Les ateliers de l'éthique The Ethics Forum The Invisibility of Privilege: A critique of intersectional models of identity

IIn this paper, I argue that intersectionality, the prevailing way of conceptualizing the relation between axes or systems of oppression (race, class, gender), illicitly imports the very model it purports to overcome: that is, the unitary model of identity. I first define “intersectionality” and distinguish between three senses that are frequently conflated. Then I subject the model to an analytic critique, revealing its hidden presuppositions about identity. Finally, I suggest that solidarity serves as a better norm for feminist practice than inclusion of “difference,” which seems to be the norm underlying many intersectional accounts.

The problem of conceptualizing political subjectivity, in an ethical and politically productive way, is an old and abiding one. It is a task that seems particularly urgent for oppositional politics, which, in one way or another, seek to transform subjectivity-in-itself into subjectivity-for-itself as a means to social emancipation. It is the broad claim of this paper that feminist theory and politics has failed to decisively address itself to this task, to constitute the subject of feminism in a truly "universal" way, 2 with the result that the emancipatory project of feminism remains -both in practice and in theory -glaringly incomplete. This is not for lack of trying: antiracist feministsmajoratively racialized women -have exposed this lacuna since the earliest enunciations of falsely universal feminist politics. They have expended an unreciprocated amount of intellectual and psychic effort articulating to race-privileged feminists what it means to be confronted, as Anna Julia Cooper put it in 1892, by both "a woman question" and a "race problem"; 3 that is, in a more contemporary political idiom, what it means to be subject to "double jeopardy," 4 "multiple jeopardy," 5 "multiple oppressions" 6 or "interlocking oppressions." 7 The most recent metaphor for the problem Cooper spoke of in 1892 is that of "intersectionality." Intersectionality "is currently the reign-ing…metaphor for complex identities insofar as they are constituted by race, ethnicity, class, and sexual orientation together with gender." 8 In this paper, I discuss this currently prevalent way for conceptualizing the political subjectivity of the hyper-oppressed. Notably, postidentitarian feminists as well as identitarians 9 deploy the language of intersectionality, and in feminist circles the term "intersectionality" has come to stand virtually as a synonym for the way in which the "litany" of oppressions (based on axes of gender, race, class, sexuality, and dis/ability) inflect and inform one another -to the extent that common usage makes it acceptable for one to use the term "intersectionality" without specifying what, in particular, is intersecting, or how. Yet, as Ladelle McWhorter points out, inasmuch as "the precise nature of such alleged 'intersections' is not made clear," it is an open question whether, in many analyses, reference to intersectionality functions as anything "other than just a strategy to avoid charges of racism or classism." 10 That is, perhaps, a stronger claim than I argue for in this paper. My aim here is more modest. I argue that the intersectional model of identity fails in its two primary analytical aims: first, to render visible the experience of hyper-oppressed subjects; and sec-ond, to supplant the normative, race-and class-privileged subject of feminist theory and politics. 11 Exponents of the intersectional model claim that it is particularly apt in capturing the experience of hyper-oppressed people, paradigmatically racialized women, in ways that its predecessors, namely the unitary conception of "woman" and "additive" models of identity, have failed to. It is maintained that, as a heuristic device, intersectionality enables a nuanced view of the ways in which axes of privilege and oppression inflect and mutually construct each other. 12 Its proponents claim that the intersectional model of identity therefore successfully guards against a politically problematic reduction of racialized women's social experience to that of racialized men, or to that of race-privileged women; consequently, the intersectional model enables the representation (in both the descriptive and political senses) of subjects oppressed on "multiple" axes. Upon giving a brief exposition of the model of intersectional identity as it manifests itself in the analyses of Kimberlé Williams Crenshaw (1991), Patricia Hill Collins (2003), and Diana Tietjens Meyers (2000) ( §1), I argue that this model inadvertently reproduces the very assumptions it claims to be redressing ( §2). In particular, I advance the claim that the intersectional model fails to make necessary substantive revisions in how we conceive the relation between oppressions, and as such fails to deliver on its promise of overcoming the conceptual errors of past models of identity. In the final section of the paper ( §3), I raise one implication of my discussion: namely, that feminist subjectivity -the ground of solidarity among "women" -does not precede but rather emerges out of concrete political practice, which involves a collective, practical confrontation with relations of privilege and oppression. crete) oppressions. Whereas monistic analyses of oppression conceive of these power relations in binary terms -"men rule women…Whites dominate Blacks" 14 -the intersectional model "references the ability of […] race, class, and gender [oppressions] to mutually construct one another." 15 This first sense of intersectionality -intersectionality of oppressions -is deployed by post-identitarian theorists as well as by identitarians. 16 An intersectional conception of oppression is often distinguished, however, from an account of systems of oppressions that theorizes them as "interlocking." Leslie McCall articulates the difference between intersectionality and interlocking of oppressions in terms of a distinction between "intracategorical complexity" and "intercategorical complexity." 17 The latter "focuses on the complexity of relationships among multiple social groups within and across analytical categories and not on complexities within single social groups [or] single categories." 18 Sherene Razack concretizes this difference in analytical emphasis: Analytical tools that consist of looking at how systems of oppression interlock differ in emphasis from those that stress intersectionality. Interlocking systems need one another, and in tracing the complex ways in which they help to secure one another, we learn how women are produced into positions that exist symbiotically but hierarchically. We begin to understand, for example, how domestic workers and professional women are produced so that neither exists without the other. First World policies of colonialism and neo-colonialism, which ultimately precipitated the debt crisis and the continuing impoverishment of the Third World and enabled the pursuit of middle-class respectability in the First World, were implemented in highly gendered ways. 19 In other words, an analysis of the way that systems of oppression "interlock" has as its point of focus the matrix of micro-and macropolitical relations that produce subjects, whereas intersectional analysis focuses on the subjects produced by those relations, conceived of in identic terms. The distinction between "interlocking" and "intersectional" analyses may appear quibbly at first glance, and it is not a stable one across the literature, but proponents of the "interlocking" approach claim they are different in kind. McWhorter argues that the conception of the relation between oppressions as interlocking predates the appearance of the intersectional model of identity. Saying that various forms of oppression are "interlocking," in McWhorter's view, means that they cannot be separated in the lived experience of hyper-oppressed subjects; that is, that "race, sex, and class are 'simultaneous factors'" in the lived experience of oppression. 20 Yet, as we will see below, this is what intersectionality also purports: that the intersectional model describes the simultaneity of oppressions in lived experience. But it should be noted that lived experience is not, originally, the terrain that the metaphor of intersectionality sought to map out: the language of "intersections" emerges from Crenshaw's ground-breaking intervention in race-critical feminist legal theory, which sought to expose the inadequacy of monistic remedies to racialized gender-based discrimination. Crenshaw argued that the discrimination that Black women face cannot adequately be captured by monistic conceptions of discrimination inscribed in U.S. antidiscrimination law. 21 The other two senses of "intersectionality" are introduced and explicitly elaborated by Crenshaw."Structural intersectionality" refers to the model of political identity that, according to Crenshaw, aptly captures the particular social location of the hyper-oppressed, paradigmatically Black women. The claim is that as an analytical model structural intersectionality illuminates the simultaneity (as well as the "complexity" and "irreducibility") of race and gender oppression in racialized women's experience. Finally, the third sense of intersectionality, which Crenshaw calls "political intersectionality" describes the empirical socio-historical fact that feminist and antiracist politics have functioned in tandem to marginalize the issues facing Black women. 22 According to identitarian theorists who adopt the intersectional model of identity, these three kinds of intersectionality I have just described are related, though there is some disagreement around the nature of this relation. Causal priority is variously attributed to structural intersectionality (Crenshaw), or to intersectionality of oppressions (Collins). In the first instance, the account runs like this: subjects have identities on the basis of which they are oppressed or discriminated against. Oppositional movements fail to represent these subjects when they reduce the "complexity" of their identities by deploying a single category of political analysis (e.g., gender). For a materialist like Collins, political relations which produce the intersectionality of oppressions have causal priority: oppression produces the intersectional identity of the hyper-oppressed subject, i.e., structural intersectionality. This subject, by virtue of the "complexity" of her experience of oppression is not adequately represented by oppositional movements that rely on monistic analyses of oppression which privilege the identities and political interests of subjects oppressed on a singular axis. On both accounts political marginalization followsontologically and narratively -the formation of the identity of the hyper-oppressed subject, who by virtue of her identity is politically marginalized. Intersectional subjects come, so to speak, already formed to the space of political contestation. But this construal of the relation between the phenomenon of oppression, political subjectivity, and oppositional politics is not unproblematic. We might imagine an alternative account in which political intersectionality (i.e., the problem of political representation) precedes, and gives rise to both structural intersectionality (i.e., political identity) as well as to the intersectional conception of the relation between relations of oppression. 23 (I advance this claim in §2.) On Crenshaw's view, the intersectional model of identity mitigates against the phenomenon of political intersectionality. Crenshaw argues that Black women are, in virtue of their membership in "at least two subordinated groups that frequently pursue conflicting political agendas" inadequately represented by both. 24 But, according to Crenshaw, the problem is not "just" that "both discourses fail [Black women] by not acknowledging the 'additional' burden of patriarchy or of racism" 25 -as theorists of the additive model of political identity once purported. Rather, insofar as gender oppression is construed as the oppression suffered by white women and race oppression as that suffered by Black men, the resources of these discrete discourses are simply inadequate to the task of conceptualizing the gender oppression and the race oppression suffered by Black women. Black women, in Crenshaw's terms, are consigned to the "intersection," "a location that resists telling." 26 But if the intersection resists telling, those systematically and concretely located there have resisted the intersection. 27 Racialized, minoritized, working-class, lesbian and majority-world women have long protested their relegation to the margins resulting from the exclu-sions of race-and class-privileged feminism. The unitary conception of "women" had achieved notoriety in academic feminism by the early 1980s for the false universalization of white bourgeois women's experience on which it was predicated. However, this critique is not recent, even if the (white) feminist mainstream only recently came to acknowledge it, to the extent that it has. Albeit belatedly, the unitary conception of "women" was conclusively deemed inadequate to the task of properly representing concrete women, variegated as they are by micro-and macro-relations of racialized class and implicated as they are in geopolitical relations of imperialism and (neo)colonialism. But it was not mere empirical inadequacy that underwrote the rejection of a unitary conception of "women" as the basis of feminist politics: it was a charge of exclusion and of subordination -an ethicopolitical charge -levied against white bourgeois feminists, a charge that entails a call to confront their race-and class privilege, and the ways in which it informs their analysis of gender oppression and constitutes the parameters of their political practice. What this call amounts to then, is, first, an epistemological demand to make transparent -to make visible -the way in which the particular social experience of certain women has been falsely universalized as the experience of all women. But this critique is motivated by a second demand, this one political, namely, to bring this critical epistemology of privilege to bear on feminist praxis. Meeting the second, "political" demand entails actually transforming feminist practice. For reasons that I will elaborate in §3, it is important to distinguish what I am calling the epistemological task from the political one. For if white bourgeois feminists have made efforts to respond to the first demand, through conceptual moves like the adoption of putatively non-unitary models of identity, it is not clear that this has translated into a significant transformation of political practice on the ground, nor its prerequisite, an uncompromising interrogation of one's own privilege and complicity in structures of domination. As Razack puts it, merely "[s]peaking about difference […] is not going to start the revolution." 28 Indeed, I think there is something problematic about models of identity which proceed by "laying the groundwork for handling difference as the real problem, instead of the power relations that construct difference." 29 And if, as I want to claim, political identities do not precede, but issue from oppositional political practice, it is not clear that merely conceptualizing political identities -or "difference" -differently is sufficient to realize the political demand that feminists transform their falsely universal, parochial politics into truly universal politics of solidarity that work to undermine systems that produce social contradictions. (I will return to this in §3.)

THE INVISIBILITY OF PRIVILEGE
Until now, I have been speaking rather abstractly about the intersectional model of identity. To help concretize this model, I want to examine the illustrations that its exponents provide. It has been suggested that intersectional identities may be visualized as "points on a plane plotted in relation to axes […] of oppression such as racism and sexism"; 30 a metaphor more evocative of the injurious nature of that social location, is one that compares "sexism and racism [to] vehicles on a collision course [… which] converge at an intersection where a pedestrian who is crossing cannot get out of the way." 31 Albeit less dramatically, the intersectional model of identity can also be schematized with a Venn diagram. Meyers warns that "if visualized as lifeless Venn diagrams, [the] trope [of intersectionality] represents inert items sorted into various overlapping categories." 32 But I submit that the "lifelessness" of the schema is a consequence not of the illustration, but of the model of identity it serves to illustrate. In this sense, the Venn diagram is actually a particularly apt schematization, insofar as it reveals the inadequacy of the intersectional model to render the experience of oppression as it is lived. 33 In particular, much like the "additive" model of oppression, the intersectional model is impotent to represent the irreducibility of the oppression uniquely facing Black women, and the simultaneity of compounded oppressions that Black women share with other groups. I hope to show that the criticism that Angela Harris makes of additive models of oppression applies to the intersectional model as well: both effectively reduce the lives of people who experience multiple forms of oppression to addition problems: 'racism + sexism = straight black women's experience' or 'racism + sexism + homophobia = black lesbian experience' […] black women's experience [is] forcibly fragmented before being subjected to analysis, as those who are 'only interested in race' and those who are 'only interested in gender' take the separate slices of our lives. 34 Harris' targets here are "feminist essentialists" who in their theorizing and in their legal and political practice deploy categories of "race" and "gender" in a way that presumes that it is possible to describe or account for the latter in isolation of the former. 35 But I want to suggest that Harris' critique has a wider extension than her own parsing of the problem suggests. In particular, it extends beyond conceptualizations of gender oppression which aspire to totality, and arguably applies to those -like the intersectional model -which ostensibly privilege "difference." Harris warns that "as long as feminists [...] continue to search for gender and racial essences, black women will never be anything more than a crossroads between two kinds of domination [...] we will always be required to choose pieces of ourselves to present as wholeness." 36 But this is precisely what the intersectional model of identity trumpets as its theoretical strength: that is, its professed ability to articulate the experience of subjects located at the intersection -the crossroads -of race and gender oppressions.
In order to flesh out this problem, let us look at a schematization of the intersectional model of identity (see Appendix A, Figure 1). This model purports to be an intervention in theorizing political identity that aims to articulate "a location that resists telling." The claim that the identity of the Black woman is produced by the intersection of gender and race is viable only if we can think "Black" without thinking "woman," and if we can think "woman" without thinking "Black." On the intersectional model, if the Black woman occupies the border, who takes residence -who is at home -in the non-intersecting zones? Not only does the model fall prey to the widespread practice of aligning certain groups with particularity, or "visibility," and others with universality, or "invisibility"; upon closer inspection, it becomes apparent that the intersectional model of identity conjures the very ontology that its exponents set out to undermine. In construing race and gender as analytically separable and relegating Black women to their intersection, it implicitly perpetuates the racialization of gender and the gendering of race. (see Appendix A, Figure 2). To put it another way, on the intersectional model, "race oppression" unmodified is implicitly conceptualized as the oppression suffered by racialized men, and "gender oppression" unmodified is implicitly con- ceptualized as that to which race-privileged white women are subject. To say that "race" and "gender" oppressions intersect in the experience of the racialized woman is to preserve a unitary conception of oppressions the normative subjects of which are relatively privileged on some axes. It should be noted that although racialized women are the paradigm intersectional subjects, it is claimed that, in fact, all individuals in invidiously stratified societies have intersectional identities. All individuals have intersectional identities produced through the convergence and mutual inflection of relations of oppression and privilege (on Collins' view) or through the confluence of "identity determinants" (on Crenshaw's and on Meyers' view). However, as Meyers notes, insofar as "our discourse exaggerates the significance of some group memberships," making certain identity determinants "mandatory and salient categories of self-description," while rendering other determinants invisible, it seems that identities are viscerally experienced as intersectional only by those who are hyper-oppressed. 37 As Crenshaw, Collins and Meyers recognize, the asymmetry in discursive emphasis on certain "identity determinants" is a function of privilege. Not only do people privileged on axes of race or gender consistently fail to identify themselves with their race (white) or their sex/gender (male/masculine), but, moreover, and what in practice amounts to the same thing, they often fail to understand themselves as racialized or as gendered. As Collins puts it, "[c]urrent assumptions see African-Americans as having race, White women as having gender, Black women as experiencing both race and gender, and White men as experiencing neither." 38 These are neither accidental, nor innocent omissions: indeed, they constitute the wages of privilege. Of course, Crenshaw, Collins and Meyers are explicitly critical of such tendencies. But it is not clear that the intersectional model of identity overcomes them. In fact, it is my suggestion that the intersectional model effectively collapses, under the weight of criticism, into the unitary model, despite the very different intentions motivating each.
To the extent that a dualistic logic continues to inform our thinking about race and gender, it is still, on the intersectional model, implicitly the masculine that poses as the generic Black person, and the white that poses as the generic woman. The ostensible mutual exclusivity of the categories of race and gender is the condition for the possibility of their intersecting; but this is a function of the invisibility of their respective gendering and racialization. These categories can be said to be analytically pure only insofar as they are unmarked, uninflected, by one another; but this purity is an impossibility, even as an abstraction (for what would it mean to say, under prevailing social conditions, that one can conceive a racialized body that was not s to the extent that it presupposes the analytic purity of the very categories it seeks to intersect, inadvertedly reproduces this construction of whiteness and of masculinity. Let us try to test this claim by interrogating the notion that all subjects in stratified societies have intersectional identities. What happens when one attempts to schematize the (putatively) "intersectional" identity of the white man (the paradigmatic "unified" subject)?(See Appendix A, Figure 3.) I would suggest that the claim that the white man is produced by the intersection of whiteness and maleness fails to induce a response symmetrical to that generated by the claim regarding the identity of the Black woman. If there are any implications to be deduced from the attribution of an intersectional identity to the white man, they are that whiteness is implicitly gendered masculine, and that maleness/masculinity is implicitly racialized white. As "abstract" analytical categories, whiteness and maleness are already co-extensive or mutually implicated.Therefore, insofar as the "intersection" of whiteness and maleness is always already assumed unless otherwise specified, as an analytical model, intersectionality contributes nothing novel to our conception of the "white man" -except, ironically, further confirmation of the "unified" character of that identity. (see Appendix A, Figure 4). This redundancy of the intersectional analysis of "unified" identities indicates two things: first, it undermines the claim that all subjects have intersectional identities -it would seem that only hyper-oppressed subjects do, in any existentially and politically meaningful sense. Second, it illuminates the role of political intersectionality in constituting the intersectional identity of the hyper-oppressed.
This last claim calls for some elaboration. I have argued that the intersectional model of identity essentially inherits the ontology that its predecessors, the unitary conception of "woman" and the "additive" model of identity, were criticized for assuming. "Structural intersectionality," then, fails in addressing, and can even be said to reinscribe on the level of identity, "political intersectionality." The intersectional model of identity reifies "political intersectionality" by discursively producing a political subject whose stable -if contested -identity is the sedimentation of this political phenomenon: namely, the failure of existing discourses to represent (in the descriptive and the normative senses) the political experience and interests of racialized women. Insofar as they are insufficiently represented by both, racialized women report standing in a dilemmatic relation to feminist and antiracist discourses. The trope of "structural intersectionality" compels an internalization of that political dilemma, producing an internally divided subject. To say that racialized women "occupy a social location that resists telling" is to locate the cause of unrepresentability in the subject of politics, in this case, in racialized women. It is to fail to recognize that race, gender, and class are not the identic properties of individuals or of groups, but rather, are political relations which structure the lived experience of the subjects they interpellate. The notion that the problem that Crenshaw calls "political intersectionality" can be resolved at the level of identity presupposes that the dilemma issues from the identity of the racialized woman. But this is, essentially, the assumption of a falsely universalizing unitary model of "women." The intersectional model reproduces it, insofar as it consigns hyper-oppressed subjects to an intersection of axes of oppression, which it conceives in monistic terms that qualitatively privilege the oppression faced by subjects who are oppressed on a single axis. But if, as I want to claim, structural intersectionality is a function of political intersectionality, there is no sense in which individuals "are" intersectional subjects prior to a political discourse that assigns them to that location. 40 Despite the political hope with which it is invested, intersectionality contributes to -and does not remedy -the discursive unrepresentability of hyper-oppressed subjects. It is not clear that "representation" ought to be the aim of transformative politics. At the same time, it is not clear that the task of representation can be disowned. 41 I have suggested that political identity does not precede, but is produced in and through political representation. If identities are not prior to politics, but are themselves the products of political relations, perhaps the task is not to re-describe or better represent identities, but instead, it is to unearth conceptually and transform practically those relations which produce them. With Collins, I am critical of the tendency to understand resistance to oppression as occurring only in the area of representation, as if thinking about resistance and analyzing representations can substitute for active resistance against institutional power. Quite simply, difference is less a problem for me than racism, class exploitation, and gender oppression. Conceptualizing these systems of oppression as difference obfuscates the power relations and material inequalities that constitute oppression. 42 What is needed, then, is an oppositional politics desiring to reveal, criticize, and actively transform structural relations of power.

A POLITICS OF SOLIDARITY
This critique of the intersectional model of identity does not imply that we ought to end all "identity talk" in all circumstances -as if that were possible, under prevailing conditions of domination. 43 But it is becoming increasingly apparent that the project of theorizing identity -or, for that matter, "difference" -is not, in itself, a sufficient condition for feminist solidarity; and, increasingly, I am led to question whether it is a necessary condition. Himani Bannerji distinguishes the project of theorizing identity from the project of theorizing subjectivity. Bannerji argues that the latter ought to replace the former as the ground of transformative criticism of structures of oppression: The social analysis we need, therefore, must begin from subjectivity, which asserts dynamic, contradictory, and unresolved dimensions of experience and consequently does not reify itself into a fixed psychological category called identity which rigidifies an individual's relationship with her social environment and history. 44 I have argued that the intersectional model of identity performs just such a reification of the structural relations of power in which subjects are implicated. A productive alternative is theorizing the subjective conditions necessary for enacting a feminist politics of solidarity. Such a politics has (at least) three crucial elements: (1) it performs a structural analysis of the ways in which systems of oppression "interlock" and of the ways in which subjects are located in and reproduce these systems. It (2) involves an actional commitment to transforming the structural relations that subtend these systems. Finally, (3) a feminist politics of solidarity distinguishes between being positioned or situated in relations of oppression and privilege -an ineluctable fact of life under prevailing conditions -and positioning or situating oneself in relations of solidarity with "communities in struggle." 45 In this closing section I want to gesture at what might be at stake in each of these moments of a feminist politics of solidarity, which embodies what I take to be the spirit of intersectional theories like Collins' but which transcends the conceptual and praxiological limitations of the intersectional model of identity as a foundation for feminist politics. I earlier distinguished "intersectional" accounts of oppression from analyses which theorize the ways in which systems of oppression "interlock." The latter, understood as the methodological point of departure for a transnational feminism, involves the analysis of "flows of capital, labour, cultural and knowledge production between nations and regions [… and] what is happening to the specific bodies caught in the drift" -including our own. 46 Such an analysis does not fetishize "difference"; instead, it attends to globalized relations of domination which produce social contradictions. Further, a transnational feminist analysis of interlocking systems of oppression is only possible "when there ceases to be an imperial demand for exotic voices of difference." 47 Only when we relinquish an imperialist fantasy of "difference" are we in a position to "[pay] attention to specificity": Imperialism demands that we understand women either as victims or agents, as saviours or as saved, but not as complicated subjects acting within several hegemonic systems. Our task is therefore to materialize women from the South, or racialized women in the North … and ourselves, as real women. 48 To "materialize" oneself in feminist theory, one must actionally confront one's implication -indeed, one's complicity -in structures of domination. This task -a practical interrogation of one's privilege -differs in kind from the insular "academic" project of theorizing identity (and difference). Indeed, it is not a speculative exercise: it is, instead, a practical one. And, crucially, it restores a concern about the contours of power to feminist theory, without which any feminist politics to which theory might give rise seems hardly worthy of the name -for "[w]hat type of oppositional politics emerge[s] from a focus on difference devoid of power?" 49 Fiona Probyn writes that feminists interested in enacting a politics of solidarity need to reconceptualize their own complicity in structures of domination as the "starting point and the condition of ethics itself." 50 Too often, recognition that one is complicit in the oppression of others (as well as in one's own) leads subjects with relative privilege to succumbing to cynical paralysis. But such a response -enabled by privilege -only serves to reinscribe privilege. Taking one's own complicity in interlocking systems of oppression as the starting point of the feminist enterprise requires an actional commitment -to put it bluntly, work -to transform the political relations which constitute these systems. For those of us who are of/in the global North, part of this work consists in confronting the implications of the fact that the condition of possibility of our own "liberation" -historically limited though it may be -is the continuing subjugation of the global South, in post/neocolonies, on reserves, in prisons, export processing zones, inside and outside the borders of white settler societies. 51 A politics of solidarity moves from an analysis of systems of oppression to action with the urgency that the global situation demands. If our descriptive location does not determine our normative commitments, we are all -relatively privileged and hyperoppressed subjects alike, though to different extents and in different ways -in "voluntary exile" in the space of political resistance. We might fruitfully compare this movement from analysis to action with Chandra Talpade Mohanty's distinction between, on the one hand, being positioned or situated in systems of oppression and, on the other, positioning or situating oneself in feminist solidarity with others whose social location differs from one's own. 52 The former is a passive condition; the latter is active. 53 Mohanty distinguishes between these two conditions as being-of/in and being-for: I am of the Two-Thirds World in the One-Third World. I am clearly a part of the social minority now, with all its privileges; however, my political choices, struggles, and vision for change place me alongside the Two-Thirds World. Thus, I am for the Two-Thirds World but with the privileges of the One-Third World. I speak as a person situated in the One-Thirds [sic] World, but from the space and vision of, and in solidarity with, communities in struggle in the Two-Thirds World. 54 What is ultimately at stake in this distinction is opening up a space in which to theorize and actionally engage the historical conditions of possibility of transformative agency. For Audre Lorde, the movement between these two senses of "location" -being of/in and being for -is made possible through the "creative use" of privilege. 55 What is politically crucial, then, is the emergence of transformative subjectivity in the practice of the movement between being-of/in and being-for. Indeed, transformative subjectivity is to be located precisely in that movement. This work, I want to suggest, represents the point of departure, and not the conclusive arrival of feminist politics; it is not derivative, but rather productive of feminist subjectivity. 2 Chandra Talpade Mohanty makes a useful distinction between the "false universalizing" of Eurocentric humanism (in its putatively feminist and antifeminist incarnations) and the truly "universal" project of a transnational feminism that takes "the local as specifying the universal" and seeks to establish a "decolonized, cross-border feminist community," in which "'common differences' can form the basis of a deep solidarity" (Mohanty, 2003, p 224-225). I find Mohanty's construal of the relation between the particular and the universal compelling, and it informs much of my analysis here.

31
3 Cooper in Guy-Sheftall, 1995, p 44. 4 Beale, 1970. 5 King, 1988. 6 Lugones, 2003. 7 Razack, 1998. 8 Meyers, 2000, emphasis added. 9 I deploy this received distinction between "identitarian" and "post-identitarian" feminisms provisionally, to make a point that I return to in §3, namely, that the critique of intersectional models of identity as I present it here does not imply that we should end all "identity talk" -as if this were possible. That is because even absent an "identitarian" politics (i.e., a politics that understands its fundamental demands for emancipation to issue from, and to represent the identity of a given social group), the fundamental problem with the language of intersectionality remains; that is, an inadequate conception of the relation between relations of oppression. (I expand on this in §1.) Consequently, "postidentitarian" feminists (i.e., feminists for whom the feminist project is cruc,ially predicated on the "deconstruction" of identity categories) who use the trope of "intersectionality" to conceptualize subjectification under conditions of oppression are no less the targets of my critique. I am neither philosophically nor politically committed to the distinction between "identitarian" and "post-identitarian" politics, nor to the view that the latter necessarily represents a political or theoretical advance over the former. It is interesting, as Cressida Heyes observes, that "writing that actually uses this specific phrase ["identity politics"], with all its contemporary baggage, is limited almost exclusively to the last 15 years. Thus it was barely as intellectuals started to systematically outline and defend the philosophical underpinnings of identity politics that we simultaneously began to deconstruct them" (Heyes, 2002). Given this, careful readers of feminist intellectual history need to question the extent to which "identity politics" is a category constituted in and through its deconstructive critique. See Heyes, 2002. 10 McWhorter, 2004. I am twisting McWhorter's words a little here, to express a worry she formulates much more tactfully: "But if the claim that race, sex, and class intersect is going to serve feminist theory as something other than just a strategy to avoid charges of racism or classism, some form of concrete analysis of what has been called 'intersection' is extremely important" (McWhorter, 2004, p 39). McWhorter goes on to perform such an analysis, which thematizes this claim by arguing that race and sex have a common genealogy.
11 A terminological note: I use the term "hyper-oppressed" to refer to subjects who are oppressed by more than one of many interlocking systems of oppression. The term "multiply oppressed" seems to beg the question of how we define the relation between relations of oppression -which is the (unresolved) question that forms the background of this paper. But I use the term "hyper-oppressed" tentatively and with reservations: for one, like the more common "multiply oppressed," it connotes a quantitative conception of oppression; it also implies that being oppressed on only one axis is, somehow, being "hypo-oppressed," or being oppressed just enough -this is not my intention in using the term. That said, the term guards against what Patricia Hill Collins has called "the myth of equivalent oppressions": "In the United States [or in Canada] to be a Black woman is not the same as to be a White gay man or a working-class Latino […]. Moreover, in a situation in which far too many privileged academics feel free to claim a bit of oppression for themselves […] oppression talk obscures actual unjust power relations. Within these politics, some groups benefit more from an assumed equivalency of oppressions than others" (Collins, 2003, p 212).
12 In a rare critique of intersectionality, Martha Gimenez argues precisely the opposite: that this model of identity offers no resources to "link intersectionality to its macro level conditions of possibility, those "interlocking" structures of oppression […] It is here that [it] runs into a theoretical dead end which the abundance of metaphors … can neither hide nor overcome [… T]he formal nature of the RGC [race, gender, class] perspective becomes clear: race, gender and class have become, for all practical purposes, taken for granted categories of analysis whose meaning apparently remains invariant in all theoretical frameworks and contexts […] no specific theory is invoked to define how the terms race, gender, and class are used, or to identify how they are related to the rest of the social system [… A]ll that remains to be done is empirically to document their intersections everywhere" (Gimenez, 2001, p 29).
13 A fourth sense which emerges from Crenshaw's analysis of cultural politics is "representational intersectionality," which is about the way in which "the production of images of women of color and the contestations over those images tend to ignore the intersectional interests of women of color" (Crenshaw, 1991(Crenshaw, , p 1283. Crenshaw illustrates this with the prosecution of 2 Live Crew for obscenity (1990) and the public discourse surrounding this event (Crenshaw, 1991(Crenshaw, , p 1282(Crenshaw, -1295. I do not discuss representational intersectionality in the body of the paper, as it seems to me to be just the convergence of structural intersectionality and political intersectionality at the level of (aesthetic or cultural) representation.
14 Collins, 2003, p 207 15 Collins, 2003 16 For example, in the Preface to Gender Trouble, that now-classic statement in post-identitarian feminist theory, Judith Butler deploys the language of intersectionality to argue for the inseparability of gender from other forces in subject-constitution: "If one 'is' a woman, surely that is not all one is; the term fails to be exhaustive, not because a pregendered 'person' transcends the specific paraphernalia of gender, but because gender intersects with racial, class, ethnic, sexual, and regional modalities of discursively constituted identities. As a result, it becomes impossible to separate out 'gender' from the political and cultural intersections in which it is invariably produced and maintained" (Butler, 1990, p 3, emphasis added).
17 McCall, 2005, p 177318 McCall, 2005, p 1786. According to Patricia Hill Collins, "interlocking" and "intersectional" refer, respectively, to macro-level and micro-level phenomena: "[t]he notion of interlocking refers to the macro level connections linking systems of oppression such as race, class and gender […]The notion of intersectionality describes microlevel processes -namely, how each individual and group occupies a social position within interlocking structures of oppression described by the metaphor of intersectionality" (Collins, 1997, p 74).However, in a later text (Collins, 2003), Collins uses the term "intersectionality" to refer to micro-meso-and macro-level processes.
19 Razack, 1998, p 13, emphasis added. Elsewhere, Razack goes into a bit more detail about the concrete ways in which "women are produced into positions that exist symbiotically but hierarchically": "The debt crisis and its antecedent in colonialism have left Caribbean women with no choice but to seek work in the countries of the bankers. North American women negotiate their own reproductive futures in a context of inadequate childcare options or flex-time […] When middle-class women as employers of Black domestic workers discipline themselves to ignore the fact that their domestic workers are also mothers, when they internalize middle-class norms of childrearing and gender roles for professional women that requires [such] economic and social arrangements […] and when racial discourses handily sustain the idea that a Black domestic worker is not the same kind of mother, we see the many levels at which hegemonic systems make subjects" (Razack, 2000, p 51). McWhorter, 2004, p 38n1 21 Crenshaw, 1989. That the intersectional conception of oppression, implied in the intersectional model of identity, was first introduced to mitigate the inadequacy of abstract legal categories is not philosophically inconsequential. I submit that intersectionality has a different tenor as a purposive intervention into legal theory, than it does as a stand-alone theory of oppression or of political subjectivity. This is because in the former context, the intersectional model of identity (and the conception of the relation between oppressions that implies) is engaging an abstract, liberal democratic discourse of "difference" and "inequality" more or less on its own terms. But clearly this discourse lacks a substantive conception of oppression (which is why critical interventions like Crenshaw's are so important). And, inasmuch as the law is a justificatory discourse of an oppressive social structure, this is no accident (which is why such interventions, however well-intentioned and incisive, arguably have, in the last instance, limited effects). I am led to wonder whether when intersectionality is imported without modification (and without comment on its origins) into feminist theory, as a panacea to "problems of exclusion in feminist thought" -to use Elizabeth Spelman's (1988) phraseit does not, in fact, inevitably bring with it this theoretical baggage -whatever the intentions of the theorist who attempts such a translation may be.
22 Crenshaw, 1991Crenshaw, , p 1245 23 This last claim points to the need for yet another disambiguation of the use of the term "intersectionality": in the literature, "intersectionality" is used to refer both to the phenomenon of convergence and mutual construction of relations of oppression in "the real world" and to the analytic method or social-scientific methodology we might use to identify and understand that "real world" phenomenon. Not only is it often unclear which sense is operative, but this double-usage sometimes leads to the formulation of apparently tautological statements, e.g., "I suggest, however, that intersectionality has introduced new methodological problems and, partly as an unintended consequence, has limited the range of methodological approaches used to study intersectionality" (McCall, 2005(McCall, , p 1772. 24 Crenshaw, 1991, p 125225 Crenshaw, 1991, p 125226 Crenshaw, 1991, p 1242. In the passage from which this phrase is taken, Crenshaw comes exceptionally close to acknowledging the causal priority of politics to identity: "Feminist efforts to politicize experiences of women and antiracist efforts to politicize experiences of people of color have frequently proceeded as though the issues and experiences they each detail occur on mutually exclusive terrains. Although racism and sexism readily intersect in the lives of real people, they seldom do in feminist and antiracist practices. And so, when the practices expound identity as woman or person of color as an either/or proposition, they relegate the identity of women of color to a location that resists telling" (Crenshaw, 1991(Crenshaw, , p 1242. But notice how her concluding appeal to "structural intersectionality" effectively reproduces the very logic she is criticizing (i.e., that racism and sexism are construed by feminist and antiracist politics as discrete phenomena).
27 Historically, this resistance has had two "prongs." The other "prong" of resistance, which I do not discuss in the body of the paper, is to racialized women's marginalization or exclusion from antiracist politics. For example, as Crenshaw and others have observed, Black feminists engaged in the civil rights and black power movements in the United States saw the continual marginalization of women in those movements. Feminist politics has been cast variously as an imported (or, "white") discourse, destructive of racial solidarity, or as concerned with secondary, derivative social phenomena, which racial emancipation would eradicate. For example, "efforts to stem the politicization of domestic violence" in racialized communities "are often grounded in attempts to maintain the integrity of [those] communit[ies]" (Crenshaw, 1991(Crenshaw, , p 1253.Alternately, "gender domination within the [Black] community is [sometimes] reconfigured as a consequence of racial discrimination against men" (Crenshaw, 1991(Crenshaw, , p 1257. Oyèronké Oyewùmí writes that "the rhetoric of the loss of manhood […] is articulated as the main theme of Black history, with the restoration of manhood being the goal of anti-racial [sic] political movements and even some mainstream social policy. The impact of slavery, joblessness, colonization, and white supremacy in all its manifestations is constantly interpreted by reference to its effects on this nebulous thing called manhood. It follows, therefore, that if the goal of liberation for the race is the restoration of manhood, who is better placed to fight this fight than the vessel from which manhood has been taken -the Black man" (Oyewùmí, 2001, p 63). When gender is foregrounded in the masculinist antiracist discourse to which Oyewùmí refers, it is to decry the emasculation of Black men under conditions of systemic racism. In Crenshaw's view, in such discourses the oppression of Black women is often rendered as, at best, an epiphenomenal manifestation of Black men's oppression; or, at worst, it is constructed as an affectation introduced by the white feminist movement that threatens solidarity between Black men and women around the fact of racial oppression.
28 Razack, 1998, p 166. 29 Collins, 1995, p 493. 30 Meyers, 2000, p 154. 31 Meyers, 2000. Here Meyers is paraphrasing Crenshaw, who first uses the metaphor: "Consider an analogy to traffic in an intersection, coming and going in all four directions.Discrimination, like traffic through an intersection, may flow in one direction, and it may flow in another. If an accident happens in an intersection, it can be caused by cars traveling from any number of directions and, sometimes, from all of them. Similarly, if a Black woman is harmed because she is in an intersection, her injury could result from sex discrimination or race discrimination […] But it is not always easy to reconstruct an accident: Sometimes the skid marks and the injuries simply indicate that they occurred simultaneously, frustrating efforts to determine which driver caused the harm" (Crenshaw, 1989, p 149). 32 Meyers, 2000, p 155. 33 Dorthe Staunaes (2003 argues that the "concept [of intersectionality] does not include a consideration of how these categories [gender, race, class, etc.] work and intersect in the lived experience of concrete subjects" (Staunaes, 2003, p 101). But she thinks intersectionality can be redeemed by relating it to "post-structuralist and social constructionist concepts of 'subjectivity,' 'subjectification,' 'subject position' and 'troublesome subject position'" (Staunaes, 2003, p 103).Elizabeth Butterfield (2003) is similarly optimistic about the political possibilities that intersectionality opens up, but she argues that "a new understanding of oppression will not be enough -we also need to formulate a new conception of the person" (Butterfield, 2003, p 1) if we are to understand how the intersectionality of oppressions functions in the lived experience of concrete subjects.
34 Harris, 1990, p 588-589. 35 Harris, 1990, p 589. 36 Harris, 1990, p 589. 37 Meyers, 2000, p 157. 38 Collins, 1998. It would seem that the logic of these "current assumptions" has pervaded Collins' own formulation of them here; presumably the "African Americans" to whom she refers are male, to be distinguished (but only implicitly) from "Black women." On this question, and the relation of operative western concepts of difference and Otherness to the "somatocentricity" of western society, see Oyewùmí, 2001.
39 I do not mean to suggest that whiteness (or maleness/masculinity) are phenomenally invisible; they certainly are not, especially to nonwhite (and nonmale) subjects who report being "terrorized" by hegemonic whiteness. See bell hooks' essay, "Representations of Whiteness in the Black Imagination" (hooks, 1992). The hypervisibility of whiteness is an enduring theme of the literary and philosophical production of racialized and minoritized peoples, which betrays vestiges of white solipsism in the often decontextualized assertion that whiteness is "invisible" -as opposed to appearing invisible to the white gaze. As Sara Ahmed writes, "[i]t has become commonplace for whiteness to be represented as invisible, as the unseen or the unmarked, as a noncolour, the absent presence or hidden referent, against which all other colours are measured as forms of deviance […] But of course whiteness is only invisible for those who inhabit it. For those who don't, it is hard not to see whiteness; it even seems everywhere" (Ahmed, 2004, §1).
40 Elizabeth Spelman observes that "[i]nsofar as she is oppressed by racism in a sexist context and sexism in a racist context, the Black woman's struggle cannot be compartmentalized into two struggles -one as a Black and one as a woman. Indeed, it is difficult to imagine why a Black woman would think of her struggles in this way except in the face of demands by white women or by Black men that she do so" (Spelman, 1988, p 124). Yet Spelman's own formulation "racism in a sexist context / sexism in a racist context" seems to reinscribe precisely this demand. 41 Spivak, 1988. 42 Collins, 1995  43 Razack warns against such a move: "[a]long with bell hooks, I am suspicious of those who warn of the dangers of identity politics, race essentialism, or ethnic particularism without paying attention to the specific relations of domination and subordination in any one context, and without contextualizing the responses subordinate groups make to domination, thus distinguishing acts of resistance from acts of domination. As hooks suggests in her assessment of Diana Fuss' exploration of the misuses of essentialism by minority students in the classroom, critiques of identity politics may be 'the new, chic way to silence students from marginal groups" (Razack, 1998, p 169). 44 Bannerji, 1995, p 88. While I am sympathetic to Bannerji's distinction, I still want to keep the above worry in view: we need to remain vigilant about the way in which in classrooms and in political organizing the prescription to "go beyond identity" when articulated by race/class-privileged feminists reinscribes that privilege at the same time as it renders it -and complicity in systems of oppression -(self-)invisible.
45 Mohanty, 2003, p 228. 46 Razack, 2000, p 44. 47 Razack, 2000, p 44. 48 Razack, 2000, emphasis in original. 49 Collins, 1995, p 494 50 Probyn, 2004. Similarly, Razack suggests that an "important point of entry [into a transnational feminist politics] is to examine hierarchical relations among women" (Razack, 2000, p 50); "interrogating hierarchical relations should at least make clear how our privileges are intertwined with our penalties, and how both are structured on the backs of other women" (Razack, 2000, p 52). I am wary of the metaphor ofthe "intertwining" of privilege and penalty, since I take it that Razack's point is that the two are inextricable (whereas "intertwining" conjures some of the same conceptual problems that I have identified with the language of intersectionality).
51 "A white settler society is one established by Europeans on non-European soil. Its origins lie in the dispossession and near extermination of Indigenous populations by conquering Europeans.
As it evolves, a white settler society continues to be structured by a racial hierarchy. In the national mythologies of such societies, it is believed that white people came first and that it is they who principally developed the land; Aboriginal peoples are presumed to be mostly dead or assimilated. European settlers thus become the original inhabitants and the group most entitled to the fruits of citizenship. A quintessential feature of white settler mythologies is, therefore, the disavowal of conquest, genocide, slavery, and the exploitation of the labour of peoples of colour" (Razack, 2002, p 1-2).
52 Aimee Carrillo Rowe has recently suggested that we effect a move from thinking of transnational feminism as a "politics of location" to conceptualizing it as a "politics of relation": she recommends "deep reflection about the selves we are creating as a function of where we place our bodies, and with whom we build our affective [and political] ties" (Carrillo Rowe, 2005, p 16