Options
Article

Oppression, Domination, and the Structure of Graded Inequality

Author
  • Yarran Hominh orcid logo (Bard College)

Abstract

What structure do paradigm cases of oppression and domination, like racism and capitalism, have? Most theories of oppression and domination take them to have a binary structure. There are the oppressors and the oppressed, the dominators and the dominated. I argue that a better model for many paradigm cases of oppression and domination is a structure of graded inequality. Such a structure comprises multiple groups arranged in hierarchically ascending and descending order. A model of graded inequality has both descriptive and explanatory advantages over a binary model. It better illuminates some mechanisms by which oppression and domination stabilize themselves. Understanding those mechanisms allows us better to destabilize them. That is so even though good reasons may hold for adopting the binary (or some other) model for other purposes. The larger methodological upshot of the paper is to connect more closely social philosophy to moral psychology.

Keywords: oppression, domination, graded inequality, B. R. Ambedkar, stability

How to Cite:

Hominh, Y., (2024) “Oppression, Domination, and the Structure of Graded Inequality”, Philosophers' Imprint 24: 15, 1-18. doi: https://doi.org/10.3998/phimp.3192

Downloads:
Download PDF
View PDF
Download XML

166 Views

32 Downloads

Published on
2024-12-26

Peer Reviewed

License

Creative Commons Attribution-NonCommercial-NoDerivs 4.0

It is widely accepted that paradigm forms of oppression and domination — like those of racism, capitalism, and patriarchy — are structural. But what structure do they have? Many theories of oppression and domination are based on a binary, group-based structure. Every oppressed or dominated group has a correlative oppressor or dominator group. I will argue that a better model for the structure of systems of group-based oppression and domination is one of graded inequality. Structures of graded inequality comprise multiple groups arranged in a hierarchically ascending and descending order. It is a better model insofar as it helps us identify how structures of oppression and domination stabilize themselves. I take this notion of graded inequality from B. R. Ambedkar, who uses it to describe the Indian caste system.

Modelling systems of oppression and domination with the structure of graded inequality helps us understand the operation of these systems. In this paper, I will focus on how it illuminates mechanisms by which systems of oppression and domination stabilize themselves in ways that a binary model does not. First, it shows how even people disadvantaged overall by systems of oppression and domination can nonetheless have some group interest in maintaining it. Second, it reveals the stabilizing role of what I will call affective misdirection — the redirection within the system of affective energies that could be otherwise devoted toward undermining the whole system. These are illustrations of the more general point that social structures shape the moral psychology of those agents who live, think, feel, and act within those structures in ways that stabilize those social structures. So as a general methodological point, understanding structures of oppression and domination involves understanding the moral psychology of the people within those structures. Social philosophy should be connected to moral psychology.

This paper will proceed as follows. In section 1, I will outline the role that group relations play in structuralist views of oppression and domination. In section 2, I will set out examples of binary models from the literature. Such models, I will argue, contain some truth. Yet it should be immediately apparent, from a glance at various existing and historical systems of oppression and domination — such as racism, capitalism, and colonialism — that such systems are not simply binary. So in section 3, drawing on B. R. Ambedkar’s analysis of caste in India, I describe graded inequality and argue that such systems of oppression and domination are better understood by appeal to that structure. I do not imply that is all those systems are, nor that they are forms of caste. Rather, I use the abstract structure of caste as a model to understand better systems of oppression and domination. In section 4, I will draw out some of the consequences of this view of the structure of oppression and domination for understanding the stability of such systems. Section 5 will respond to three concerns about graded inequality.

1. Structuralism and Group Relations

What is meant by the claim that oppression and domination are structural? The claim has several layered meanings. At its most general, the structuralist claim is negative: a claim about what oppression and domination are not. They are not, according to structuralists, merely individual-level phenomena. They are not created or maintained just by intentional individual action, nor do they comprise just individual level attitudes. ‘Structure’ in this negative sense is whatever is “beyond-the-individual”.1 This category of the beyond-the-individual includes norms, laws, institutions, shared conceptual schemes, and culture.

Yet the ordinary meaning of ‘structure’ implies more than an ontological grab-bag of beyond-the-individual things. Those things must fit together to form a structure, a complex and systematic whole.2 One common way to fit them together is to see structure in terms of a network of interrelated roles or positions, the properties of which are largely independent of the particular individuals who fill those roles.3 A structure is defined in terms of these social positions and the relations between them. Those positions, in virtue of their relations to other positions, bear normative properties, including duties, prerogatives, goals, and so on.

For example, teachers are necessarily in relation with students. That relation gives rise to normative properties: the responsibility to impart correct information, the power to grade, the negative duty not to engage in certain relationships with students, the positive duty to be a good mentor, etc. These normative properties are tied together by role-based purposes, e.g., “civic education”. To be a teacher or a student is also to be related to a variety of other social positions, including parent, principal, etc. The network of these normatively laden social positions forms the structure of a school.

This picture of social structure ties together the collection of beyond-the-individual things. An institution comprises a network of interrelated social positions to which are attached certain norms. While some norms attach to specific positions, others attach to the structure as a whole. The school’s governing norms may be a school motto or a commitment to a particular vision of education. The structure as a whole may also be characterized by other shared epistemic resources — methods of teaching, conceptual schemes, and ways of categorizing material things (this counts as a textbook; an education comprises this set of subjects) that together form a culture. Institutions are layered in systematic normative relations. A law is a norm produced by some formalized political institutions (legislatures, judiciaries), and other institutions like schools are governed by some of those laws.

That is a broad picture of what is meant by ‘structure’. My interest is in something more specific. Theorists of structural oppression and domination often claim that paradigm forms of oppression and domination are group-based. That is another way in which these paradigm forms of oppression and domination are not individual-level phenomena. Individuals are oppressed or dominated qua their membership in a relevant social group. Each relevant social group defines a social position in the system. Members of oppressed or dominated social groups are oppressed or dominated in virtue of the relations that hold between their social group and other social groups. On these views, oppression and domination exist in virtue of the relations that hold between social groups, and only secondarily as a result of the intentional actions taken by particular individuals who occupy the positions defined by those groups.4

Marilyn Frye writes that “[i]f an individual is oppressed, it is in virtue of being a member of a group or category of people that is systematically reduced, molded, immobilized”.5 Ann Cudd defines oppression as “a harm through which groups of persons are systematically and unfairly or unjustly constrained, burdened, or reduced… [oppression] is perpetrated through social institutions, practices and norms on social groups by social groups”.6 Charles Mills and Carole Pateman use the device of a “domination contract” to model the “collusion among themselves of a social group with far greater influence, who have their own self-seeking agenda… [namely, to] subordinate women and people of color under the banner of a supposedly consensual contract”.7 Similar claims could be listed: blacks are dominated by whites, patriarchy oppresses women on behalf of men, the capitalist class dominates the proletariat, and the Hebrews were oppressed by the Egyptians.

We are now in a position to briefly characterize oppression and domination as group-based structural phenomena. With respect to oppression, I follow Marilyn Frye and Ann Cudd: oppression is a harm in which groups of people are systematically and unjustly constrained, burdened, and reduced, and in which other groups receive corresponding benefits.8 Domination, at least at the group level, is a harm in which some groups are subordinated to others such that the latter can exercise arbitrary power over the former against their interests.9 These characterizations are for the purposes of argument, in the sense that (within the constraints of the argument above regarding structure and group) it is inessential exactly what ‘oppression’ and ‘domination’ mean. Perhaps domination need not involve wholly arbitrary power; perhaps oppression has five faces or four. My focus is on the idea of group-based structure and not on the definition of oppression or domination.

The paradigm contemporary cases of oppression and domination — racism, patriarchy, ableism, capitalism — are generally understood as structural in this group-based sense. (I speak of racism, patriarchy, etc., and not race, gender, etc., to emphasize that I am focusing on the systems of oppression and domination and not primarily on the identities that are formed by and against those systems.) This group-based sense of structure holds also of paradigm historical cases of oppression and domination, even those central to individualist theories of those phenomena. Take slavery — the historical example par excellence of domination for neo-republicans like Philip Pettit. Neo-republicans treat domination as a relation between individuals, consisting in one party’s unconstrained power over another, their capacity for intentional arbitrary interference with the person dominated.

Yet as Orlando Patterson notes in Slavery and Social Death, “the master-slave relationship cannot be divorced from the distribution of power throughout the wider society in which both master and slave find themselves”.10 The individualized domination of master over slave is embedded in and partially constituted by a wider social order.11 That social order legitimates individual domination relations, and slaves (as a group) were subject to masters (as a group).12 Even third parties had power over slaves. Slaves in slave-owning societies were dominated (with legal sanction) by people other than their masters.13 For example, slaves were more severely punished than non-slaves for delicts committed against third parties, yet they were able to be wronged almost with impunity by third parties. In virtue of the larger system, slaves were subordinated to non-slave society as a whole and not merely individual masters.

To recap, structuralism about oppression and domination comprises three related claims. The first is negative: that for oppression and domination to be structural means that they are phenomena that are not merely individual. The second and third are positive: that oppression and domination are matters of the network of relations between social positions; more specifically, that they involve relations between groups.

2. The Binary Model of Oppression and Domination

A common model takes oppression and domination to have a binary group-based structure, insofar as they involve relations between two groups: a subordinate group and a superordinate group.14 Here are some examples.

For Marilyn Frye, oppression “is a system of interrelated barriers and forces which reduce, immobilize and mold people who belong to a certain group, and effect their subordination to another group… Logically, it presupposes that there are two distinct categories”.15 Catharine Mackinnon defines the correlate gender groups of “male” and “female” in terms of the “erotization of dominance and submission”.16 Frank Lovett defines domination as “a condition experienced by persons or groups to the extent that they are dependent on a social relationship in which some other person or group wields arbitrary power over them”.17 For a neat general statement, see Gary Okihiro: “The social formation of race, gender, sexuality, class, and nation segregates by inventing and policing discrete binaries and hierarchies of white and non-white, man and woman, straight and queer, capitalist and worker, and citizen and alien”.18

Binary models are common. Yet it should be obvious upon a moment’s reflection that the paradigm cases of systematic oppression and domination are not binary. The system of racial oppression and domination in the contemporary U.S. comprises not just whites and blacks, but a variety of other racialized groups. Patriarchy oppresses not only cis women, but trans, non-binary, genderqueer, gender-non-conforming, and other queer groups, including gay cis men. Colonialism involved domination not only of the colonized by the colonizers, but drew distinctions between different colonized groups, including local comprador elites and peasants. And contemporary capitalism as a system of oppression and domination involves not simply workers and the owners of the means of production, but the petty bourgeoisie, skilled as well as non-skilled laborers, franchisors, and franchisees.

These paradigm systems of oppression and domination thus have structures with multiple group positions and complex relations between those groups, not just between the groups at the top and bottom. And, as theorists of intersectionality have pointed out, individuals are positioned in multiple systems and their experience is shaped by the interaction of those systems. I will say more about the significance of this multiplicity and how to understand it in the next section, and I will return to intersectionality in the final section. But it is worth saying something first about the prevalence of the binary model and the reasons it is commonly used in the face of this multiplicity.

First, the binary model contains some truth. It captures the way in which group identity often forms, or at least coalesces, because of relations of oppression or domination rather than being antecedent to them. This is true both of oppressors and the oppressed. Both gain a sense of group identity through that relation. That is part of what Hegel recognized with his analysis of the master-slave dialectic. Both the master and the slave recognize themselves as such in contrast to the other.19 Orlando Patterson also makes this point when he claims that the significance of being part of a group defined by the possession of freedom was an epistemic consequence of the unfreedom embodied in slavery. Those who were “not-slave” recognized themselves as “free” only in contrast to the slave. It is a point also recognized sociologically in terms of in-group/out-group dynamics, in theories of racial formation, and in social constructionist accounts of gender. Groups form because of those processes by which one set of people come to oppress or dominate another. Those processes generate a world that can be understood along binary lines.

For example, processes of racialization in the United States cast Latinxs, Asian Americans, and others as either “near-black” or “near-white”. Chinese Americans were defined as “black” in the case of The People v. George W. Hall 4 Cal. 399 (1854). Hall, who was white, was convicted of murder on the testimony of Chinese witnesses. Blacks were not permitted to give evidence in court. Hall’s lawyers argued successfully that ‘black’ meant “nonwhite”, on the basis that the purpose of the exclusion was to protect (white) men “from the testimony of the degraded and demoralized caste” — everyone who was not white.20 Yet Asian Americans are now in many cases considered “near-white”. The “model minority” myth casts Asians as possessing the “white” capacity for hard work (contrasting them with the stereotype of “lazy blacks”), allowing them to assimilate into white American society. Linda Martín Alcoff notes that Latinx people are now increasingly being racialized as “white” in order to maintain a white majority.21 At the same time, darker-skinned Asians, people of Arab and Middle Eastern background, and darker-skinned Latinx people are racially profiled at levels that mean they can be classified as “near-black”. The constructive effect of the black/white racial binary makes it to some extent descriptively accurate.

Second, focusing on particular binary relations of oppression and bracketing other features of the system can lead to a greater specificity of analysis, though at the expense of comprehensiveness. One can zoom in on certain features of that particular oppressive relation without being encumbered by the complexity of introducing other groups. The binary model, like all models and theories, directs our attention in certain ways that can serve particular goals. But we ought to be aware when we do so, and of what we might miss by doing so.

Third, there may be good political reasons to posit a binary model. It can be politically powerful to single out one group as the target of political activity and to figure the rest as resisting them together in solidarity. For example, take the contemporary phrase “the 1%”. It picks out a political target, one which has motivational and rhetorical power. Casting a complex reality into stark relief can clarify the political stakes and act as a kind of consciousness-raising.22

While the binary model has its purposes, I will argue that there are descriptive and conceptual benefits to the graded inequality model that the binary model does not have. And this suggests at the least that we could utilize the graded inequality model in conjunction with the binary model — after all, as model pluralists argue, even inconsistent models need not exclude each other in practical use.23

3. Ambedkar, Caste, and the Structure of Graded Inequality

I have pointed to several paradigm cases of oppression and domination that are obviously not binary. In this section, I draw on B. R. Ambedkar’s theory of caste to propose that what Ambedkar calls graded inequality is a model with more explanatory power. I do not imply that these paradigm cases of oppression and domination are forms of caste (though race, class, and citizenship are often described as forms of caste), nor that they are fully explicable through the model of graded inequality.24 Graded inequality is also just a model — and a model of the structural features of oppression and domination, which are not the whole of those phenomena. And, like any model, it abstracts away from complexities in the underlying phenomena. It is sufficient for my purposes that graded inequality illustrates — in ways the binary model does not — important features of these systems, particularly the means by which systems of oppression and domination stabilize themselves. I turn to these in section 4.

Graded inequality is a structure that comprises groups in multiple strata arranged in a hierarchical order. The hierarchy is one of unequal and differential rights, privileges, and duties vis-à-vis other groups in the structure, and of closeness and distance to the norm set by the highest group.

The paradigm of graded inequality is caste. Insofar as it is a paradigm, caste (in my analysis) plays a role similar to that of an idealized model. It reveals clearly the operation of the structure in ways that may be more difficult to see in other systems. I will illustrate the structure of graded inequality through the case of caste in India.

A caste system is one divided into distinct, closed, and hierarchically structured groups. The groups are distinct, insofar as any one person belongs only to one caste from birth, and their social role is (largely) defined by their belonging to that caste. The groups are closed, insofar as a person cannot move between castes, the castes are endogamous, and certain interpersonal relations between castes are regulated or prohibited. Distinctness and closure are matters of degree. A caste system is more closed the higher the restrictions on movement and interaction between castes and the higher the endogamy rate. And lower closure leads to less distinctness. Castes are hierarchically structured, insofar as castes are higher or lower than others.25 They are defined relationally and in contradistinction to other castes. Castes do not exist in the singular; only in the plural. For there to be a higher caste, there must be a lower.26 It is this hierarchical and relational feature of caste’s structure that I will refer to, following Ambedkar, as graded inequality. Distinctness and closure will play little role in my argument.

The word ‘caste’, in the Indian and more broadly the South Asian context, draws together two different categorizations: jati and varna. Jati refers to caste groups defined by occupation and often geographical region. These occupation-level castes are themselves grouped into a fourfold division according to the old Vedic principle of varna. This fourfold varna categorization comprises, from “highest” to “lowest”, Brahmins (priests, intellectuals), Kshatriyas (warriors, administrators), Vaishyas (merchants, agriculturalists), and Shudras (laborers). Excluded from that fourfold division are Dalits, or “untouchables”, as well as other avarna (not being part of a varna) Indigenous tribes.

The fourfold varna division as superimposed on jati divisions is what Ambedkar primarily has in mind when speaking of caste in terms of graded inequality.27 “The four classes [varnas]”, Ambedkar writes, “are not on a horizontal plane, different but equal. They are on vertical plane…unequal in status, one standing above the other”.28 The castes are unequal; each of them has differential and graded rights, privileges, and duties. This inequality is further structured, Ambedkar notes, according to Gabriel Tarde’s principle of sociological “distance”.29 The “highest” caste, the Brahmins, sets the norm for all other castes. The “higher” a caste, the more they share with the Brahmin norm, and the “closer”, sociologically, they are to the Brahmins. The inequality is given its structure top-down by the norm set by the highest caste.

Dalits, as one of the groups excluded from the caste hierarchy, suffer the most intense forms of oppression and domination. Dalits are considered “polluting”. Caste Hindus, those of the four varnas, refuse to come into contact (even indirectly, through contact of shadow or sharing of food or drink or space) with the “untouchable” Dalit.30 Anything “polluted” has to be ritually cleansed before use. Dalits are the subject of routinized violence, marginalization, humiliation, and exclusion perpetuated by caste Hindus for the purpose of maintaining caste hierarchy.31 They are denied access to education, religious and community institutions, job opportunities, and welfare. These instances of domination are enabled by the social and religious norms of caste and the legal and political system that continues to propound the caste system. It is this oppression and discrimination that leads some theorists to compare race to caste; in particular, being black to being Dalit.

Contemporary paradigm systems of oppression and domination can usefully be understood in terms of graded inequality. Let me analyze briefly two such systems: capitalism and racism in the United States. These analyses are intended solely as illustrations of the general point; hence, there may be legitimate and for other purposes important disputes as to the details (the number of grades and how they are divided, the precise order of the hierarchy) that do not materially affect my main point: that these systems of oppression and domination can be usefully understood in terms of graded inequality.

First, capitalism. Capitalism’s hierarchy of graded inequality has the large capitalist owners of the means of production at the top, with the unemployed — what Marx called the “industrial reserve army” — at the bottom. Above the industrial reserve army are employed unskilled laborers, with skilled laborers above them, and the petty bourgeoisie and the professional classes second from the top. Let me distinguish graded inequality from the binary model (capitalist/proletariat) by examining the role played within capitalism by the bottom three grades. Although I draw particularly on Marx in this analysis, my intention is to describe the underlying phenomena in ways the binary model cannot. And so I could have drawn on any number of analyses of differentiation within the working classes to make this point, for example Third Worldist accounts of the proletariat/peasantry distinction or that of labor historians who focus on “labor aristocracy”.

A “reserve army” of labor — what Marx describes as “a mass of human material always ready for exploitation” — is necessary for the reproduction and accumulation of capital, since any given increase in fixed capital requires an increase in exploitable labor-power to make that fixed capital productive.32 The process of capitalist accumulation itself creates this reserve army, insofar as centralization of capital and technological advance in the means of production render, at least temporarily, parts of the workforce unnecessary. So the industrial reserve army is necessary for the reproduction of capital, and in turn the reproduction of capital creates and maintains the reserve army. The industrial reserve army serves an additional purpose: to enable the greater exploitation of those who have work, in order that they keep the work that is necessary for their livelihood. Those who are employed have a motivation to maintain their structural advantage over the unemployed. Competition among the proletariat is a motor of oppression: “work harder or be replaced!”33

The processes of capitalist manufacture and the division of labor inherent in it, to quote Marx, “develops a hierarchy of labor-powers, to which there corresponds a scale of wages”.34 As the process of production becomes broken up into specialized and discrete parts, there arises a set of tasks that do not require any skills, “certain simple manipulations, which every man is capable of doing”.35 In contrast to processes of guild and independent craft production, in which one person had to learn and in the end undertake all stages of the production of goods, capitalist manufacture “begins to make a speciality of the absence of all development”.36 The division of labor into skilled and unskilled labor is desirable for the capitalist. It decreases the costs of labor for both groups and thus increases the surplus value derived from the exploitation of that labor. It decreases the cost of skilled labor because of the simplification of functions — the skilled laborer does not have to deal with the boring minutiae of “simple manipulations”. It decreases the cost of unskilled labor because it does away with the need for apprenticeship and training of that labor.37 Marx describes this division into skilled and unskilled as analogous to the creation of caste.38 The skilled thus work to maintain their position ahead of the unskilled, forming one version of a “labor aristocracy”.39 A labor aristocracy uses its increased bargaining power with capitalists (only “we” have the skills to perform this set of functions) to protect its structural privileges over less skilled workers.

I have made two points. The first is that capitalism contains these further gradations, and not merely the distinction between owners and workers. The second is that those further gradations are generated by capitalism and serve the operation of that system, including (among other things) the oppression and domination of workers for the exploitative generation of wealth. That general operation includes a structural relation between lower grades whereby intermediate grades gain structural advantages over those lower. Operating with a simple binary model (capitalist/proletariat), occludes our vision of how solidarity between workers is structurally discouraged.

What about racism? Racial hierarchy in the United States comprises not only the binary of white and black, but also other racialized groups (Asians, Latinxs, Native Americans, etc.). There is much debate about where these “other” racialized groups fit into the American racial hierarchy, and by what principles and by what sociological mechanisms they are placed where they are.40 That task is made far more complex by a number of factors, for example multiraciality and the fact that there are further differentiations (along lines of skin color and ethnicity) within racialized groups that, depending on the relevant measures used, may lead to different groups occupying particular gradations. And that complexity is only increased by the multiple dimensions of oppression that characterizes the structure of racism in the contemporary United States.41

It is outside of the scope of this paper to delve into the processes of racialization that support white supremacy in the United States, or the complexities of racial divisions. It will serve my limited purposes to examine a simplified model of racial graded inequality and compare it to a binary model. I will analyze Eduardo Bonilla-Silva’s notion of the post-Civil Rights era United States as a tri-racial order, which categorizes a variety of racial and ethnic groups together as a single intermediate gradation between “white” and “black” (where “black”, in particular, is not limited to African-Americans).42 I do not endorse Bonilla-Silva’s model as the best or most accurate model for theorizing racism in the U.S. But treating his model as a (simplified) graded inequality model enables me to identify some mechanisms that stabilize oppression and domination in ways that a binary model cannot.

Bonilla-Silva argues that the contemporary United States, with its increasing Asian and Latinx demographic, is better understood not as a bi-racial but as a tri-racial order. The third racial category he proposes between white and black is that of “honorary whites” — those who are not-black, but who are granted certain “white” racial privileges in order to “buffer racial conflict” and to serve various ideological functions (assimilationist liberalism, Protestant narratives of upward mobility) that maintain white supremacy. In this category, Bonilla-Silva places south Asians and east Asians, lighter-skinned Latinxs, and Middle Eastern Americans. He includes in the “black” gradation south-east Asians, darker-skinned Latinxs, and some Native Americans. He argues, following Omi and Winant, that racialization occurs in order that some maintain the benefits of oppression and domination that define a racist system.43 One mechanism by which these benefits are maintained is the selective extension of some racialized benefits to these intermediate groups. “Honorary white” groups are thus “closer” to and less distinct from whiteness than “black” groups.

Compare this to a binary model: Andrew Hacker’s Two Nations: Black and White, Separate, Hostile, Unequal.44 Hacker argues that U.S. racism effectively renders white and black America two separate and unequal nations: a clear statement of the binary model of oppression. Hacker admits explicitly that the structure of race in the U.S. involves other racialized groups, including Native Americans, Latinxs, and Asian Americans. Hacker even admits that blacks and whites “now comprise a dwindling share of the nation’s population.”45 Yet Hacker focuses exclusively on black/white race relations in the United States. He justifies this by saying that “other groups find themselves sitting as spectators, while the two prominent players try to work out how or whether they can co-exist with one another”.46 The binary model is qualified, but in a way which just reasserts the model. And so Hacker, unlike Bonilla-Silva, is forced to see the other racial groups as “spectators” and not as actors in the U.S. racial order. Bonilla-Silva’s model allows us to see how racism as a system of oppression stratifies racialized groups in order that dominant groups maintain their advantages over others through extending benefits down the hierarchy.

These claims raise the question: How exactly does a structure of graded inequality serve the purposes of maintaining oppression and domination? In the next section, I will outline two general mechanisms by which a structure of graded inequality helps stabilize systems of oppression and domination.

4. Graded Inequality and the Stabilization of Oppression and Domination

There are many consequences of graded inequality specific to particular systems of oppression and domination. For example, there are particular forms of alienation that arise from the hierarchy of labor. The specialization of labor (including of unskilled labor) means that workers are not fully human; they become the “disjecta membra” — the mere scattered fragments — of a full person. There are specific forms of stigmatization and stereotyping that attend racial categories. And in the case of caste in India, there is the distinctive phenomenon of untouchability that characterizes Dalit oppression.

I will be concerned with more general features of graded inequality that arise across systems of oppression and domination. In particular, I focus on how graded inequality stabilizes oppression and domination. Political philosophers have long been concerned with the stability of fundamental institutions that are thought to benefit society. I extend this question of stability to institutions, like racism and capitalism, that are equally fundamental to our society but that are clearly unjust.47 Understanding that stability is a step toward destabilizing them.

Drawing again on Ambedkar, I outline two related general mechanisms: group interest and affective misdirection. Graded inequality fosters group interests in maintaining the system, even in those who are disadvantaged overall by the system, to protect what limited comparative advantages they have over others. Graded inequality is a form of divide-and-conquer. Graded inequality also creates means of affective misdirection — fostering certain kinds of positive and negative affect between groups that could otherwise be directed toward overturning the system as a whole. Both of these mechanisms involve the shaping of moral psychology by the system in ways that support the continued operation of that system.

Group interest

Graded inequality fosters a moral psychology of group interest. Groups above the bottom of the graded hierarchy derive some comparative benefit from the hierarchy in being above others. A moral psychology of group interest is one in which members of the group seek to protect that comparative interest against what they perceive as the interests of other groups. It is one in which, Ambedkar says, their “prevailing purpose is protection of what it has got”.48 Since protection of this interest involves maintaining the structure of graded inequality, this moral psychology of group interest serves to stabilize the system as a whole, since a group cannot maintain those benefits unless the structure exists. This mechanism is not available on a binary model, since the subordinate group has no comparative interest in maintaining the system.

Take Ambedkar’s analysis of the lowest of the varna castes, the Shudra (laboring caste), who are above Dalits and other avarna tribes. As the lowest of the varnas, Shudras are disadvantaged, oppressed, and exploited almost like Dalits. So they should theoretically be in solidarity with the cause of annihilating caste. But, Ambedkar argues, the “atrocities that are committed upon the Untouchables, if they commit any breach of the rules and regulations of the established order…are all the doings of the Shudras”.49 The Shudras act as the “police force of the Brahmins”.50 They directly enforce the caste order upon those lower than them. The Shudra play this enforcer role, Ambedkar argues, because while the Shudra “is anxious to pull down the Brahmin, he is not prepared to see the Untouchable raised to his level”.51 They have their positional status in the caste hierarchy to lose. As I described in section 3 (and expand on below), both racism and capitalism set lower groups against each other, both in the pursuit of material interests (money, power) and positional interests (status).

This notion of a group interest is of a specific sort. It is not what we might call a true interest, insofar as groups in the middle of the hierarchy would more truly benefit from the abolition of the entire system of oppression and domination. Shudras would be far better off if caste were abolished. Yet it is not merely a subjective or perceived interest, one taken by an agent as being in their true interest. By this I mean two things. First, the interest exists whether it is grasped subjectively as such. It exists in virtue of a group possessing a particular position in the hierarchy. In this sense, we can say that it is an objective interest, inhering in the social structure. Second, the interest need not be grasped and acted on as that interest in order to exist and to stabilize systems of oppression or domination. The interest can manifest subjectively in the psychology of individuals in a variety of emotions, beliefs, desires, ways of thinking and feeling, self-conceptions and so on, not simply as a belief that “my group interest is X”. For example, as Ambedkar says of castes, groups may want to emulate those higher than them in the hierarchy, and to mark themselves off as distinct from those lower. They may take pride in their particular position in the caste order, and consolation in the fact they are not at the bottom.52 Group interest may manifest subjectively in these more affective forms and not in a propositional belief that they must “protect what they have got”.

Conceiving of group interests in this enlarged way is not meant to deny that those interests may be, and are often, grasped subjectively as such. Members of immigrant groups often hold explicitly discriminatory views about refugees and other, later, migrant groups perceived as “jumping the queue”; after all, they came in “the right way”, they have assimilated, and they don’t want those “others” taking “their” jobs. So, too, Asians and Latinxs can hold explicitly anti-black racist attitudes that can (for example) underlie disapproval of affirmative action and welfare policies.53 And parts of unionized labor have a history of resisting unionization of those lowest paid, for fear that they will lose their bargaining position with the bosses.

However this group interest is subjectively felt and acted upon, it serves to stabilize the entire system. Ambedkar, comparing a structure of graded inequality to one of “simple inequality”, argues that the former is more stable than the latter, precisely in virtue of its divide-and-conquer structure. “If the Hindu social order was based on inequality,” Ambedkar writes, “it would have been over-thrown long ago. But it is based on graded inequality … [and so] there is nobody to join the Untouchable in his struggle.”54 Graded inequality explains the stability of oppression and domination in ways that a binary model cannot.

Affective misdirection

I suggested earlier that group interests can manifest subjectively otherwise than as the usual philosophical intentional attitudes: beliefs and desires. They can manifest in various affective states, in ways of seeing and feeling, and in hidden self-conceptions. A general and related point is that structures of graded inequality enable means for misdirecting people’s affective energies. Emotions that could motivate action against the system become co-opted into stabilizing that system. Call this affective misdirection. Affective misdirection is a mechanism available to the binary model. But its workings become clearer when we move to a structure of graded inequality.

What kinds of affects are misdirected, and how? Let me give two examples: the positive affect of hope and the negative affect of anger. Both could provide important resources for resistance to oppression and domination: hope for a better world for all; anger directed against the system in the forms of protest and direct action. I will just assume these productive uses of these emotions, accepting that there is ongoing debate about both (is hope always bourgeois, always too indefinite? Is anger counterproductive, overly divisive?).55

Take hope. Hope can be misdirected within a system of oppression insofar as it becomes not hope for a better system (or no system at all, depending on one’s views), but hope for more and better benefits within the system. Call this a hope for upward mobility. This hope can take many forms. The most recognizable to us is individual upward mobility, say between classes. Protestant ideologies of hard work and thrift support hope in the idea that one can rise through the ranks. If one puts in the hard yards, if one “invests in oneself” properly, one can learn the skills, start one’s small business, have one’s big idea tech start-up bought by private equity, be “self-made”. And the possibility of individual upward mobility and associated doctrines of meritocracy are one of the key ideological planks stabilizing capitalism. If one is exploited enough, eventually one will be able to exploit others.

But there are other, group-based forms of upward mobility. Entire groups can move up within the hierarchy by adopting or being granted privileges and cultural capital associated with the higher grades. In the case of race, we can think of the (selective) incorporation of groups like the Irish and southern Europeans into the racial category of “white”.56 Or take the myth of the “model-minority”. Asians are taken to possess the “white” characteristics of hard work and intelligence, sufficient to make them “honorary whites”. Asians and other racialized groups in the middle of the American racial hierarchy draw on these tropes in the hope that they can “whiten” themselves and assimilate.

These possibilities of upward mobility rely on leaving the structure largely intact. One can only rise within a system if that system continues to exist. They localize energies to bettering one’s position, not to overturning the system as a whole. In this way, this misdirection of hope away from achieving systemic change and toward achieving local improvements serves to stabilize the system as a whole.

What goes for the positive emotion of hope goes also for the negative emotion of anger. Anger that is properly directed against the entire system of oppression or domination can come to be directed toward particular groups within that system. One motor of this mechanism is the fact that comparative status is a positional good. Insofar as anger is an apt response to loss of status or threat to status, then anger at other groups will arise in a system of comparative status where there is competition for status.

This anger often takes the form of “punching down” against groups lower in the hierarchy, those seeking to “get out of their place”. Ambedkar’s example of the Shudra acting as the police force of the Brahmins is apt here. Or take one phenomenon central to right-wing populism: anger against immigrants and other racialized groups for “taking our jobs” or for other negative economic phenomena. In reality, such job losses and economic downturns are a function of systematic macroeconomic tendencies and not of the actions of any one group.

It can also take the form of “punching up”, for example the violence enacted by blacks against Korean shopowners in Los Angeles in 1992 or in the racial conflict between blacks and Latinos in Los Angeles and the Bay Area around the same time.57 And there can be a “jostling for position” in the middle, as may be evidenced by antipathy between, for example, Asians and Latinxs.58 In this vein, Ambedkar describes caste as “an ascending scale of hatred and a descending scale of contempt.”59 The existence of a graded hierarchy directs anger and frustration at other groups in the hierarchy as a way of protecting one’s group interest. And that anger and the consequent group animosity stand in the way of the solidarity that is necessary in order to overturn the system of graded inequality as a whole. In that way the misdirection of this anger — its channeling into inter-group resentment — serves to stabilize the system. Both of these mechanisms operate through agents’ moral psychology. Social structures shape agents’ moral psychology in ways that stabilize those structures. So to understand oppression and domination, we must pay attention not only to structure but also to moral psychology.

5. Response to three criticisms

So far I’ve argued that graded inequality is a better model than the binary model for understanding systems of oppression and domination. It helps us to see certain of the mechanisms by which those systems stabilize themselves. I want to close by responding to three worries. The first is that the gradations in the contemporary systems of oppression and domination on which I’ve focused are less clearly delineated than in the case of caste: that graded inequality oversimplifies. The second is that graded inequality either ignores or is just a manifestation of intersectionality. The third is that positing a structure of graded inequality is politically inapt insofar as it involves objectionable ranking of oppressions.

Delineation and oversimplification

The first worry is that, in contemporary systems of oppression and domination, the gradations are not so clearly delineated as in the case of caste. Patriarchy, we might think, is a good example. There are multiple oppressed groups defined by gender and sexual orientation. There is a clear “top” (meant literally and only slightly tongue-in-cheek), but not necessarily a clear bottom or middle gradations. Are ENBYs lower in the gender hierarchy than binary trans people? Bisexuals higher than pansexuals? That is, we might worry that caste is unique insofar as it displays this “great chain of being” so clearly.

I have two responses to this worry. First, there are systems of oppression and domination that do instantiate the “great chain of being”. Take racism in many Latin American countries, for instance, which is structured according to what Edward Telles calls a “pigmentocracy”, according to which inequalities and discrimination are better predicted by skin color than by historical ethnoracial group belonging.60 Or citizenship categories, which in many countries are intentionally set up to distinguish degrees of distance to “true” citizenship.

This first response isn’t in itself sufficient to allay the concern, insofar as it just provides further examples. The second, deeper response is this. The worry arises from treating these other systems as direct analogies to caste. But I am not analogizing caste directly to these other systems. Rather, I am abstracting a formal structure from caste and using that abstract structure to theorize these other systems. In that sense it is precisely because caste is so unique that it is useful. It is a paradigm or exemplar of graded inequality that, like the idealization of a frictionless plane in physics, allows us to see the underlying mechanisms at work. In other cases, there may be all sorts of complications. Yet once we have identified how graded inequality works (through group interest and affective misdirection) to stabilize the system, we can see better how those complications may actually serve those workings. Some mechanisms of stability may work better where the demarcations are murkier — for example, jostling for position. So in the case of patriarchy, we do see inter-group competition and the affective redirection of anger against other groups. On the binary model, we cannot so easily see these. And the graded inequality model explains why murky divisions serve stability: precisely because it is in the interests of each group to have a clear hierarchy where they are above some other.

But, one might say, doesn’t graded inequality still oversimplify how these systems work? There are multiple, perhaps incommensurable, dimensions of oppression even within a single system. Doesn’t the graded inequality model illegitimately collate these into one all-things-considered hierarchy?

We need not collate so quickly. Instead, we might (depending on the system and on our particular analytic purpose) disaggregate these dimensions. And for each dimension of oppression, we can identify a hierarchy, the details of which would be obviously dependent on the measure to be adopted and so on. With respect to racism, it is true that educational attainment does not necessarily correlate with racialized violence, for example. Yet we can identify gradations within each dimension (say, by measuring “diversity” in college admissions, or inner-city racial violence). And while it is true that even here there is some oversimplification, the use of more complex models (e.g., Claire Jean Kim’s “field” model of Asian Americanness along two dimensions: “superior/inferior” and “foreigner/insider”) becomes quickly system- and even group-specific, whereas graded inequality can be applied to identify more general mechanisms of stability.61

It is also worth remembering, with E.P. Thompson, that structure is not everything.62 There are all sorts of features of oppression and domination that do not directly correlate to, or are not solely caused by, structure. For example, racialized educational inequalities in the U.S. are also a function of histories of migration, of state-specific educational policies, and a variety of other historical processes that are not directly reflected in group structure. So, too, the ideological or stereotypical burdens that different racialized groups bear are not necessarily directly explicable by structural analysis. And this is just to say that structural analysis, like all forms of social analysis, is partial and directed toward particular forms of understanding.

Intersectionality

The second worry is that what I have described as graded inequality is just another word for intersectionality. Don’t these gradations arise because of intersecting dimensions of oppression? For example, the patriarchy oppresses gay men and straight women, and gay men can still be misogynistic and benefit from patriarchy at the expense of straight women, but that arises from the intersection of gender and sexuality. Or, if graded inequality and intersectionality are not the same thing, surely they are related in some way — and perhaps intersectionality can do all the work that graded inequality can do.

I think it is clear that graded inequality and intersectionality are related, insofar as they both concern social positioning and are a corrective to simple binary models of oppression and domination. And it is true that some forms of graded inequality can arise because of intersectionality. But they are distinct. Graded inequality can operate within one dimension of oppression or domination, whereas intersectionality arises from the intersection of multiple dimensions. Graded inequality is (at least in the first instance) a feature of objective structure in formal terms, whereas an important aspect of intersectionality is its focus, whether substantive or methodological or both, on subjective qualitative experience — though I do not, of course, want to suggest wrongly that intersectionality is not about objective social structure.63 Most importantly, I think graded inequality and intersectionality are complementary insofar as they point us toward different features and different dynamics of oppression and domination.

One way to see how they are distinct is to note that a binary model of oppression and domination can still be intersectional. There can be multiple dimensions of oppression or domination, each of which comprises a binary structural relation. Perhaps there are only two sexes and two classes, yet sex and class can still intersect such that the experience of working-class women is qualitatively different from the experience of upper-class women, such that to speak of “women’s experience” without adverting to class will lead us wrongly to focus only on the latter.

What this example shows is that intersectionality (in one central, though not exhaustive sense) concerns the qualitative way in which intersecting dimensions of oppression or domination create new effects that are not merely the sum of the dimensions taken individually, whereas graded inequality concerns the structure of those dimensions of oppression and domination (perhaps taken individually, or perhaps modelled over intersecting axes).

Even where interlocking systems of oppression create gradations that are only revealed through that intersectional approach, intersectionality and graded inequality denote different aspects of that phenomenon. Intersectionality in this sense is methodological — pay attention to the interstices! — and graded inequality is what is revealed through that methodology.

Intersectionality and graded inequality can thus be seen to be complementary. Intersectionality as a methodology points us toward the shape and quality of how oppressions are experienced, and from there back to the interlocking structures that lie behind those experiences. Graded inequality points us initially toward the structures, and from there to the subjective shape of how those structures react back into the structures themselves, to stabilize them.

Ranking Oppressions

This discussion leads us to a third worry. Doesn’t graded inequality run up against one of intersectionality’s important lessons: that oppressions qualitatively differ in ways that resist a ranking of oppressions? Graded inequality, we might worry, does rank oppressions in a flatfooted and objectionable way.

I agree that ranking oppressions comes with some obvious dangers both political and philosophical, though its contrary — refusing to say anything comparative about kinds of oppressions — falls into something comparable to a blind multiculturalism in which all “diversity” is identical in form and only different in substance. And in an obvious sense, the very idea of gradation implies something like the notion of ranking. Yet the way I intend the concept of graded inequality is not about ranking the severity of groups’ experience of oppression or of suffering, but of identifying structural dynamics that shape (but do not determine) those experiences — dynamics that arise from a ranking at the level of structure. One cannot read any simple ranking of severity of oppression directly off that ranking at the level of structure. That is in part because how oppression manifests even within a single dimension can be qualitatively different across gradations. For example, Asian Americans and Latinxs in the middle of the racial hierarchy suffer from alienness stereotypes in ways that African Americans don’t. And even that claim concerns a single dimension of oppression, and does not license (let alone entail) the claim that one can rank oppressions across dimensions of oppression. Much depends on the particular histories and details of forms of oppression. It is to those details and to the experiences of others that one must look to understand, in what ways one can, the oppression of others that one cannot directly experience, as well as for similarities and solidarities. And that lesson of intersectionality as methodology is not contradicted by the idea of graded inequality. Structure is only one determinant of experience.

6. Conclusion

I’ve argued in this paper that the structure of systems of oppression and domination is better understood not as binary, but in terms of graded inequality. Seeing them in this light reveals certain mechanisms by which those systems stabilize themselves. Those mechanisms include the mobilization of group interest and affective misdirection — mechanisms of stability that work through the moral psychology of agents. These mechanisms can operate even where the gradations are not clearly defined. Identifying those mechanisms through the notion of graded inequality is a process that is distinct from, though it can complement, the processes of inquiry and the phenomena picked out by the concept of intersectionality and through other forms of social analysis. And in doing so, one is not committed to any objectionable form of ranking the severity of different kinds of oppressions. Greater understanding of these mechanisms (and others) may help destabilize the systems of oppression and domination to which many are subject. In this way, resistance to the systems of oppression and domination that characterize modern life counsels connecting structural analysis to moral psychology.

Notes

  1. Ayala-López & Beeghly 2020, 213. ⮭
  2. Haslanger 2016, 118. ⮭
  3. E.g., Haslanger 2016; Haslanger 2018. ⮭
  4. On oppression, see Frye 1983; Young 1990; Cudd 2006. On domination, see Hasan 2021; Gourevitch 2013; Gourevitch 2015. That oppression is group-based has reached almost the level of definition. There is more debate about whether domination is best understood as a group-based phenomenon. See Lovett 2010. ⮭
  5. Frye 1983, 8. ⮭
  6. Cudd 2006, 23. ⮭
  7. Pateman & Mills 2007, 87. ⮭
  8. Cudd 2006, 23; Frye 1983, 8. ⮭
  9. See, e.g., Wartenberg 1990. See also List & Pettit 2011. ⮭
  10. Patterson 1982, 35. ⮭
  11. I use the terms ‘master’ and ‘slave’ (and not, e.g., ‘enslavers’ and ‘enslaved people’) here following Patterson. ⮭
  12. See Hasan 2021. ⮭
  13. Patterson 1982, 172ff. ⮭
  14. Binary models are often qualified in various ways, and there are theories of oppression (less often of domination) where there need not be a correlate superordinate group for each subordinate group. I address qualified binary views later in section 3. For views where there is no correlate oppressor group in the sense that there is not a group that intentionally acts to maintain another group’s oppression, there is still a correlate privileged group which benefits from oppression. See, e.g., Young 1990, 41. See also Haslanger 2012, 316. ⮭
  15. Frye 1983, 33. Frye can be read as saying that oppression presupposes that there are at least two distinct categories. ⮭
  16. MacKinnon 1983, 635. ⮭
  17. Lovett 2010, 2. ⮭
  18. Okihiro 2016, 79–80. ⮭
  19. Hegel 2018. ⮭
  20. See Okihiro 2014, 50–51. ⮭
  21. Alcoff 2015. ⮭
  22. Thanks to Taylor Rogers and Thimo Heisenberg for this suggestion. ⮭
  23. See Weisberg 2013; Veit 2019. ⮭
  24. See the large literature on caste and race, e.g., Cox 1959; Wilkerson 2020. ⮭
  25. Ambedkar 2014a, 72. See also Shah 2019, Chapters 1, 5, 8. ⮭
  26. See Ambedkar 2014b. In Ambedkar 2014c, 211, “there cannot be caste in the single number. Caste can exist only in the plural number”. Theorists draw on this conception of caste when they describe other systems of subordination and domination as caste systems. See, e.g., Myrdal 1944; Du Bois 1982 [1904]. More recently, see Wilkerson 2020. For a critique, see Cox 1959. ⮭
  27. See, e.g., Ambedkar 2014d, 106 and following. ⮭
  28. Ibid, 107. ⮭
  29. Ambedkar 2014b, 19–20. ⮭
  30. Untouchability has been theorized in a number of ways. See Dumont 1980; Guru & Sarukkai 2018. Although untouchability has been formally abolished with Article 17 of the Constitution of India (1950), supported by other anti-discrimination constitutional provisions and the Untouchability Offences Act 1955, it is still widely practiced and enforced. See Teltumbe 2020; Yengde 2019; Jodhka 2018; Jaaware 2018. ⮭
  31. The Indian National Crime Reporting Bureau reported over 50,000 caste-based crimes against Dalits and other avarna in 2020, a rise of 9.3% from 2019. See NCRB 2019; NCRB 2020, Table 7A.1. Ninety percent of these were committed by caste Hindus. The kinds of offences are indicative of the underlying intention: to maintain caste hierarchy, to stigmatize. Crimes reported included “forcing victims to eat or drink obnoxious substances; dump[ing] excreta, sewage, carcasses into their homes or compounds; land grabbing; humiliation; sexual abuse”. See Deshpande 2017. Police are often complicit in caste-based violence. A report by the Dalit grassroots organization Swabhiman Society and international women’s rights organization Equality Now 2020 suggests that:

    police frequently failed to record or investigate crimes when initially reported and were sometimes abusive or put pressure on survivors to drop cases. Survivors were not given adequate advice about their legal rights, and some did not receive compensation payments they were entitled to.

    Human Rights Watch 2021 reports police complicity in caste-based and anti-Muslim violence, which is ever-increasing under the present Hindu supremacist Modi government. Dalit women are subject to gang rapes and murders used as a tool to keep Dalit resistance to a minimum (Ganesan 2018; Banerjee 2021). ⮭
  32. Marx 1967, 632. ⮭
  33. One central difference between caste and class in this respect is the role of competition and the commodification of labor as motors of oppression. ⮭
  34. Ibid, 349. ⮭
  35. Ibid, 350. ⮭
  36. Ibid. ⮭
  37. Ibid. See also Braverman 1998; Tinel 2013. ⮭
  38. Marx 1967, 339–40. ⮭
  39. See, e.g., Hobsbawm 2012; Field 1978; Barbalet 1987. ⮭
  40. See Kim 2004. ⮭
  41. I thank an anonymous reviewer for making this point. ⮭
  42. See Bonilla-Silva 2004; Bonilla-Silva 2015. ⮭
  43. Omi & Winant 2014. ⮭
  44. Hacker 2003; Perea 1997; Perea 1998. ⮭
  45. Hacker 2003, 21. ⮭
  46. Compare Kim 1999. ⮭
  47. See Mills 2017. ⮭
  48. Ambedkar 2014a. ⮭
  49. Ambedkar 2014c. ⮭
  50. Ibid, 115. ⮭
  51. Ibid, 116. ⮭
  52. Ibid. ⮭
  53. See Bonilla-Silva 2004. ⮭
  54. Ambedkar 2014c, 116. ⮭
  55. For hope, see Warren 2015; Blöser et al. 2020; Stockdale 2021. For anger see Nussbaum 2013; Pettigrove 2012; Srinivasan 2018. ⮭
  56. See Alcoff 2015. ⮭
  57. Camarillo 2007; Kim 2003. ⮭
  58. E.g., O’Brien 2008. ⮭
  59. Ambedkar 2014c, 384. ⮭
  60. Telles 2014. ⮭
  61. Kim 1999. ⮭
  62. Thompson 1978. ⮭
  63. See, e.g., Collins 1995, 492: “… the notion of intersectionality describes micro-level processes – namely, how each individual and group occupies a social position within interlocking structures of oppression described by the metaphor of intersectionality.” See also Carastathis 2014. Thanks to Taylor Rogers and Annette Martín for pushing me on this point. ⮭

Acknowledgments

My gratitude and appreciation to the Social Philosophy Workshop: Eric Bayruns García, Daniel Brinkerhoff Young, César Cabezas, Megan Hyska, Annette Martín, Laura Martin, Shivani Radhakrishnan, and Phil Yaure. This work has been shaped by our conversations in more ways than can be expressed in footnotes. Thanks must also go to Taylor Rogers for giving wonderful comments at the 2022 Social Philosophy Workshop at University of Illinois Chicago, to two anonymous reviewers and the editors at Philosophers’ Imprint, and to Akeel Bilgrami, Robert Gooding-Williams, Thimo Heisenberg, Philip Kitcher, Zoey Lavallee, Michele Moody-Adams, and Kathryn Tabb for feedback and support at various times in the writing process.

References

Linda Martín Alcoff (2015), The Future of Whiteness (Polity Press).

B. R. Ambedkar (2014a), “Annihilation of Caste”, in Writings and Speeches of B. R. Ambedkar Volume 1, ed. Vasant Moon (Dr. Ambedkar Foundation).

B. R. Ambedkar (2014b), “Castes in India”, in Writings and Speeches of B. R. Ambedkar Volume 1, ed. Vasant Moon (Dr. Ambedkar Foundation).

B. R. Ambedkar (2014c), “Untouchables or the Children of India’s Ghetto”, in Writing and Speeches of B. R. Ambedkar Volume 5, ed. Vasant Moon (Dr. Ambedkar Foundation).

B. R. Ambedkar (2014d), “The Hindu Social Order”, in Writing and Speeches of B. R. Ambedkar Volume 3, ed. Vasant Moon (Dr. Ambedkar Foundation).

Saray Ayala-López and Erin Beeghly (2020), “Explaining Injustice: Structural Analysis, Bias, and Individuals”, in An Introduction to Implicit Bias: Knowledge, Justice, and the Social Mind, ed. Erin Beeghly and Alex Madva (Routledge).

Deeplina Banerjee (2021), “Gang Rape Exposes Caste Violence in India and the Limits of Me Too”, The Conversation.

Jack M. Barbalet (1987), “The ‘Labor Aristocracy’ in Context”, Science & Society 51(2), 133–53.

Claudia Blöser, Jakob Huber, and Darrel Moellendorf (2020), “Hope in Political Philosophy”, Philosophy Compass 15(5).

W. E. B. Du Bois (1982 [1904]), “Caste: That is the Root of the Trouble”, in Writings by W. E. B. Du Bois in Periodicals Edited by Others: 1891 1909, ed. Herbert Aptheker (Kraus-Thompson Organization).

Eduardo Bonilla-Silva (2004), “From Bi-Racial to Tri-Racial: Towards a New System of Racial Stratification in the USA”, Ethnic and Racial Studies 27(6): 931–50.

Eduardo Bonilla-Silva (2015), “The Structure of Racism in Color-Blind, ‘Post-Racial’ America” American Behavioral Scientist 59(11): 1358–76.

Harry Braverman (1998), Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century (NYU Press).

Albert M. Camarillo (2007), “Cities of Color: The New Racial Frontier in California’s Minority-Majority Cities”, Pacific Historical Review 76(1): 1–28.

Anna Carastathis (2014), “The Concept of Intersectionality in Feminist Theory”, Philosophy Compass 9(5): 304–14.

Patricia Hill Collins (1995), “SYMPOSIUM: On West and Fenstermaker’s ‘Doing Difference’”, Gender & Society 9(4): 491–94.

Oliver Cromwell Cox (1959), Caste, Class and Race: A Study in Social Dynamics (Monthly Review Press).

Ann E. Cudd (2006), Analyzing Oppression (Oxford University Press).

Ashwini Deshpande (2017), “The Ugly Reality of Caste Violence and Discrimination in Urban India”, The Wire.

Louis Dumont (1980), Homo Hierarchicus: The Caste System and Its Implications (University of Chicago Press).

Equality Now (2020), “India’s Justice System Is Failing Survivors of Cast-Based Sexual Violence”, Press Release.

John Field (1978), “British Historians and the Concept of the Labor Aristocracy”, Radical History Review 19: 61–85.

Marilyn Frye (1983), The Politics of Reality: Essays in Feminist Theory (Crossing Press).

Deekshitha Ganesan (2018), “Caste Discrimination in India: A Study of NCRB Data (Part IV)”, Centre for Law & Policy Research.

Alex Gourevitch (2013), “Labor Republicanism and the Transformation of Work”, Political Theory 41(4): 591–617.

Alex Gourevitch (2015), From Slavery to the Cooperative Commonwealth: Labor and Republican Liberty in the Nineteenth Century (Cambridge University Press).

Gopal Guru and Sundar Sarukkai (2018), The Cracked Mirror: An Indian Debate on Experience and Theory (Oxford University Press).

Andrew Hacker (2003), Two Nations: Black and White, Separate, Hostile, Unequal (Scribner).

Rafeeq Hasan (2021), “Republicanism and Structural Domination”, Pacific Philosophical Quarterly 102(2): 292–319.

Sally Haslanger (2012), “Oppressions: Racial and Other”, in Resisting Reality: Social Construction and Social Critique (Oxford University Press).

Sally Haslanger (2016), “What Is a (Social) Structural Explanation?”, Philosophical Studies 173(1): 113–30.

Sally Haslanger (2018), “What Is a Social Practice?”, Royal Institute of Philosophy Supplement 82: 231–47.

G. W. F. Hegel (2018), The Phenomenology of Spirit, trans. Michael Inwood (Oxford University Press).

Eric Hobsbawm (2012), “Lenin and the ‘Aristocracy of Labor’”, Monthly Review 64(7): 26–34.

Human Rights Watch (2020), “India: Events of 2020”.

Aniket Jaaware (2018), Practicing Caste: On Touching and Not Touching (Fordham University Press).

Surinder S. Jodhka (2018), Caste in Contemporary India, Second Edition (Routledge).

Claire Jean Kim (1999), “The Racial Triangulation of Asian Americans”, Politics & Society 27(1): 105–38.

Claire Jean Kim (2003), Bitter Fruit: The Politics of Black-Korean Conflict in New York City (Yale University Press).

Claire Jean Kim (2004), “Imagining Race and Nation in Multiculturalist America”, Ethnic and Racial Studies 27(6): 987–1005.

Christian List and Philip Pettit (2011), Group Agents: The Possibility, Design, and Status of Corporate Agents (Oxford University Press).

Frank Lovett (2010), A General Theory of Domination and Justice (Oxford University Press).

Catharine MacKinnon (1983), “Feminism, Marxism, Method, and the State: Toward Feminist Jurisprudence”, Signs 8(4): 635–58.

Karl Marx (1967), Capital: Volume I, ed. Friedrich Engels, trans. Samuel Moore (International Publishers).

Charles W. Mills (2017), Black Rights/White Wrongs: The Critique of Racial Liberalism (Oxford University Press).

Gunnar Myrdal (1944), An American Dilemma: The Negro Problem and Modern Democracy (Harper & Brothers).

NCRB (2020), Crime in India Report 2020.

NCRB (2019), Crime in India Report 2019.

Martha C. Nussbaum (2013), Political Emotions: Why Love Matters for Justice (Harvard University Press).

Eileen O’Brien (2008), The Racial Middle: Latinos and Asian Americans Living Beyond the Racial Divide (NYU Press).

Gary Y. Okihiro (2014), Margins and Mainstreams: Asians in American History and Culture (University of Washington Press).

Gary Y. Okihiro (2016), Third World Studies: Theorizing Liberation (Duke University Press).

Michael Omi and Howard Winant (2014), Racial Formation in the United States (Routledge).

Carol Pateman and Charles Mills (2007), Contract and Domination (Polity Press).

Orlando Patterson (1982), Slavery and Social Death: A Comparative Study (Harvard University Press).

Juan F. Perea (1997), “The Black/White Binary Paradigm of Race: The ‘Normal Science’ of American Racial Thought”, California Law Review 85(5): 1213–58.

Juan F. Perea (1998), “The Black/White Binary Paradigm of Race”, in The Latino/a Condition: A Critical Reader, ed. Richard Delgado and Jean Stefancic (NYU Press).

Glen Pettigrove (2012), “Meekness and ‘Moral’ Anger”, Ethics 122(2): 341–70.

A. M. Shah (2019), The Structure of Indian Society: Then and Now, Second Edition (Routledge).

Amia Srinivasan (2018), “The Aptness of Anger”, Journal of Political Philosophy 26(2): 123–44.

Katie Stockdale (2021), Hope Under Oppression (Oxford University Press).

Edward E. Telles (2014), Pigmentocracies: Ethnicity, Race, and Color in Latin America (University of North Carolina Press).

Anand Teltumbde (2020), Dalits: Past, Present and Future (Routledge, 2020).

E. P. Thompson (1978), The Poverty of Theory and Other Essays (Monthly Review Press).

Bruno Tinel (2013), “Why and How Do Capitalists Divide Labour? From Marglin and Back Again through Babbage and Marx”, Review of Political Economy 25(2): 254–72.

Walter Veit (2019), “Model Pluralism”, Philosophy of the Social Sciences 50(2): 91–114.

Calvin L. Warren (2015), “Black Nihilism and the Politics of Hope”, CR: The New Centennial Review 15: 215–48.

Thomas Wartenberg (1990), The Forms of Power: From Domination to Transformation (Temple University Press).

Michael Weisberg (2013), Simulation and Similarity: Using Models to Understand the World (Oxford University Press, 2013).

Isabel Wilkerson (2020), Caste: The Origins of Our Discontents (Penguin).

Suraj Yengde (2019), Caste Matters (India Viking, 2019).

Iris Marion Young (1990), Justice and the Politics of Difference (Princeton University Press).