri Samba, Self-Portrait, 1991 type, but that is persistently involved in the proliferation of ever new hybrid identities. Thus, Gilroy's analysis is both centred around identity and insistent on the fundamental hybridity and openness of identity.24 Against those who would emphasize the enduring manifestation of roots in black culture, and against the association of black liberation with a return in some sense to those roots, Gilroy uses the imagery of the diaspora precisely to articulate a mobile and mediated identity, internally heterogeneous, and whose very survival and ability to flourish has been predicated on its character as always open to new mutation. In opposition to ... nationalist or ethnically absolute approaches, I want to develop the suggestion that cultural historians could take the Atlantic as one single, complex unit of analysis in their discussions of the modern world and use it to produce an explicitly transnational and intercultural perspective. 25 His choice of the word 'produce' rather than 'discover' is clearly intentional. The second major purpose of this book is to show how this perspective has been and can be brought to bear on an account of modernity generally. Along the lines of Mill's argument above, Gilroy claims that a critique of modernity which is entirely immanent is insufficient. That is, a critique which uses the Enlightenment's 12 Thus, it is through 'the slaves' perspective' that a more thoroughgoing critique of the Enlightenment can advance. This perspective begins from a more sceptical position on 'the democratic potential of modernity' than, for example, Jiirgen Habermas is said to have. It would insist that Columbus accompany Luther and Copernicus as the standard bearers of modernity, with all the repercussions that must then follow concerning how we assess that standard. Locke's Second Treatise could no longer be taught without a mention of his contribution to writing the Carolina slave constitution. 27 Gilroy uses Frederick Douglass' s slave memoirs to suggest a revision of Hegel's Lord and Bondsman narrative, wherein it is the slave that 'actively prefers the possibility of death' rather than the master. Douglass' s version reveals the prior structure of enslavement which mandates 'the· slave's survival in bondage over the possibility of death, and locates the slave's first moment of agency in his determination to counter violently the violence which has already been inscribed in the social relation. This retelling of the narrative more correctly locates the origin of institutional violence as prior to the slave's enslavement, and thus raises 'queries about the assumption of symmetrical intersubjectivity' which grounds so many modernist accounts of self-formation. Gilroy's point is not to draw a sharp border between slave and non-slave perspectives, and at one point he even aligns Habermas with 'a good many ex-slaves' in his commitment to 'making bourgeois civil society live up to its political and philosophical promise' .28 But, although there is much intermixture and overlap between perspectives, they are not all coextensive, and one can shift the horizons of visibility by occupying the centre as opposed to the periphery of a black Atlantic perspective. This understanding of identity in terms of perspective suggests a definition of identity as a social location, a location within a social structure and marked vis-a.-vis other locations which gives the identity its specificity rather than its internal characteristics. 29 Gilroy is not arguing here that the perspective engendered by identity has a singularly deterministic effect on thought. He rejects Patricia Hill Collins' s 'collapse' of being and thinking 'so that they form a functional unity that can be uncritically celebrated' .30 And he suggests, rightly, that a determinist view of the impact of identity on thought would inhibit the scope of critical reflection on that thought. Moreover, he argues that such an account of knowledge would 'simply end up substituting the standpoint of black women for its forerunner rooted in the lives of white men', simply replacing white men with black women in the myth of 'stable, ideal subjects' .31 And as a myth, postulating a convenient but specious concept of the self, such an account cannot last very long. However, despite his hybrid, postmodern-influenced, problematized notion of identity, for Gilroy 'identity' and in this book it is racial identity he is exploring remains the central term of his analysis. He repeatedly criticizes those whose critique of racial essentialism leaves them 'insufficiently alive to the lingering power of specifically racialised forms of power and subordination' or those who have been 'slow in perceiving the centrality of ideas of race and culture' to the investigation of modernity. 32 And he repudiates theories of the self which would, like Marshall Berman' s, try to conceptualize it at a more abstract, more putatively universal level, below the effect of racial configurations. The Black Atlantic does a masterful job arguing against purist, nationalist paradigms by showing how these cannot account for what is essentially an 'intercultural and transnational formation' .33 It makes the unifying theme not an internal core or original historical moment or homogeneous cultural elements, but the 'well-developed sense of the complicity of racialised reason and white supremacist terror' which provides a perspective informing literary, musical and philosophical creativity.34 In this way, hybridization and identity can coexist, at least as long as global white supremacy continues to structure intersubjective relations through racialized identities. Gilroy's book also serves as an empirical rejoinder to the metaphysical arguments of postmodernism, which set a priori limits on the plasticity of identity concepts. 35 Gilroy's argument comes out of specific analyses of cultural products, rather than accounts of the limits of language, and demonstrates the usefulness of a unifying concept like the Black Atlantic to understand and appreciate a wide range of forms. There are others besides Gilroy who are making similar moves. Kobena Mercer's recent collection of essays exhibits the same reluctance either to embrace nationalist or Afrocentric treatments of black identity or to dispense with identity as irrelevant. 36 And, like Gilroy, Mercer negotiates between these conceptions through a diasporic aesthetic, which relies on analogous positionality and historical experience rather than a deep self or unified politics to establish identity. Roberto Fernandez Retamar's 'Caliban' is another example, proposing Caliban, a figure from English literature, as the symbol of Latin Amer:ican identity. Fernandez Retamar's account this exemplifies what Trinh T. Minh-ha calls the fearless affirmation of the hyphen, as well as a conception of identity primarily in positional terms: To assume our condition as Caliban implies rethinking our history from the other side, from the viewpoint of the other protagonist. [Quoting Marti:] 'We must stand with Guaicuipuro, Paramaconi [heroes of Venezuela, probably of Carib origin], and not with the flames that burned them, nor with the ropes that bound them, nor with the steel that beheaded them, nor with the dogs that devoured them. '37 I would argue that the concept of identity found in these works, as in Gilroy, Mills and Gordon, is a concept not organized around a claim to sameness, which is what invites much of the criticism of identity concepts. Rather, as in Gordon's diagnosis of racism as positing an absence, what the claim of identity here is organized against is the assumption of lack. In this context identity is put forward not as sameness opposed to difference but as substance opposed to absence. It is also opposed to notions of the self which formulate it primarily as an abstract form without content, a decontextualized ability to reason without any interested positionality. Examples here would be the Cartesian sum, a self as a thinking, abstract process or ability, and the early Sartrean model of the self as the ability to negate. Against such contentless models, works such as Gilroy's could be understood as consistent with a more substantive understanding of the self which can sustain particular identities in a way I have time here just to sketch. From Bourdieu one might take the concept of the self as a sedimentation of dispositions and practices developed through a personal history, and understand that history in terms of an experience which is always carried forth even if interpreted anew (as in the later Sartre). Racial identifications will affect the particular manifestation of both these elements (practices, experience), but in order to account more fully for race we must also include the element of visibility, as an embodied manifestation that invites and elicits determinate though contextually variable meanings. On this kind of account, race can be understood to 13 figure in identity formation not as a metaphysical necessity but as a necessity within a given historical context. And from here one might go on to develop a phenomenology of racial identity as, for example, a differentiation or distribution of felt connectedness to others. This will necessarily be a complex issue, undetermined solely by phenotype. The felt connectedness to visibly similar others may produce either flight or empathic identification or other possible dispositions. Gilroy's description of the black Atlantic identity has the power to incorporate this openness and constant mixture with the connecting elements of a post-slavery diasporic perspective, such that phenotypic race is never sufficient yet never completely absent. This provides us with a metaphysically more accurate, and politically less problematic, formulation of racial identity, not based on purity or the continuity of original essence, and one not closed to new incarnations. I began this paper with the example of a non sequitur exchange, where an Inuit questioned the point of mapmaking and a white man responded with a reassurance of accuracy. The assumption in some of the anti-identity dismissals of race seems to take the form of an inverse of this exchange, such that a denial of the possibility of accuracy is somehow taken to entail a denial of the possibility of maps. My argument would be that this response is no less of a non sequitur than the one before. Notes 1. David Theo Goldberg, Racist Cultures: Philosophy and the Politics of Meaning, Basil Blackwell, Oxford, 1993; Comel West, Prophesy Deliverance!, Westminster Press, Philadelphia, 1982. 2. West, Prophesy Deliverance!, p. 55. 3. Goldberg, Racist Cultures, p. 6. 4. Michel Foucault, The Archaeology of Knowledge, trans. A. M. Sheridan Smith, Pantheon, New York, 1982, p. 45. 5. Michel Foucault, The Birth of the Clinic, trans. A. M. Sheridan Smith, Vintage Books, New York, 1975, p. 30. 6. See also Adrian Piper, 'Passing for White, Passing for Black', Transitions 58, pp. 4-32. 7. Anthony Appiah, In My Father's House: Africa in the Philosophy of Culture, Oxford University Press, New York, 1992, pp. 32,45, and chs 2 and 3 generally. 8. Richard Rodriguez, Days of Obligation: An Argument with My Mexican Father, Viking, New York, 1992, p. 1. 9. See, for example, the special issue of Identities: Global Studies in Culture and Power on '(Multi)Culturalisms and the Baggage of "Race'" edited by Virginia Dominguez, vol. 1, no. 4, April 1995. 10. Iris Young, 'The Ideal of Community and the Politics of Difference', in Linda Nicholson, ed., Feminism/ Postmodernism, Routledge, New York, 1990, pp. 303-5. 11. For a recent development of this idea, see Paul Ricoeur, Oneself As Another, trans. Kathleen Blarney, University of Chicago Press, Chicago, 1992. 12. Dan Danielson and Karen Engle, 'Introduction', in D. 14 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. Danielson and K. Engle, eds, After Identity: A Reader in Law and Culture, Routledge, New York, 1994, pp. xiiixix. Michael Steinberg, "'Identity" and Multiple Consciousness', paper delivered at the workshop 'Identity: Do We Need It?', sponsored by the Internationales Forschungszentrum Kulturwissenschaften, Vienna, Austria, May 1995. Gary Peller, 'Race Consciousness', in Danielson and Engle, eds, After Identity, p. 74. Ibid., p. 76. Lewis Gordon, Bad Faith and AntiBlack Racism, Humanities Press, Atlantic Highlands, NJ., 1995, esp. ch. 14. Charles Mills, 'Non-Cartesian Sums: Philosophy and the African-American Experience', Teaching Philosophy, vol. 17, no. 3, October 1994, p. 228. Ibid. Ibid. Ibid., p. 226. Ibid., p. 225. Hannah Arendt, On Violence, Harcourt, Brace, Jovanovich, New York, 1969, pp. 18-19; quoted in Gordon, Bad Faith, pp. 88-9. Paul Gilroy, The Black Atlantic: Modernity and Double Consciousness, Verso, London, 1993 and Harvard University Press, Cambridge, Mass., 1993. Ibid., p. xi. Ibid., p. 15. Ibid., p. 56. See Mills, Neo-Cartesian Sums, p. 226. 28. Gilroy, The Black Atlantic, p. 49. 29. See my 'Cultural Feminism vs. Poststructuralism: The 3, Identity Crisis in Feminist Theory', Signs, vol. 13, no~ 1988, pp. 405-36, for further elucidation of this point. 30. Gilroy, The Black Atlantic, p. 52. 31. Ibid., p. 53. He calls Hill Collins's account of knowledge 'experience centred', meaning to differentiate it from theoretical critique, but I would take issue with this characterization. One could have an experience-centred account of knowledge, or at least one that emphasizes the importance of experience, without either a transparent view of experience or an anti-theoretical disposition. The complex history of twentieth-century radical empiricism as well as the phenomenological tradition are counterexamples to this assumption. 32. Gilroy, The Black Atlantic, pp. 32 and 49. 33. Ibid., p. ix. 34. Ibid., p. x. 35. For a similar argument against setting a priori limits on the plasticity of social practices, see Judy Butler's review essay 'Poststructuralism and Postmarxism' in Diacritics, Winter 1993, pp. 3-11. Here she is critiquing Ernesto Laclau's 'description of the logical features by which any social practice proceeds' on the grounds that it postulates 'a logic to which social practices are subject but which is itself subject to no social practice' (p. 9). I would make a related claim that we cannot determine in advance, outside of social practice, the 'logic' of identity concepts, or their inevitable political effects. 36. Kobena Mercer, Welcome to the Jungle: New Positions in Black Cultural Studies, Routledge, New York, 1994. 37. Roberto Fernandez Retamar, Caliban and Other Essays, trans. Edward Baker, University of Minnesota Press, Minneapolis, 1989, pp. 16 and 19.