Youjin Kong

As a Korean migrant woman in the US academy, I was marginalized at the intersection of race, gender, language, and migration status; my commitment to intersectional feminism grew directly from that experience. But returning to South Korea has changed the map of my privilege: I am now part of the racial-ethnic majority, a native speaker, and a professor with institutional standing. Gender marginalization remains real, but the xenophobic racism I faced in the US is no longer directed at me. What, then, grounds my responsibility to challenge intersecting oppressions in a world where I am no longer their primary target? Drawing on María Lugones’s concept of world-traveling, I conceptualize this experience through what I call “shifting positionality”: the way one’s relation of privilege and marginalization changes as one moves between worlds. I examine three models for grounding the responsibility of world-travelers like myself. The Shared Struggle Model, drawing on Chandra Talpade Mohanty, grounds responsibility in shared marginalization across contexts. The Culpability Model, drawing on Ee Ling Quah, grounds it in our implication in sustaining unjust systems. Both models offer important insights, but each faces decisive objections: the former erases privilege and contextual differences, while the latter conflates implication with culpability and risks blaming marginalized subjects for navigating structural constraints. I then defend a modified version of Iris Marion Young’s Social Connection Model, arguing that our participation in social-structural processes that produce oppressive outcomes grounds our responsibility to address intersecting oppressions—across shifts in positionality, whether we are on the more marginalized or more privileged side. This account speaks not only to my particular trajectory but to world-travelers with shifting positionality more broadly, extending the intersectional feminist framework to account for how the intersection of privilege and marginalization changes as one moves across worlds.

Intersectionality broadly refers to the idea that gender, race, ethnicity, class, sexuality, migration, disability, and other axes of oppression and social identity are not separate but intersect and operate together. Rooted in Black feminism and popularized by legal scholar Kimberlé Crenshaw since 1989, intersectionality has become a cornerstone of feminist scholarship. Despite its wide use across disciplines, however, the term has also sparked debate, and its definition and implications remain far from agreed upon. Debates about intersectionality have raised questions such as: What does it mean to say that oppressions “intersect”? Does this imply that we cannot speak of sexism apart from racism, class exploitation, or other forms of oppression? Can we theorize “women’s” oppression when Black, Latina, Asian, and white women experience different realities? And how can we build solidarity across these intersectional differences? This annotated bibliography introduces key resources for understanding how the notion of intersectionality originated, evolved, and continues to be debated. General Overviews lists resources that provide a general understanding of intersectionality for students, educators, and researchers. Early Roots includes both recent works tracing intersectionality to earlier Black feminist thought and those foundational texts themselves. Critical Engagements surveys major critiques and responses to intersectionality, and Conceptualization and Application features works that refine or employ the concept in theory and practice. As many scholars note, intersectionality is a work in progress—an evolving framework continually reshaped as feminists practice intersectionality in their scholarship and activism. This article is likewise a living document, expanding alongside ongoing feminist and anti-oppression projects.

This review engages with Ee Ling Quah's Fire Dragon Feminism: Asian Migrant Women's Tales of Migration, Coloniality and Racial Capitalism (2025), a timely contribution that theorises 'fire dragon feminism' through the lived experiences of Asian migrant women in the Australian academy. Quah examines their shifting positions of privilege and vulnerability within global racial capitalism, colonialism and neoliberalism, analysing how racialised myths condition migrant women's lives and how myths embedded in corporate equity, diversity and inclusion (EDI) programs further shape their experiences. In this review, I highlight how Quah's framework captures the structural dimensions of emotions such as burnout, hypervigilance and isolation in the white-dominant academy. While appreciating Quah's critique of binaries that portray Asian migrant women as either helpless victims or self-sufficient agents, I raise concerns about her use of 'culpability' to describe migrant women's complicity in oppressive systems. Drawing on Marilyn Frye's and Sukaina Hirji's accounts of 'double binds' and Iris Marion Young's 'social connection' model of responsibility, I argue that Asian migrant women are responsible for structural injustice without being blameworthy. By distinguishing implication and complicity from culpability and blameworthiness, I suggest a more adequate framework for understanding the collective responsibilities fire dragon feminism calls us to embrace.

Fairness in machine learning (ML) has become a rapidly growing area of research. But why, in the first place, is unfairness in ML wrong? And why should we care about improving fairness? Most fair-ML research implicitly appeals to distributive equality: the idea that desirable benefits and goods, such as opportunities (e.g., Barocas et al., 2023), should be equally distributed across society. Unfair ML models, then, are seen as wrong because they unequally distribute such benefits. This paper argues that this exclusive focus on distributive equality offers an incomplete and potentially misleading ethical foundation, especially in the context of text- and image-generation models. Grounding ML fairness in egalitarianism—the view that equality is a fundamental moral and social ideal—requires challenging structural inequality: systematic, institutional, and durable arrangements that privilege some groups while disadvantaging others. Structural inequality manifests through ML systems in two primary forms: allocative harms (e.g., economic loss) and representational harms (e.g., stereotypes, erasure). While distributive equality helps address allocative harms, it fails to explain why representational harms are wrong—that is, why it is wrong for ML systems to reinforce social hierarchies that stratify people into superior and inferior groups—and why ML systems should aim to foster a society where people relate as equals (i.e., relational equality). To address these limitations, the paper proposes a novel multifaceted egalitarian framework for ML fairness that integrates both distributive and relational notions of equality. Drawing on critical social and political philosophy, including the work of Anderson, Young, and Fraser, this framework offers a more comprehensive ethical foundation for tackling the full spectrum of harms perpetuated by ML systems. The paper also outlines practical pathways for implementing the framework across the entire ML pipeline.

The author explores the metaphysics of social identities by using non-ideal theory as a method. She aims to understand what social identities are by examining the experiences of marginalized people – the experiences of people having a social identity as X (e.g., “Latina,” “Muslim woman”) in the non-ideal world, where they are marginalized by virtue of being X. To this end, the author delves into the decolonial feminist philosophies of Uma Narayan, Mariana Ortega, and María Lugones, and engages their conceptual frameworks with Asian and Asian American experiences. Drawing on these analyses, she develops an account of social identities as processes of navigating power: A social identity as X can be best understood as the process as a whole, in which the subject lives as an X by navigating power dynamics, bargaining with them, and highlighting different meanings of being X. By employing non-ideal theory, which centers on power-as-oppression as a factor shaping marginalized people’s lives, we can achieve this nuanced understanding of social identity. Ideal theory, by excluding power from its analysis, is incapable of capturing how social identity is experienced as a process.

Does intersectionality divide marginalized groups (e.g., women) along identity lines (e.g., race, class, and sexuality)? In response to the criticism that intersectional approaches to feminist and critical race theories lead to fragmentation and division, this paper notes that it relies on an ontological (mis)understanding of identity as a fixed entity. I argue against this notion of identity by engaging in a detailed case study of how Asian American women experience their Asian identity. The case study demonstrates that identity is lived not as a pre-given static object but rather as a fluid and flexible process: what Asian identity means varies according to how this identity is related to the dynamics of power. I propose a tripartite taxonomy of identity-power relationships exhibited in Asian American women’s lives, in which identities are 1) constructed by and 2) used to reproduce power as oppression, as well as 3) used to build power as resistance. In particular, I explore Black and Asian feminist solidarities to show that those at the margins recreate their identities and, in doing so, build positive forms of power such as resistance and solidarity. The criticism that intersectionality leads to fragmentation does not hold, as it is grounded in a flawed notion of identity that fails to consider how identities are actually lived in their changing relationships to power—especially how marginalized groups, such as women of color, build solidarity to resist oppression by recreating their identities, rather than being fragmented into smaller and mutually exclusive subgroups.

This paper develops an account of intersectional feminist theory by critically examining the notion of identity implicitly assumed in major critiques of intersectionality. Critics take intersectionality to fragment women along the lines of identity categories such as race, class, and sexuality. Underlying this interpretation, I argue, is the metaphysical assumption that identity is a fixed entity. This is a misunderstanding of identity that neglects how identity is actually lived. By exploring how Asian American women experience their “Asian” identity in their everyday lives (e.g., the “Asian-as-patriarchal vs. White-as-gender-progressive” stereotype, growing anti-Asian racism amid COVID-19, and Asian-Black feminist solidarities), I demonstrate that Asian identity is not fixed but changing according to how it is related to power. I identify and discuss three characteristic types of the identity-power relationship: manifestation of power-as-oppression through the construction of identity, reproduction of power-as- oppression, and creation of new forms of power, namely resistance and solidarity, through the reconstruction of identity. The lives of multiply-oppressed women (e.g., Asian women) can be understood as the locus at which the identity-power relationship is worked out, that is, the power dynamics of oppression are manifested, reproduced, and resisted through the (re)construction of identity. Building on this analysis and engaging discussions on non-ideal theory in social/political philosophy, I argue that intersectional feminist theory can be best explained as a non-ideal theory in a strong sense: a theory that, by focusing on the lives of the multiply oppressed, presents the intersecting dynamics of oppression as central and theory-guiding.

A growing number of studies on fairness in artificial intelligence (AI) use the notion of intersectionality to measure AI fairness. Most of these studies take intersectional fairness to be a matter of statistical parity among intersectional subgroups: an AI algorithm is “intersectionally fair” if the probability of the outcome is roughly the same across all subgroups defined by different combinations of the protected attributes. This paper identifies and examines three fundamental problems with this dominant interpretation of intersectional fairness in AI. First, the dominant approach is so preoccupied with the intersection of attributes/categories (e.g., race, gender) that it fails to address the intersection of oppression (e.g., racism, sexism), which is more central to intersectionality as a critical framework. Second, the dominant approach faces a dilemma between infinite regress and fairness gerrymandering: it either keeps splitting groups into smaller subgroups or arbitrarily selects protected groups. Lastly, the dominant view fails to capture what it really means for AI algorithms to be fair, in terms of both distributive and non-distributive fairness. I distinguish a strong sense of AI fairness from a weak sense that is prevalent in the literature, and conclude by envisioning paths towards strong intersectional fairness in AI.

Feminist critics of evolutionary psychology are often accused of committing the naturalistic fallacy, that is, of inferring certain normative conclusions from evolutionary psychology’s purely descriptive accounts. This article refutes the accusation of the naturalistic fallacy, by showing that evolutionary psychology’s accounts of human behavior are not purely descriptive, but rather grounded on biased value judgments. A paradigmatic example is Randy Thornhill and Craig Palmer’s well-known book A Natural History of Rape. I argue that at least three biased judgments are at work in Thornhill and Palmer’s evolutionary psychology account of rape: (1) adaptationist approaches to evolution, (2) willful ignorance of cultural and social context, and (3) binary understanding of gender and sex. In that Thornhill and Palmer’s account cannot hold without the three biased value judgments, it is not a purely factual description of rape. Therefore, feminist concerns over its normative implications do not commit the naturalistic fallacy, as it is not the case that they infer a normative conclusion from a pure description.

This dissertation addresses the question of how to reconceptualize “women” in order to do a more intersectional feminism. Intersectionality—the idea that gender, race, class, sexuality, and so on operate not as separate entities but as mutually constructing phenomena—has become a gold standard in contemporary feminist scholarship. In particular, intersectionality has achieved success in showing that the old conception of women as a single, uniform concept marginalizes women and others who exist at the intersecting axes of multiple oppressions (e.g., women of color, women in the global South, working-class women, and/or queer and trans people), and thus, demonstrating the need to develop a new way of conceptualizing women. However, the question of what such a new conception would be remains unanswered. Specifically, if feminist theory today is to destabilize the old notion of “women” that relegates multiply oppressed women to the margin of feminism, and yet still needs to use some notion of “women” to critically analyze how the social structure of sexist oppression operates to subordinate women and to dismantle this oppression, how should the concept “women” be reformulated? In this dissertation, I argue that we need to understand women as a concept that is open to constant redefinition, which is socio-historically situated in the actualities of oppression. This is what I refer to as the “situated redefinition” model of women, which is formulated in more detail as follows: The Situated Redefinition Model of Women (SR) – The concept “women” should be always open to being redefined in a way that it could better serve political goals grounded in the actual, daily lives of the marginalized. That is, “women” should be always open to new meanings and provisional definitions that the marginalized would find more useful to achieve political goals, which grow out of their concrete experiences in the current intersecting structures of oppression. My central argument is that we need to understand the concept “women” according to the situated redefinition model, in order to employ this concept for doing intersectional feminism. To support this thesis, the dissertation is divided into two main parts. By engaging with the intersectionality literature, critical race feminisms, Asian/American feminisms, and recent critiques of intersectionality, the first part of the dissertation elucidates what exactly it means for feminism to be more “intersectional.” The second part develops the situated redefinition model and articulates why this model is needed to do intersectional feminism, drawing on the social/political philosophy literature on non-ideal theory, postmodern discussions of universality, and feminist discussions of identity politics. Broadly, this dissertation seeks to shed new light on how we understand and use the concept women for feminist ends. My conceptual model offers a way for feminist scholar-activists to subvert the essentialist/exclusionary notion of “women” without abandoning altogether feminist-political deployments of the concept “women” for the purpose of ending sexist and intersecting oppressions.

Recently, social and political philosophers have shown increased interest in the ideological nature of ideal theory and the importance of non-ideal theory. Charles Mills, who sparked recent critiques of ideal theory, invokes the notion of “non-idealizing abstractions” and argues that these are helpful when applying non-ideal theory. In contrast, I argue that the notion of non-idealizing abstractions is not a helpful tool for non-ideal theory. I suspect that it pays insufficient attention to the actual power dynamics of oppression, which significantly influence judgments about whose experiences and interests are worth being reflected by an abstraction. Failing to take account of the power differences among the oppressed, what Mills considers non-idealizing abstractions falls into ideology, which cannot reflect the experiences or interests of less-privileged minorities, and only concerns those of more-privileged minorities. I examine cases in which the concept of “patriarchy,” which Mills alleges to be a non-idealizing abstraction, functions as what I refer to as the “Colonialist Concept of Patriarchy” that marginalizes Third World women’s experiences of intersectional oppression. I suggest that a more suitable and less ideological way to apply non-ideal theory should avoid asserting that an abstraction is “non-idealizing” and should, instead, protect “resignifiability” of the abstraction.

Richard Rorty, a neo-pragmatist well known for his anti-universalist philosophy, applies his anti-universalist approach to feminism in the paper titled “Feminism and Pragmatism” (1991). In this paper, Rorty claims that universalism is not helpful for feminists in making changes to a masculinist society. In contrast, the main point of my paper is to defend universalism as appropriate to feminism. It is not, however, argued in the form of advocacy for all versions of universalism. I will classify universalism into two distinguished types, ahistoricist-essentialist universalism and historicist, anti-essentialist universalism, and I will defend the latter as consonant with feminism and with Rorty’s own pragmatist approach.