Race, Gender, and Affirmative Action

Resource Page for Teaching and Study

Updated July 2008


This is an annotated bibliography of resources on race, gender, and affirmative action intended for the use of faculty who are designing courses concerning race, gender, and affirmative action, and for students and browsers interested in a guide to the literature.  This bibliography cites three types of sources:  (1) short, accessible articles suitable for undergraduate teaching (noted with an asterisk); (2) longer and more technical works useful for graduate instruction or faculty reference; (3) sources available on the web (underlined, with links). The literature on race, gender, and affirmative action is much too vast to cover on any website.  This bibliography aims to identify a few exemplary works for each identified theme, concentrating on works that are representative of the literature, particularly suitable for teaching or for providing faculty with a background on the issues, and that are especially relevant for assessing the arguments for and against affirmative action policies. Because this bibliography is thematically organized, the same work may appear more than once, sometimes with page references to the portion of the work particularly focused on the given theme.  The focus is on race, particularly on whites and blacks in the U.S., although gender and class are also briefly considered.

Concepts of Race, Gender, and other Social Groups
Cosmopolitanism and Race Eliminativism
What are Racism and Sexism?

Race and Gender Inequality in the United States
Arguments for Affirmative Action Policies

Discrimination-blocking defenses of affirmative action

Arguments Against Affirmative Action Policies
Affirmative Action in Education

Outcome Studies of Affirmative Action

Alternatives to Affirmative Action

Legal Cases about Affirmative Action

Recent Legal Decisions on Affirmative Action

Sources on the Web


Concepts of Race, Gender, and other Social Groups

Concepts of race, gender, and other ascriptive identities (e.g., ethnicity, caste) are understood and used in different ways.  Because the same words are used to refer to different conceptions of these identities, confusion is apt to result when speakers don't realize that they are using these words with different senses.  For instructional purposes, it is therefore worthwhile to survey and address students' understandings of these concepts and to explicitly define your own use of these concepts.  Concepts of race, gender and other ascriptive identities are "cluster" concepts:  they refer to disparate elements, and not all uses of the concept need include every element.
For example, consider the concept of "race."  Any concept purporting to be "racial" must include the following elements:  (a) body type differences (real or imagined):  primarily, skin color, but also hair texture, color, and distribution, stature, musculature, differences in expression of sexual characteristics, etc.; (b) ancestry (real or imagined):  "racial" identity is passed on from parents to children in a lineage (although how identity is passed on may vary--e.g., from ideas of mixed-race to the contrasting "one drop" rule of hypodescent, according to which anyone with a black ancestor is black); and (c) geographical territory of origins (real or imagined):  in the U.S., membership in a racial group is defined by the continent in which one's ancesters, prior to the era of European colonialism, were born (whites from Europe, blacks from Africa, etc.).  Racial concepts were more fine-grained in the 19th and early 20th c., corresponding more closely to ideas of ethnicity or nationality--e.g., Semites, Celts, Slavs. Note that one could have a concept of race which identifies ideas (a)-(c) as imaginary or arbitrary, just as one can have a concept of a "unicorn," understood to be an imaginary animal, or a concept of  "countries lying on 0° longitude," which, while it refers to real entities, is entirely arbitrary.  Thus, one's concept of race need not endorse the reality of races.  Skeptics about human races generally believe that some of the features ascribed to different races are imaginary (for example, the idea, held by some Americans, that blacks have an extra muscle in their foot), but more importantly, that the grouping of individuals into "races" is arbitrary--as arbitrary as political boundaries.
Although (a)-(c) constitute necessary elements of anything that can count as a racial concept, what makes the concept of "race" so politically consequential is the additional elements that may go along with it:  (d) cultural differences (real or imagined):  superficially, in preferences over food, dress, music, etc.; more consequentially, in religion, levels of "civilization," values, virtues and vices (e.g., work ethic, aggression); (e) biological subspecies:  the idea that ancestry defines genetically distinct and isolated breeding populations that have different body types and different capacities or dispositions for manifesting culturally valued or disvalued traits; (f) actual social stratification: the idea that the races are arranged in a social hierarchy of power, privilege, prestige, or wealth; (g) normative social stratification: the idea that the races ought to be arranged in a social hierarchy; (h) subjective identity:  individuals' conceptions of themselves as "raced" and affirmation or repudiation of this identity (e.g., through "passing").
These elements may be linked, and thus support different uses of the concept of race.  The concept of race may be used to:
(1) Justify and explain racial hierarchy.  According to theories of biological racism, (a) - (c) define (e) human subspecies that have different capacities or dispositions for (d) culturally valued or disvalued traits and therefore (f) are and (g) ought to be arranged in a cultural hierarchy.  Much teaching about race focuses on refuting biological racism, especially on rejecting (e).  But rejecting (e) is not sufficient to undermine (g) normative social stratification.  For according to theories of cultural racism, racial groups superficially identified by (a) - (c) constitute (d) distinctive subcultures that foster different culturally valued and disvalued traits that both (f) cause and (g) justify social stratification by race.  Recent survey research indicates that most racially prejudiced Americans lean more towards cultural than biological racism.  Thus, attempts to undermine (g) by attacking (e) will not go very far.
(2) Explain racial hierarchy without justifying it:  According to theories of social construction, racial groups are artifacts of (f) social systems of subordination and inequality that are rationalized by racist ideologies.  Racial groups exist as elements of a social hierarchy of birth, although the racial concepts embedded in racist ideologies are fictitious (refer to nonexistent biological subspecies (e), or systematically distort the character or origins of purported cultural differences (d) among racial groups) and/or arbitrary (reify and naturalize arbitrary political boundaries in the real or imagined ancestral territories of origin of different "races").  Ideologically asserted (a) physical, (b) ancestral, (c) geographical and (d) cultural differences among races may be real or imagined, and may be consequences of the system of social stratification itself.  Social constructionist conceptions of race reject elements (e) and (g) from their concept of race.
(3) Form a basis of subjective racialized identity.  Racist conceptions of race supply a relational subjective identity, based on (g) normative racial hierarchy, which in turn is rationalized in terms of differentially valued purported (a) physical, (b) ancestral, (c) geographical and (d) cultural differences among racial groups.  However, liberatory antiracist movements linked to "identity politics," while rejecting (g), have sometimes also adopted "essentialist" views of race based on a reinterpretation and revaluation of purported group differences of types (a)-(d).  Liberatory antiracist movements opposed to (g) may alternatively be based on subjective identities that conceive of race as socially constructed.  Such identities are justified strategically, and need not survive the eradication of racism.  Eliminate (f) and (g) (actual and normative social inequality) as elements of social structure and subjective understanding, and racial identities might be transformed into relatively benign bases of cultural or ancestral affiliation--much as certain varieties of white ethnicity have become in relation to one another (for example, in the U.S., being of Irish compared to Scandinavian ancestry).
(4) Justify a principle of "colorblindness":  According to this principle, race is an "irrelevant" characteristic of individuals and therefore never a justified basis for treating people differently.  This position follows from a conception of race reduced to (a) - (c) alone.  It is evident that (a) superficial physical differences, such as skin color, are morally irrelevant.  And since the demise of aristocracy, which justified the assignment of rank and privilege according to birth rather than merit, it has become equally evident that (b) ancestry, (c) linked to a geographical territory of origins, is a morally irrelevant basis for distributing basic goods and opportunities.  Rejecting (f) is crucial to the colorblind position.  For if society did in fact systematically assign advantages according to race, this would be unjust from a colorblind point of view, and thus turn a person's race into a morally relevant characteristic (if membership in certain racial groups means that one is unjustly disadvantaged, this may justify remedies directed toward those groups).  Advocates of colorblindness either reject (d) the idea that there are cultural differences among racial groups, or advocate individualism:  the view that people should be judged according to their individual merits, not according to the stereotypical traits ascribed to or manifested by (many or most but not all members of) their group.
 It is important to keep the distinct definitions corresponding to distinct uses of racial concepts clearly in mind when considering policies that respond to group subordination.  For example, Jim Crow laws in the U.S. singled out blacks, understood as members of a biological race, for discriminatory treatment.  Compensatory affirmative action policies single out blacks, understood as occupying disadvantaged social positions in virtue of racist practices, for favorable treatment.  The concept of race upon which Jim Crow laws were built is a fiction, but the concept of race upon which affirmative action policies are built reflect a social reality.
The concept of "race" may be usefully compared to the concept of "caste."  Caste includes, as essential elements, (b) ancestry and (f) actual social stratification, and typically also includes ideas about (a) real or imagined body type differences and (d) socially significant (sub)cultural differences.  Both race and caste are, fundamentally, modes of constructing social hierarchy on relations of birth.  The key difference is that race essentially involves the idea of groups originating in different territories, whereas members of different castes may be conceived as originating from the same territory.

Comprehensive Accounts of Racial Concepts

Haslanger, Sally. "Gender and Race: (What) Are They? (What) Do We Want Them To Be?"  (Department of Philosophy and Linguistics, MIT)
By far the best, most theoretically sophisticated available article-length reference, online or in text, on the concept of race.  It stands out for clearly distinguishing the many different uses and corresponding definitions of race, and also for the comparisons it makes with the concept of gender.  A shorter version of this paper is forthcoming in Nous.

Haslanger, Sally. On Social Construction. (Oxford: Oxford University Press, forthcoming).

This book offers a definitive treatment of the idea of socially constructed phenomena, and applies it to categories
 of race and gender.

*Appiah, K. Anthony. "Race, Culture, Identity: Misunderstood Connections," in Appiah and Gutmann, Color Conscious: the Political Morality of Race. (Princeton: Princeton University Press, 1996).

Superb, highly accessible critique of historical concepts of race as they figured in natural history and biology in the U.S. and Europe.  Defends subjective racial identities as contingent bases for affiliation in a postmodern spirit.  Outstanding for its jargon-free exposition, and for its critique of the claim that racial identities are marked by distinct cultures.

*Zack, Naomi. Thinking about Race. (Belmont, Cal.: Wadsworth, 1998).

Useful accessible undergraduate text covering the concepts of race and racism, mixed race, ethnicity, racial and ethnic identity, the intersections of race, gender, and class, whiteness, race and sexuality, and other topics.  Balances discussion of black, white, Asian, Hispanic, Indian, Jewish, and mixed-race concerns.  Contains excellent suggestions for further reading.

Race as a Biological Concept

*Marshall, Gloria. "Racial Classifications: Popular and Scientific." In C. Loring Brace and James Metress, eds., Man in Evolutionary Perspective, 364-72. New York: John Wiley and Sons, 1973.
Exposes the ways in which supposedly "scientific" concepts of race relied upon confused popular representations of racial groups, and how these popular representations depend upon social rather than biological factors.  Useful discussion of varieties of racial classifications, inside and outside the U.S.

*Livingstone, Frank. "On the Nonexistence of Human Races." In Ashley Montague, ed. The Concept of Race (New York: Free Press, 1964).  Reprinted in Sandra Harding, ed., The "Racial" Economy of Science (Bloomington: Indiana University Press, 1993), 133-141.

Classic critique of the biological concept of race from a genetic and evolutionary point of view.  The Montague volume in which this appears collects many important articles from the postwar scientific critique of "scientific" racism.

*Stepan, Nancy Leys and Gilman, Sander, "Appropriating the Idioms of Science: The Rejection of Scientific Racism," in Dominick LaCapra, ed., The Bounds of Race: Perspectives on Hegemony and Resistance (Ithaca, N.Y.: Cornell University Press, 1991).    Reprinted in Sandra Harding, ed., The "Racial" Economy of Science (Bloomington: Indiana University Press, 1993), 170-193.

Documents the critiques made by African-Americans and Jews of racial concepts and "scientific" racism from 1870-1920.

*Olson, Steve, "The Genetic Archaeology of Race," Atlantic Monthly April 2001.

Excellent online article on the latest understanding of human genetics, debunking the idea that the racial classifications used by different societies have a genetic basis.  Elegantly explains how this can be true even though there are some genetically based physical differences (e.g., in skin color) broadly associated with membership in different groups of people popularly classified as belonging to different races.
Risch, Neil;  Burchard, Esteban; Ziv, Elad and Tang, Hua, "Categorization of humans in biomedical research: genes, race and disease," Genome Biology 3 (2002).

Dissents from the dominant social science view that racial categories are social constructs, arguing that they
have medical significance for understanding the incidence and treatment of disease.  An important dissenting view:
Risch is a leading geneticist at Stanford University.
Bamshad, Michael; Olson, Steve, "Does Race Exist?" Scientific American, December 2003.

Very useful popular article, that makes a nice complement to Risch et al's more technical article, above.  Offers
a nuanced answer to the question:  "If races are defined as genetically discrete groups, no. But researchers can use some genetic information to group individuals into clusters with medical relevance."

Andreasen, Robin, "A New Perspective on the Race Debate," British Journal of the Philosophy of Science 49 (1998): 199-225.
Andreasen, Robin, "Race: Biological Reality or Social Construct?," British Journal of the Philosophy of Science 67 (2000): S653-S666.

Andreasen argues that, given more sophisticated cladistic conceptions of human population groupings, that a biologically significant concept of race can be identified consistent with social constructionist theories of race.

Race as a Social Construct

*Blauner, Robert. "Colonized and Immigrant Minorities." In Ronald Takaki, ed. From Different Shores, 2d. ed., 149-60. Oxford: Oxford University Press, 1994.
Explains the difference between "racial" and "ethnic" minorities in terms of the circumstances of their incorporation into the U.S.  The groups currently thought of as "racial" were "colonized"-- forcibly incorporated by conquest or kidnapping, deprived of legal equality, forced into subordinated socioeconomic roles, and deprived of their cultures.  By contrast, "ethnic" or "immigrant" minorities chose to come to the U.S. (although they may have been forced out of their home countries, they usually had a choice of destinations), enjoyed legal equality, and were permitted to advance and integrate themselves in U.S. institutions after relatively brief struggle.

*Takaki, Ronald, ed. From Different Shores, 2d. ed. Oxford: Oxford University Press, 1994.

Highly accessible teaching anthology on the social construction of race and ethnicity in the U.S., paying attention to the intersections of race and ethnicity with class, gender, and age, and to public policy controversies.  In addition to the outstanding Blauner article cited above, contains useful articles discussing the particular experiences of a wide variety of ethnic and immigrant groups from Asia, Africa, Latin America, and Europe.

Ignatiev, Noel. How the Irish Became White. (New York : Routledge, 1995).

Traces one story in the social construction of whiteness: the ways Irish immigrants to the U.S. gained access to the status of whites by advancing their interests at the expense of African-Americans.

Race/Ethnicity as a Subjective Identity

 Zack, Naomi. American Mixed Race: the Culture of Microdiversity (Lanham: Roman and Littlefield, 1995).
Explores the social construction of race through mixed-race identities.  Discusses  reaffirmations of pure racial categories, suggestions for new multiracial identities, and arguments for eliminating "race."

*Boxill, Bernard. "Separation or Assimilation?" in John Arthur and Amy Shapiro, eds., Campus Wars: Multiculturalism and the Politics of Difference (Boulder, Col.: Westview, 1995), pp. 235-248.

Nuanced discussion by a distinguished African-American philosopher of the dilemma between separation and assimilation posed by DuBois and black cultural nationalists.  Rejects the terms of the dilemma, arguing that there is no duty of blacks to identify or affiliate particularly with other blacks, but that there may be good reasons to so identify.

*Kennedy, Randall. "My Race Problem -- And Ours: A consideration of touchy matters -- racial pride, racial solidarity, and racial loyalty -- rarely discussed," The Atlantic, May 1997.

In contrast with Boxill, rejects racial pride, racial solidarity, and racial loyalty as inappropriate attitudes to take toward one's race.  This paper, along with Boxill's, above, would make an excellent pair of articles contrasting two points of view on subjective racial identity.

Chavez, Linda. Out of the Barrio : Toward a New Politics of Hispanic Assimilation. (New York: Basic Books, 1992).

Famous attack, by a conservative Hispanic political activist, on bilingualism and similar policies aimed at preserving cultural differences of immigrants from Latin America.  Defends an assimilationist ideal for Hispanics.

*Wasserstrom, Richard. "On Racism and Sexism," in Philosophy and Social Issues (Notre Dame: Notre Dame University Press, 1980), pp. 11-29.

Classic philosophical discussion of a color-blind ideal of society.  Argues that a sex-blind ideal is not as plausible, and considers a pluralist (non-assimilationist) alternative ideal of society with respect to both sex and race.

*Young, Iris Marion. "Social Movements and the Politics of Difference," in John Arthur and Amy Shapiro, eds., Campus Wars: Multiculturalism and the Politics of Difference (Boulder, Col.: Westview, 1995).

Criticizes assimilationist ideal of liberation, conceived as the transcendence of group difference.  Defends liberatory movements that stress "a positive self-definition of group difference", applying the ideal of a "heterogeneous public" to cases involving women, American Indians, and non-English speakers in the U.S.  This paper, combined with a selection from Hollinger, below, would provide an outstanding pair of papers for teaching purposes that contrast two distinct visions of multiculturalism.

*Hollinger, David. Postethnic America: Beyond Multiculturalism. (New York: Basic Books, 1995).

Contrasts a cosmopolitan ideal of voluntary ethno-racial identity with more particularistic conceptions of fixed social identity that the author takes to characterize much multicultural discourse and politics.  Criticizes the idea that ethno-racial identity ought to take precedence over national identity, while also rejecting ideals of assimilation and "color-blindness."

Taylor, Charles. Multiculturalism: Examining the Politics of Recognition. (Princeton: Princeton University Press, 1994).

Sophisticated philosophical defense of multiculturalism by noted Canadian philosopher Charles Taylor. Includes commentary by Jurgen Habermas, Amy Gutmann, Anthony Appiah, and others.

Schlesinger, Arthur M., Jr. The Disuniting of America : Reflections on a Multicultural Society (New York: W.W. Norton, 1998).

Attacks multiculturalism for creating a Balkanized society.  Poses a useful contrast to Taylor's more favorable view, above.
Tommie Shelby, We Who Are Dark: The Philosophical Foundations of Black Solidarity (Cambridge, Mass.: Belknap Press, 2005)

While rejecting biological accounts of race and the excesses of multiculturalism, Shelby defends the importance of black solidarity in the pursuit of racial justice.  An important statement.

Frankenburg, Ruth. White Women, Race Matters : The Social Construction of Whiteness. (Minneapolis: University of Minnesota Press, 1993).

Interviews with white American women explore their racial identities "from the inside".  This book helped shift scholarly work on race toward greater attention to the "unmarked" white racial identity.

Cosmopolitanism, Race Eliminativism, and the Ethics of Identity

Recent critiques of identity politics have led some to advocate eliminativism about racial concepts--simply dropping them from our conceptual frameworks.  Others have been inspired by such critiques to advocate cosmopolitanism as an attractive ideal.  Cosmopolitanism does not require the elimination of subjective racial identites, but does require pursuit of wider identifications.  (This notion of cosmopolitanism, tied to subjective identities, should be distinguished from cosmopolitanism, understood as an approach to global justice that takes individuals rather than nations or other collectives as the units of fundamental moral standing for purposes of judging the justice of the whole world, and regards principles of justice as including everyone and applying to everyone.)

Appiah, Kwame Anthony.  The Ethics of Identity. (Princeton: Princeton University Press, 2004).

Important attempt to reconcile the claims of individualism and community within a perfectionist liberal framework that encourages people to engage with those of different identities.

Appiah, Kwame Anthony. Cosmopolitanism: Ethics in a World of Strangers. (W.W. Norton, 2006).

Continuation of Appiah's argument, above, for a "rooted cosmopolitanism" whereby each person can simultaneously identify with parochial groups and seek common identifications on a global scale.  A vigorous rebuttal of critiques of cosmopolitanism.

Callon, Eamonn. "The Ethics of Assimilation," Ethics  115 (2005): 471-500.

Defends the claim that it is not wrong to assimilate into another culture, even if this means abandoning identification of the community of one's birth.  Don't miss the discussion of Callon's article by Lawrence Blum and Lionel McPherson at MIT's Symposia on Gender, Race, and Philosophy.

Kleingeld, Pauline, Brown, Eric, "Cosmopolitanism", The Stanford Encyclopedia of Philosophy (Fall 2002 Edition), Edward N. Zalta (ed.).

Excellent overview, with bibliography, of contemporary conceptions of cosmopolitanism and its critics.

Nussbaum, Martha. For Love of Country: Debating the Limits of Patriotism. (Boston: Beacon Press, 1996).

Nussbaum's lively defense of a cosmopolitan ideal joined with numerous sharp critiques, mostly defending more parochial notions of identity.  An excellent introduction to debates on the ethics of ethno-national vs. cosmopolitan identification.

Sen, Amartya. Identity and Violence: the Illusion of Destiny. (W.W. Norton, 2006).

Incisive critique of identity politics.  Rejects the idea that any particular identity--especially of the ascriptive (involuntary) sort, lays a claim on the individual's loyalties, as opposed to others she might choose.  Stresses that we all are subjects of multiple, overlapping identities and capable of constructing new identities through engagement with others that may be more valuable than the ones we are born with.
Webster, Yehudi. The Racialization of America (New York: St. Martin's, 1992).

Argues that, since races don't exist, and the use of racial concepts is harmful, we should eliminate racial concepts and racial discourse from our lives.

What are Racism and Sexism?

"Racism" and "sexism" are probably the most contested terms in debates over affirmative action.  This is because, while they are descriptive terms, they also carry strong evaluative implications.  Few people are willing to apply these terms to beliefs, attitudes, or practices that they do not condemn.  Consequently, disputes over definition flow from disagreements about what is wrong with racism and sexism.  There are three key axes of disagreement:
(1) Individual or group harm?  Some people argue that harm can happen only to individuals, not groups.  On this view, a harm or injustice inflicted on someone is not made more objectionable because it has a racist motive, purpose, or cause.  A racist lynching is wrong because it is a lynching, and is not made more wrong because it is racist.  In addition, on this view there is no reason to care about inequalities between groups; the only unjust inequalities obtain among individuals.  Others argue that groups can suffer harms, or, more specifically, that individuals can share in a harm in virtue of their group membership.  As an example, they point to group stigmatization.
(2) Classification or inequality?  Some people hold that any belief, attitude, or practice that classifies people by race (and to a lesser extent, gender) is per se morally objectionable.  Such classifications are thought to be wrong because race is an involuntary characteristic, or thought to be irrelevant to what really matters about a person.  On this view, all acceptable attitudes and practices must be colorblind.  Others argue that some uses of racial classifications are benign or at least not of great consequence, and that what is objectionable is the use of such classifications in practices that create or sustain group inequality or oppression.
(3) Purpose or impact?  Some people argue that, as long as no person specifically intends that their policies treat people differently according to their race--that is, as long as they did not consciously engage in racial classification in their action--then their action is not racist.  Others argue that policies that have an unequal impact on different people because of their race are objectionable, even if the unequal impact was not explicit or even intended.  The key point is that race played a causal role in generating the inequality, not whether the causal role passed through people's conscious deliberations.  There are 3 basic cases:  (a) Unconscious or covert racism.  People might treat people differently according to unconsciously held racial stereotypes or cognitive schemas that structure their perceptions and habits, although they would disavow such representations if they were brought to consciousness.  Or, they may harbor unconscious feelings of antipathy toward people on account of their perceived race, and treat them accordingly.  (b) Secondary racism, or racism by proxy. This occurs when a facially race-neutral basis for discrimination is accepted at least in part because it tracks race.  For example, the Social Security system for its first several decades did not cover domestic servants or agricultural workers.  This was the price white Southern politicians exacted from President Roosevelt for their support for the Social Security Act.  Most of the excluded workers were black or Hispanic, and white Southerners did not want to pay for the retirement of their nonwhite employees.  Here the racist motive was explicit, but it is not hard to imagine covert racist motives operating in a similar fashion.  As long as the facially race-neutral classification would have been rejected, had its negative impact been felt more strongly by a socially advantaged race, the policy depending on the classification is considered a form of racism by proxy.  (c) Institutional racism which is neither overt, covert, nor secondary, includes policies that perpetuate the legacy of racial discrimination by means of classifications that disproportionately impact disadvantaged racial groups, but are not accepted because they do this.  An example would be the practice of colleges and universities giving preference in admissions to the children of alumni ("legacies").  If these schools practiced racial discrimination in the past, their alumni will be disproportionately white, and so will their legacies.  Thus, certain applicants will have an edge in admission because their parents are white.  This comes closest to a pure differential impact standard, since it applies regardless of present (but not past) purpose or motive.  It is the most controversial standard of racism for this reason.  Fundamentally, it claims that we ought to avoid not only intentional racial harm, but negligent racial harm.  It asserts a duty of care to avoid reinforcing or magnifying the harmful consequences of past racial discrimination.
For teaching purposes one might want to consider whether the word "racist" is too emotionally charged to be worth using.  Many students resent the application of this term to cases lacking hostile intent, even if they might accept that a policy could be objectionable because its causal history and impact are inappropriately tied to race.
General Works
*Lawrence Blum, "I'm Not a Racist, But. . .": The Moral Quandary of Race (Ithaca: Cornell University Press, 2002).
Sophisticated, sensitive, nuanced discussion of moral problems concerning race.  Especially good at resisting the tendency to think that any issue involving race is one involving racism.  Argues that to avoid expanding the meaning of the term to the point of reducing its normative force, we need to distinguish racism (antipathy or contempt for people on account of their race) from other phenomena involving race (e.g., racial anxiety, racial ignorance).  The other phenomena are also of moral concern, but not proper grounds for moral condemnation.  Also contains an excellent discussion of what "race" is.

Individual vs. Group Harm

*Levin, Michael. "Is Racial Discrimination Special?," Journal of Value Inquiry 15 (1981): 225-232.  Reprinted in T. Mappes and Jane Zembaty, eds. Social Ethics (McGraw-Hill, 1987), pp. 192-199.
Denies that racist motives add to the wrongness of any action, because only individuals, not groups, are subjects of moral harms and duties.  Argues that it follows from this fact that there is no case to be made that wrongs against racial groups deserve special rectification measures, such as affirmative action.

*Brooks, D. H. M., "Why Discrimination is Especially Wrong," Journal of Value Inquiry 17 (1983): 305-312.  Reprinted in T. Mappes and Jane Zembaty, eds. Social Ethics (McGraw-Hill, 1987), pp. 199-204.

Argues, against Levin, that systematic, institutionalized racial discrimination is especially wrong and deserving of special rectification measures.  A key point of his argument concerns the ways group identities magnify harms motivated by group antipathy:  the terror of a racist lynching is experienced not just by the particular victim, but by everyone else in the group who is symbolically targeted by the lynching as considered fit for subhuman treatment.

*Steven M. Cahn, The Affirmative Action Debate, 2d. ed. (New York: Routledge, 2002)

Contains 8 short pieces on individual vs. group treatment.  Because this volume has just been published, the papers are more accessible than those listed above.  However, none of them in combination is as good in representing the issues than the Levin-Brooks pair above, because they are all so brief.

Owen Fiss, "Groups and the Equal Protection Clause," Philosophy and Public Affairs 5 (1976): 107-177.

Classic paper defending a group-based view of equal protection; advocating group remedies as solutions to group-based harms.

 Classification vs. Inequality

*Eastland, Terry. "The Fight for Colorblind Law," ch. 2 of Ending Affirmative Action: the Case for Colorblind Justice (New York: Basic Books, 1997)
Fundamental defense of a colorblind ideal of justice.

*Loury, Glenn. "The Divided Society and the Democratic Idea," University Lecture, Boston University, October 7, 1996.

Argues that "explicit attention to inequality between
Blacks and Whites, as distinct from a purely color-blind
concern about inequality among individuals, is essential to
the attainment of social justice."  Important speech by a noted African-American economist who is usually labelled "conservative" for his criticism of the contemporary civil rights racial agenda, including affirmative action, but whose views are in fact not easy to classify.

Purpose vs. Impact

Garcia, Jorge. "The Heart of Racism," Journal of Social Philosophy 27 (1996): 5-46.
Defends the view, against theories of institutional racism or oppression theories, that racism requires bad intentions or attitudes toward racial groups.

Piper, Adrian. "Higher-Order Discrimination," in Owen Flanagan and Amelie Rorty, eds. Identity, Character, and Morality: Essays in Moral Psychology (Cambridge, Mass: MIT Press, 1990).

Important philosophical discussion of covert racism, whereby people may express racial antipathy in their behavior without being aware of or endorsing such attitudes.

*Ezorsky, Gertrude. "Overt and Institutional Racism," ch. 1 of Racism and Justice: the Case for Affirmative Action. (Ithaca, N.Y.: Cornell University Press, 1991).

Perhaps the single best, brief, most accessible source for undergraduate instruction on what institutional racism is.

*Frye, Marilyn. "Sexism" in The Politics of Reality: Essays in Feminist Theory. (Freedom, Cal.: Crossing Press, 1983), pp. 17-40.

Highly accessible teaching source; defines sexism as a system of oppression based on classifying people by sex.  This careful paper, along with "Oppression" in the same volume, are feminist classics.

*Young, Iris Marion. "Five Faces of Oppression," ch. 2 of Justice and the Politics of Difference (Princeton: Princeton University Press, 1990).

Discusses oppression as a structural phenomenon expressed in exploitation, violence, cultural imperialism, marginalization, and subordination.  Defends the relevance of considering social groups rather than individuals as the focus of moral and political concern.

Alfred Blumrosen, "Strangers in Paradise: Griggs v. Duke Power Co. and the Concept of  Employment Discrimination," in Paul Burstein, ed., Equal Employment Opportunity (New York: Aldine de Gruyter, 1994), pp. 105-120.

Important legal defense of the disparate impact standard of racial discrimination, by the principal legal inventor of the standard.  The author, as head of EEOC, successfully persuaded the Supreme Court to adopt the disparate impact standard for Title VII employment discrimination claims in Griggs.  Blumrosen usefully develops the different standards of discrimination by analogy with the different standards of liability in torts (malice, negligence, strict).  He also makes it clear that the prime rationale for the differential impact standard is not that racial proportional representation is a first principle of justice, but that gratuitous impositions of burdens on those already disadvantaged on account of race are unfair.

Race and Gender Inequality in the United States

Diagnosis of the causes of race and gender inequality is indispensible to evaluating affirmative action policies.  These policies are proposed as a partial remedy for the effects of past and present race- and gender- based disadvantage.  However, such disadvantages ought to be compensated or corrected for only if they are unjust.  This requires a demonstration that the actual distribution of advantages in the U.S. does not follow principles of just distribution.  It may seem that this determination depends more on disagreements about principles of justice than on disagreements about the causes of inequality.  Thus, advocates of affirmative action are often thought to support "equality of outcome" while critics support "equality of opportunity" or perhaps a libertarian theory that permits private organizations to engage in race and gender discrimination in a free market.  However, the impact of differences in underlying theories of justice on the debate over affirmative action is less than often supposed.  (1) There is widespread consensus in the U.S. on the principle of equality of opportunity; thus, most people evaluate the justice of current distributions by this standard.  (2) Even advocates of equality of outcome agree that inequalities due to the fully voluntary choices of individuals among equivalent opportunity sets are unobjectionable.  (3) Even libertarians agree that disadvantages due to state-sponsored racism and sexism are unjust, and that reparations are due to those whose property has been taken away in a history of unjust (coercive) transfers.
In practice, then, most disputants agree on the normative implications for affirmative action of different causal accounts of socioeconomic inequality.  The case for affirmative action is undermined if the fundamental causes of race and gender inequality can be traced to (a) an innate, biological inferiority of the intended beneficiaries of affirmative action in intelligence, motivation to work hard, prudence and self-control, or other attributes needed for success; (b) cultural pathologies of the intended beneficiaries that interfere with their ability or willingness to take advantage of opportunities; or (c) voluntary choices to pursue less rewarding opportunities.  The case for affirmative action is advanced if the fundamental causes of race and gender inequality can be traced to racism and sexism as discussed above--that is, to (a) overt, intentional discrimination by private parties or the state, on grounds of race and gender; (b) unintentional discrimination by individuals due to their unconscious acceptance of unjust stereotypes, evaluative biases, or group-based antipathies; or (c) structural obstacles to the advancement of nonwhites and white women whose causal antecedents can be traced to (a) or (b).  The literature on causes of group inequalities is so vast that only a brief sampling of theories can be offered here.

Racial Inequality

Biological Theories
Levin, Michael. Why Race Matters : Race Differences and What They Mean (Human Evolution, Behavior, and Intelligence) (Praeger, 1997).
Defense of unreconstructed biological racism.  Argues, on the basis of black-white differences in intelligence tests and crime rates, that racial inequalities in society are caused by genetic differences in intelligence and that higher black crime rates are due to genetically based black impulsiveness.  Expresses more certainty and stridency about its conclusions than the more famous and circumspect Bell Curve. Includes a fairly comprehensive survey of the literature on race differences in intelligence and criminality oriented toward genetic hypotheses.

Herrnstein, Richard and Murray, Charles. The Bell Curve: Intelligence and Class Structure in American Life. (New York : Free Press, 1994).

Most controversial for its suggestion that blacks are genetically determined to be less intelligent than whites.  Its central argument, however, is class-based:  that differences in IQ are a primary cause of differences in socioeconomic success.

Fraser, Steven, ed. The bell curve wars : race, intelligence, and the future of America (New York : BasicBooks, 1995).

Comprehensive critique of the science, data, and ideology of The Bell Curve.

Ned Block, "How Heritability Misleads about Race," Boston Review 20 (6) 1996: 30-35.

Outstanding online article on the misuse of concepts of genetic heritability in relation to claims about genetic causes of racial differences.

Sowell, Thomas. "Ethnicity and IQ," American Spectator, Feb. 1, 1995, pp. 32.

While defending the authors of the Bell Curve from liberal attackers, argues that the "Flynn effect"--the demonstrated historical increase of many groups' I.Q.'s over a few decades (too short a time for significant changes in their gene pool) of orders of magnitude comparable to the black-white IQ gap--is devastating to their thesis that blacks are genetically inferior in intelligence to whites.
Cultural Pathology/Voluntary Choice Theories
Murray, Charles. Losing Ground (New York: Basic Books, 1985).
Blames the welfare state for inculcating dependency, bad behavior, and consequently poverty in blacks and the underclass because of its perverse incentives and destruction of moral standards.

Jencks, Christopher. Rethinking Social Policy (Cambridge, Mass.: Harvard University Press, 1992).

Criticizes Murray's Losing Ground data and logic in ch. 2.

D'Souza, Dinesh. The End of Racism: Principles for a Multicultural Society. (New York: Free Press, 1996).

Blames the relative lack of black socioeconomic success on black cultural pathology, denying that white racism is any longer a significant factor.  In odd contradiction to this, acknowledges and defends "rational" (i.e., self-interested) discrimination against blacks based on negative statistical generalizations about them.  Reconciles the contradiction by denying that such discrimination constitutes racism.  Calls for ending affirmative action and repealing antidiscrimination laws.

Sowell, Thomas. Race and Culture: a World View. (New York: Basic Books, 1994).

Considerably more sophisticated than D'Souza.  Argues that the primary determinant of the socioeconomic success of ethnic groups worldwide is neither genes nor prejudice, but their "cultural capital" (work habits, orientation to the market, education, etc.).

Anderson, Elijah. "The Code of the Streets," The Atlantic, May 1994.

A different take on cultural pathology: argues that inner city blacks, the vast majority of whom do not endorse the "street ethics" of crime, drugs, and violence, are trapped into "street" norms of honor and revenge as a condition of survival.  This creates a vicious cycle, whereby hopelessness of marginalization leads to the quest for "respect" via violent street norms, and conformity to these norms in turn ensures deeper marginalization and isolation from mainstream opportunities.

Thernstrom, Steven and Thernstrom, Abigail, America in Black and White (New York: Simon and Schuster, 1997).

By far the most important book arguing that reponsibility for improving their situation today lies mainly with blacks.  Contains a lengthy and widely acclaimed history of antiblack racism in the U.S.  Argues that racism has ceased to be a major force in blacks' lives in recent years.  Instead, blacks need to study harder, adapt to an entrepreneurial culture, and reject crime and out-of-wedlock childbearing.  Argues that reliance on the civil rights agenda has disserved blacks.  The great strength of the book is its extensive historical evidence, some of which poses sharp challenges to William J. Wilson's hypothesis that black unemployment is due to a spatial mismatch of blacks and jobs.  If jobs are lacking in the inner cities, then how have Asian immigrants managed to succeed there?  The great weakness of the book is its failure to consider the continuing importance of racial segregation to the fate of African-Americans, and its consistent downplaying of evidence of continuing racism in the U.S.
Discrimination/Stigmatization Theories
Dovidio, John and Gaertner, Samuel. "On the Nature of Contemporary Prejudice: the Causes, Consequences, and Challenges of Aversive Racism," in Jennifer Eberhardt and Susan Fiske, eds. Confronting Racism: the Problem and the Response (Thousand Oaks, Cal.: Sage Publications, 1998).
Dovidio, John and Gaertner, Samuel.  Prejudice, Discrimination, and Racism (San Diego: Academic Press, 1986).
Surveys extensive empirical evidence for the phenomenon of "aversive racism", a form of covert racism whereby self-described colorblind individuals manifest racially discriminatory attitudes and behaviors of which they are unaware.

*Kirschenman, Joleen and Neckerman, Kathryn. "We'd Love to Hire Them, But. . . ": The Meaning of Race for Employers," in Christopher Jencks and Paul Peterson, eds. The Urban Underclass (Washington, D.C.: Brookings Institution, 1991), pp. 203-232.

Documents complex patterns of racial discrimination among Chicago-area employers, based on interviews with them.  Useful for considering the interactions of race with ethnicity, class, and other factors.

*Glenn Loury, The Anatomy of Racial Inequality (Cambridge, Mass.: Harvard University Press, 2002)

Identifies racial stereotypes and racial stigma as fundamental causes of racial inequality in the U.S., offering a model of the dynamics of these psychological constructs in reproducing inequality through the interactions of individuals.  Loury's model is not a "discrimination" model, but rather a "stigmatization" model. His point is that racial stigmatization plays a more important role than direct discrimination in producing racial inequality, by depriving blacks of opportunities for development.  Inequalities initially produced through intentional discrimination can become entrenched, perpetually reproducing themselves via voluntary market interactions of individuals thereafter. A very important book.

Ian Ayers, Pervasive Prejudice?: Unconventional Evidence of Race and Gender Discrimination (Chicago: Chicago University Press, 2001).

Important empirical study of racial and gender discrimination in retail markets. Ayers is one of the leading researchers who employs the audit testing method to detect market discrimination. This involves sending out carefully matched black and white, or male and female, prospective consumers to different retailers and finding out if they are treated differently by the firms they visit. Ayers found evidence of pervasive discrimination against blacks and women in retail car negotiations, and discrimination against blacks in bail bonding and access to kidney transplants. This constitutes very strong empirical evidence against Gary Becker's famous hypothesis that markets drive out discrimination. Glenn Loury, in The Anatomy of Racial Inequality, cited above, offers theoretical models of market interactions that explain Ayers' empirical results.

*Steele, Claude. "Race and the Schooling of Black Americans," The Atlantic, April 1992.

Argues that the lower educational performance of black relative to white students is due to racial stigma. Under the sway of racial stereotypes, teachers fail to notice and encourage black academic performance.  Internalization of racial stereotypes of black intellectual inferiority threatens black students with the prospect that their failures will be interpreted as reflecting the failures of their race.  Disidentification with school is a strategy for coping with the stereotype vulnerability that threatens black students' self-esteem.

Barbara Reskin, "The Proximate Causes of Employment Discrimination," Contemporary Sociology 29 (2000): 319-329.

Excellent brief summary of the theory and evidence of how discrimination works through unconscious cognitive biases--stereotypes, attributions, ingroup preference, etc.--and how workplaces can be changed to block the operation of these biases.

*"True Colors" ABC Primetime Live Segment, Sept. 26, 1991.

By far the best, most vivid demonstration available today of the persistance of racial discrimination in the U.S.  Perfect for instructional purposes at 20 minutes in length.  Available in many university video libraries.  ABC sent two carefully matched "testers", one white, one black, to St. Louis to see how they faired in numerous encounters:  shopping for shoes, renting a video, searching for an apartment, seeking employment, buying a car, hailing a taxi, and so forth.  They captured egregious cases of worse treatment of the black tester in every setting.  The cumulative impact of these cases over a short period of time is stunning.
Institutional (Structural) Theories
Tilly, Charles. Durable Inequality (Berkeley and Los Angeles: U of California Press, 1999).

Important general theory of the causes of group inequality, by one of the most distinguished, creative, and prolific sociologists of the 20th century.  Tilly argues that the basic causes of systematic group inequality are the same, regardless of the type of group at issue--race, gender, caste, ethnicity, etc.  A must read for anyone interested in social stratification.

Bonilla-Silva, Eduardo. "Rethinking Racism: Toward a Structural Interpretation," American Sociological Review 62 (3) (1997): 465-480.

A structural view of racism from the perspective of sociological theory.

*Hacker, Andrew. Two Nations : Black and White, Separate, Hostile, Unequal. Rev. ed. (Ballantine Books, 1995).
Most valuable for its extensive statistical documentation of black-white racial inequality in the U.S.  Readers not already persuaded by theories of institutional racism tend to doubt the author's causal claims that black disadvantages can be attributed to white racism, in part because of his neglect of alternative hypotheses.

*Massey, Douglas and Denton, Nancy. American Apartheid : Segregation and the Making of the Underclass (Cambridge, Mass.: Harvard University Press, 1993).

Focuses on pervasive residential segregation as a linchpin of continuing black socioeconomic disadvantage.  Extremely useful source of data on housing segregation, including detailed and rigorously documented evidence of continuing systematic and extensive housing discrimination practiced by landlords and real estate agents.  Discusses the ways housing segregation underwrites other explanations of black disadvantage, such as difficulties in accumulating wealth, poor access to public services, especially to good schools, cultural/linguistic isolation, difficulties in forging political coalitions with other groups, the spatial mismatch between where blacks live and where job growth is occurring, etc.  Chapter 4 contains a particularly illuminating discussion, suitable for undergraduate instruction, of the causes and consequences of institutionalized housing discrimination, including an important comparison of white and black Hispanics.

Massey, Douglas. Categorically Unequal: The American Stratification System. (New York: Russell Sage, 2007).

First-rate account of the fundamental causes of social stratification by race, class, and gender in the U.S.  Extensively documented discussion of the latest studies, integrating sociology, social psychology, political science, and economics.  If you need a single source covering it all, this is it.  Particularly good on blacks and Latinos, especially on the effects of mass incarceration on blacks.

Roithmayr, Daria, "Locked In Inequality: The Persistence of Discrimination," 9 Michigan Journal of Race and Law 31 (2003)
Roithmayr, Daria, "Locked in Segregation" . Virginia Journal of Social Policy and the Law, Vol. 12, No. 957 Available at SSRN: http://ssrn.com/abstract=627724

Important papers explaining the persistence of racial inequality even in the absence of continuing racial discrimination by means of a "market lock-in theory," according to which early white advantages due to discrimination get "locked in" via network effects.  The second paper uses the lock-in theory to explain the persistence of residential racial segregation in the US.

Wilson, William Julius. When Work Disappers : The World of the New Urban Poor (New York: Vintage Books, 1997).

Argues that black socioeconomic disadvantage can be attributed to the disappearance of jobs in the inner cities, where blacks disproportionately live.  An important work, especially for the way it addresses the argument that black disadvantages can be attributed to black cultural pathology.  Wilson acknowledges cultural pathology, but argues that its cause is the inability of black men to find work.

Oliver, Melvin and Shapiro, Thomas. Black Wealth, White Wealth : A New Perspective on Racial Inequality. (Routledge, 1997).

Documents the large disparities between black and white wealth.  Argues that the causal impact of wealth differences is often overlooked in analyses of racial disadvantage that focus on much smaller racial differences in income. The wealth gap is not explained by lower black savings rates, but by the fact that whites inherit more wealth, have greater opportunities to enjoy housing appreciation, and own more businesses.

Kousser, J. Morgan. Colorblind injustice : minority voting rights and the undoing of the Second Reconstruction. (Chapel Hill, N.C. : The University of North Carolina Press,1999).

Comprehensive account of disenfranchisement of African-Americans and Latinos in the U.S.  The first chapter, covering the decline of African-American enfranchisement from the end of Reconstruction until the Voting Rights Act of 1965, is a tour de force.  That chapter gives an outstanding account of the innumerable strategies used by Southern states to disenfranchise blacks without using overt racial classification.  The focus of the book is on post- Voting Rights Act litigation, emphasizing the ways in which African-American and Latino voters continue to have their voting power diluted by various state and political party actions, and how the Supreme Court has failed to protect these groups against effective political marginalization.  Of special interest for its focus on Mexican-American as well as black disenfranchisement.

Kennedy, Randall. Race, Crime, and the Law. (New York: Vintage Books, 1997).

The best comprehensive study of racial discrimination in law enforcement, on every level:  differential protection of racial groups from crime, differential prosecution, racial profiling, race-based jury selection, racism in the application of the death penalty.  A mark of intellectual integrity in the book is that the author toes no party lines:  while finding that the evidence supports the claim that there is racism in the application of the death penalty, Kennedy is more skeptical of arguments that the war on drugs--in particular, the sentencing differential between those convicted of possessing crack and powder cocaine--is racist.

Gender Inequality

*Okin, Susan. "Vulnerability by Marriage,"  Justice, Gender, and the Family, ch. 7. (New York: Basic Books, 1989).
Probably the single best brief account of the systematic socioeconomic disadvantages women face due the institution of  marriage.  Outstanding undergraduate teaching source, highly accessible and clearly written.  Focuses on the interactions of norms of marriage and a sexist division of labor within the family, the socialization of girls to expect marriage and large childrearing responsibilities, the relatively weak bargaining power of women within marriage, and women's limited options in the labor market.

Catherine MacKinnon, Sexual Harassment of Working Women (New Haven: Yale University Press, 1979)

Classic text that made the case that sexual harassment in the workplace constitutes a violation of antidiscrimination law.  Revolutionized the law of discrimination.  Analyzes sexual harassment as a means by which the subordination of women in the workplace is sustained.  Even more than 20 years later, MacKinnon's original analysis of sexual harassment remains important.

Rhode, Deborah. Speaking of Sex: the Denial of Gender Inequality. (Cambridge, Mass.: Harvard University Press, 1997).

Documents the continuing systematic disadvantages women face in U.S. society across a wide range of legal, social, and public policy settings.  Explains why people have difficulty recognizing women's disadvantages.

Valian, Virginia. Why So Slow?: the Advancement of Women. (Cambridge, Mass.: MIT Press, 1998).

The most comprehensive and empirically rigorous account to date of the relatively slow advancement of women in the professions.  Argues, on the basis of extensive data from psychology, sociology, economics, and biology, that people employ hypotheses about gender differences, embodied in "gender schemas" that create small sex differences in characteristics, behaviors, perceptions and evaluations of men and women.  These small differences accumulate, resulting in systematic underrating of women relative to men in professional settings.  An excellent reference work, with numerous highly useful citations.

Reskin, Barbara. "Sex Segregation in the Workplace," Annual Review of Sociology 19 (1993): 241-270.

Identifies gender segregation of work (the fact that most people work in jobs overwhelmingly staffed by people of the same sex) as a major cause of gender inequality in resources.  Attributes this phenomenon to several causes, including employer stereotypes and preferences, actions of male employees to deter women from entering or persisting in male-dominated jobs, and structural features of the internal job market (if only some entry-level jobs lead to promotion opportunities, then preferential selection of men to those job will entail that they dominate the upper-level jobs as well).  The big surprise of Reskin's survey of empirical research is that economic "human capital" explanations of gender segregation are empirically disconfirmed: motherhood actually increases the probability that a woman will be in or seek a nontraditional (male-dominated) job.

Goldberg, Steven. Why Men Rule : A Theory of Male Dominance (Open Court, 1993).

Argues that biological differences between the sexes make male dominance inevitable.  Rewritten update with replies to critics of Goldberg's 1973 Inevitability of Patriarchy.  Claims that sex differences in testosterone make men more aggressive, competitive, and ambitious than women, and hence motivates men to excel in any field their culture deems prestigious.

Fausto-Sterling, Anne. Myths of Gender: Biological Theories About Women and Men (New York: Basic Books, 1985.

Highly accessible critique by a biologist of biological theories (genetic, hormonal, evolutionary) that purport to explain male dominance.  Ch. 5 on hormones and aggession makes an effective pairing with Goldberg.

Arguments for Affirmative Action Policies

Definition: affirmative action policies include any policies that (a) attempt to actively dismantle institutionalized or informal cultural norms and systems of ascriptive group-based disadvantage, and the inequalities historically resulting from them, and/or that (b) attempt to promote an ideal of inclusive community, as in ideals of democracy, integration, and pluralism (multiculturalism), (c) by means that classify people according to their ascriptive identities (race, gender, ethnicity, sexual orientation, etc.) and select people for participation in institutions using these classifications as criteria.
Arguments for affirmative action policies can be divided into 4 categories.  (1) Arguments on grounds of justice defend affirmative action as a compensation or corrective for past and continuing racism/sexism. (2) Arguments on grounds of democracy view group-conscious representational devices as necessary under certain conditions for realizing a democratic society. (3) Arguments on grounds of social utility claim that affirmative action policies promote desirable goals such as better mentoring of members of disadvantaged groups or delivering professional services to the disadvantaged.  (4) Arguments on grounds of free speech and education defend affirmative action policies for the ways they create the diverse set of participants in discourse, research, and learning that is claimed necessary to promote the internal mission of educational institutions.  Discussion of this last category of arguments is deferred to the section on "Affirmative Action in Education."

Arguments from Justice

Arguments from justice are of 3 fundamental types: compensatory (backward-looking), discrimination-blocking (present-oriented), and integrative (forward-looking). . Compensatory affirmative action aims to compensate people for past discrimination and its effects.  A main effect of past discrimination is current competitive disadvantage; affirmative action gives victims a competitive advantage to compensate for this injury.  Discrimination-blocking affirmative action aims to block current discriminatory mechanisms by imposing a countervailing force in the opposite direction.  It doesn't remove the factors--prejudice, stereotypes, stigma, intergroup anxiety--that cause discrimination; it just tries to block their discriminatory effects.  Integrative affirmative action aims to dismantle the current causes of race-based disadvantage--segregation, stigmatization, discrimination--by promoting racial integration.  It thus aims for a future in which these causes no longer operate.

That affirmative action is just compensation or reparation for historical injustice

*McGary, Jr., Howard. "Justice and Reparations," Philosophical Forum 9 (1977-8): 250-263.
Argues that African-Americans are entitled to receive preferential treatment in employment and college admissions as reparations for slavery, Jim Crow, and institutional discrimination.

*Nickel, James. “Should Reparations Be to Individuals or to Groups?” Analysis 34.5 (1974): 154–60.

Very nice paper defending compensatory affirmative action from the standard charges against it:  that it is overinclusive, in compensating members of disadvantaged racial groups who have not personally suffered discrimination or its effects; underinclusive, in failing to compensate whites who have suffered discrimination; and fails to tailor the amount of benefit to the degree of harm personally suffered.  Nickel argues that these facts do not force compensatory affirmative action to a collectivist stance, according to which racial groups as a whole are the relevant units of moral entitlement.  Rather, compensatory affirmative action can retain its individualist rationale, and justify group-based awards on grounds of administrative necessity:  the costs of case-by-case adjudication would be too high.  Where the wrongdoing against a particular class has been pervasive enough, the harm so great that the chance is negligible that compensation would overshoot the aggregate damages, and individualized compensation is excessively costly or impossible, then generalized compensation to the group comes closer to the ideal of compensatory justice than a refusal to compensate at all.  This short, snappy paper is helpfully reprinted in Cahn, Steven, ed. The Affirmative Action Debate. 2nd ed. New York: Routledge, 2002.

*Duster, Troy. "Individual Fairness, Group Preferences, and the California Strategy," in Robert Post and Michael Rogin, eds. Race and Representation: Affirmative Action (New York: Zone Books, 1998), pp. 111-133.

Updates the redress argument, exploring the history of racial discrimination from the New Deal through the civil rights era.  Makes interesting parallels with the caste systems of South Africa and India, and affirmative action systems in these countries.  Considers the political motives behind the movement to prohibit affirmative action in California.  Also contains an intriguing discussion of Brandeis University's decision to practice affirmative action for men upon its discovery that exclusive reliance on criteria of academic merit would lead to a student body that was 70% female.

That affirmative action helps block current mechanisms of discrimination

Skrentny, John. The Ironies of Affirmative Action. (Chicago: University of Chicago Press, 1996).

Not an argument for affirmative action, but an indispensable historical account of how affirmative action in employment came about.  The discrimination-blocking model of affirmative action did not arise from any fundamental theories of justice, but from the experience of administrative agencies charged with enforcing antidiscrimination laws.  They found that after the passage of antidiscrimination laws, nothing changed.  Employers continued to discriminate, and members of groups suffering from discrimination made virtually no inroads into occupations and businesses that had been excluding them.  The complaint remedy built into Title VII, based on an individualized compensation model, proved ineffective in combating this discrimination.  Individual victims of discrimination had a hard time identifying themselves and proving discrimination.   When they managed to do so and filed suit, they still saw little relief.  Huge, intractable caseloads overwhelmed the ability of courts and administrative agencies such as the EEOC to handle them in a timely way.  Case-by-case litigation imposed huge costs on plaintiffs and employers alike.  To get practical, measurable, results, federal agencies started to pressure firms to adopt race-based hiring goals.

Reskin, Barbara. The Realities of Affirmative Action in Employment (Washington, D.C.: American Sociological Association, 1998).

Useful book documenting how affirmative action operates in the workplace, confirming Skrentny's account.  In general, affirmative action hiring goals are used as a counterforce against current discriminatory tendencies in the workplace, rather than as a means to require less qualified women and minorities to be hired over more qualified white men.  Vigorously contests the charge that affirmative action in employment operates in the latter way, with supporting evidence.

Stainback, Kevin, Corre Robinson, and Donald Tomaskovic-Devey. “Race and Workplace Integration.” American Behavioral Scientist 48.9 (2005): 1200–28.

Further confirmation of Skrentny's account.  Argues that passage of discrimination laws is ineffective in stopping discrimination in the absence of concerted government pressure on firms to stop discriminating and integrate their jobs.

*Ezorsky, Gertrude. Racism and Justice: the Case for Affirmative Action. (Ithaca, N.Y.: Cornell University Press, 1991).
For those who are skeptical about reparations for harms incurred a generation or more ago, Ezorsky argues that affirmative action can be justified on the ground that the harms of discrimination are current, and require compensation.

*Axelson, Diana. "With All Deliberate Delay: on Justifying Preferential Policies in Education and Employment," Philosophical Forum 9 (1977-8): 264-288.

Widely reprinted defense of affirmative action as a necessary correction for current discrimination.  Documentation of continuing institutional racism is central to the article.  Has the advantage over Ezorsky of brevity; makes an excellent companion piece to Pojman's anti-affirmative action paper (cited below), for those who wish to present students with a pair of highly representative articles on affirmative action, pro- and con.  Its disadvantage is that it is somewhat dated; for this reason the Ezorsky book, with its updated argument and evidence, may well be preferred as a teaching source.  Relevant portions of Ezorsky can be excerpted for those who don't want to assign the entire book.

*Fish, Stanley. "Reverse Racism, or How the Pot Got to Call the Kettle Black," The Atlantic, November 1993.

Lively, highly accessible defense of affirmative action, more by replying to criticisms than by direct argument.  Argues that the charge of reverse discrimination ignores history, and that objections on grounds of merit ignore the weak justification of conventional criteria of merit such as the SAT as well as current obstacles to equality of opportunity.

Yelnosky, Michael. “The Prevention Justification for Affirmative Action.” Ohio State Law Journal 64 (2003): 1385-1425.

Provides evidence that tokenism in the workplace hurts women and minorities by exposing them to discrimination.  More fully integrating jobs so women and minorities don't stand out can prevent discrimination against them.  Affirmative action is a tool for integrating jobs.

Kang, Jerry and Banaji, Mahzarin, "Fair Measures: A Behavioral Realist Revision of 'Affirmative Action'" . California Law Review, Forthcoming Available at SSRN: http://ssrn.com/abstract=873907

Defends affirmative action as a remedy for current unconscious discrimination.  Based on the latest theories of implicit bias (cognitive discrimination). The authors argue that viewing affirmative action as a correction of current bias avoids the difficulties of backward-looking and future-benefit rationales for race-conscious selection.

That affirmative action promotes integration, and thereby helps dismantle the continuing causes of race-based disadvantage

Anderson, Elizabeth. "Integration, Affirmative Action, and Strict Scrutiny," NYU Law Review, 77 (2002): 1195-1271.

This article, by the author of this website, defends racial integration as a central goal of race-based affirmative action. Racial integration of mainstream institutions is necessary both to dismantle the current barriers to opportunity suffered by disadvantaged racial groups, and to create a democratic civil society. Integration, conceived as a forward-looking remedy for de facto racial segregation and discrimination, makes better sense of the actual practice of affirmative action than backward-looking compensatory rationales, which offer restitution for past discrimination, and diversity rationales, which claim to promote nonremedial educational goals. Integrative rationales for affirmative action in higher education could also easily pass equal protection analysis, if only the point of strict scrutiny of racial classifications were understood. Unfortunately, the development of strict scrutiny as an analytical tool has been hampered by the Court's confusion over the kinds of constitutional harm threatened by state uses of racial classification. This Article sorts out these alleged harms and shows how strict scrutiny should deal with them. It shows how the narrow tailoring tests constitute powerful tools for putting many allegations of constitutional harm from race-based affirmative action to rest, and for putting the rest into perspective. It also argues that there is no constitutional or moral basis for prohibiting state uses of racial means to remedy private sector discrimination. Integrative affirmative action programs in educational contexts, which aim to remedy private sector discrimination, can therefore meet the requirements of strict scrutiny, properly interpreted.

Sabbagh, Daniel. Equality and Transparency: A Strategic Perspective on Affirmative Action in American Law (New York: Palgrave MacMillan, 2007).

Important new book elaborating on Ronald Dworkin's defense of affirmative action.  Dworkin argues that the fundamantal purpose of affirmative action is to break the correlation between race and class by bringing more blacks and Latinos into the middle class.  This will, by eliminating the basis for forming them, break down the stigmatizing stereotypes of disadvantaged racial groups as "underclass"--stereotypes that themselves constitute a grave barrier to equality of opportunity.  Sabbagh offers a powerful empirical defense of Dworkin's view, grounded in evidence in social psychology of the ways stereotypes are formed, and in economic and sociological evidence on the devastating impacts of stereotyping and consequent discrimination on the lives of blacks.  He also provides powerful critiques of compensatory and diversity-based affirmative action, and a deep analysis of the difficulties affirmative action programs face in publicizing their operations and rationales in a  non-self-undermining way.  Superbly argued, researched and documented.  Highly recommended.
Arguments from Democracy
*Post, Robert. "Introduction: After Bakke," in Robert Post and Michael Rogin, eds. Race and Representation: Affirmative Action (New York: Zone Books, 1998), pp. 13-27.

Argues that race-based affirmative action policies are necessary in college admissions because a central mission of the university is to promote a democratic culture.  This requires building the cultural capital of all citizens, so that they have the communicative and imaginative skills necessary for creating a universally inclusive, democratic discourse.  Colleges and universities "aspire to cultivate the remarkable and difficult capacity to regard oneself from the perspective of the other, which is the foundation of the critical interaction necessary for active and effective citizenship", p. 23.  Without a diverse student body (to which end affirmative action is necessary), educational institutions will be able to inculcate only limited capacities for critical interaction across group divisions.  Post stresses that this argument does not depend on the thought that identities correspond to cultures, or that individuals have fixed identities.  Part of the point of a democratic culture is to free individuals and citizens acting collectively to engage in self-definition and self-determination, without being beholden to definitions based on birth or ancestry.

Issacharoff, Samuel. "Can Affirmative Action be Defended?" Ohio State Law Journal 59 (1998): 669-695.

Important paper by a former University of Texas law professor who was recruited by his university to help defend UT's affirmative action policies in the famous Hopwood case (they lost). Argues that the democratic state has a compelling interest in training a racially integrated elite. Race-based affirmative action is the only way to enable schools to simultaneously pursue their compelling interests in meritocracy and in integrating all groups into the nation's elite. Contains important data and arguments explaining why race-neutral attempts to secure integration either fatally compromise academic standards or fail to generate significant black and Hispanic enrollment in selective schools.

Estlund, Cynthia. "Working together: the workplace, civil society, and the law," Georgetown Law Journal 89 (2000).
Estlund, Cynthia. Working Together: How Workplace Bonds Strengthen a Diverse Democracy (Oxford: Oxford UP, 2005).

Novel democratic defense of affirmative action in employment, arguing that places of employment are major sites of civil society, in which citizens interact and share their views. The democratic interest in promoting an integrated civil society--ensuring that citizens from different socially salient groups share their views with one another--supports affirmative action in the workplace.
Anderson, Elizabeth. "Racial Integration as a Compelling Interest," Constitutional Commentary 21 (2004): 101-127.

The principle and ideal developed in Brown v. Board of Education and its successor cases lie at the heart of the rationale for affirmative action in higher education.  The principle of the school desegregation cases is that racial segregation is an injustice that demands remediation.  The ideal of the school desegregation cases is that racial integration is a positive good, without which "the dream of one Nation, indivisible"  cannot be realized.  The good of integration is directly connected to realizing democracy through the promotion of a racially integrated elite, which, in virtue of its racial diversity, is more accountable to and more knowledgable of the problems of citizens from all walks of life than a homogeneous and insular elite. Both the principle and the ideal make racial integration a compelling interest.  The Supreme Court recognized these claims in Grutter v. Bollinger.  However, it failed to take full advantage of them.  It thereby failed to answer crucial questions that must be answered by policies subject to strict scrutiny.  I display the links tying Grutter to Brown, discuss the vulnerabilities of Grutter in the absence of an explicit grounding in Brown, and demonstrate how the affirmative action policy upheld in Grutter, when explicitly grounded in Brown, survives strict scrutiny.

Anderson, Elizabeth. The Imperative of Integration. (Princeton University Press, 2010).

A comprehensive discussion of the inegalitarian and undemocratic effects of de facto racial segregation, in the context of a general theory of group inequality and a review of empirical evidence, and a comprehensive defense of integration policies including affirmative action, with a review of the empirical evidence on its effects, not only on its targeted beneficiaries, but on the democratic character of civil society.  A Précis of The Imperative of Integration  may be found in Poverty and Race 20.4 (2011), published by the Poverty and Race Research Action Council.

Weisskopf, Thomas. Affirmative Action in the United States and India: A Comparative Perspective. (Routledge, 2004).

"The most important objective of . . . [affirmative action policies] is to bring about greater ethnic integration of society's elite, on the reasonable premise that society functions more efficiently, more equitably, more democratically, and more harmoniously if its professional, managerial, academic, and political elite is ethnically well integrated." (p. 244).

Arguments from Social Utility

*Dworkin, Ronald. "Affirmative Action: Does it Work?" "Affirmative Action: Is it Fair?" in Sovereign Virtue: The Theory and Practice of Equality (Cambridge, Mass: Harvard University Press, 2002).

Important defenses of affirmative action by one of the leading legal scholars in the U.S., and one of the leading advocates of affirmative action. Vigorously denies that affirmative action is unfair to whites, using a battery of arguments. Dworkin distinguishes himself by offering a predominantly forward-looking, rather than compensatory, rationale for affirmative action.

Harwood, Sterling. "Affirmative Action Is Justified: A Reply to Newton," Contemporary Philosophy (1990): 14-17.
Defense of affirmative action with replies to a major critic (cited below).  Besides appealing to considerations of compensatory justice, offers a battery of arguments for its good consequences: (1) role models; (2) diversity in education; (3) increases in the pool of applicants and hence competition; (4) replacement of diminishing marginal utility for richer whites with more utility for poorer blacks; and (5) unskewing biased and incomplete tests of merit.

That affirmative action promotes diversity, which leads to further good consequences

"Diversity" arguments for affirmative action are a species of argument from social utility.  The general idea is that group attributes can contribute to positive outcomes over and above the attributes of individuals.  The diverse whole is greater than the sum of its individual parts.  Diverse groups are more effective at solving problems than relatively homogeneous groups, even if the average individually-measured merit of the homogeneous group is higher than the average individually-based merit of the diverse group.  There are many dimensions of diversity that have this property; the trick is to show how these dimensions are linked to race, ethnic, gender, or class diversity--the typical types of diversity sought by affirmative action programs.  This section considers diversity outside education; a later section considers diversity in the context of education.

Orentlicher, David, "Diversity: A Fundamental American Principle" . Missouri Law Review, Vol. 70, p. 777, 2005 Available at SSRN: http://ssrn.com/abstract=832724

Argues that diversity in education and employment has myriad good consequences, which are recognized in the American system of law.

Hong, L., and S. Page. 2004. "Groups of diverse problem solvers can outperform groups of high-ability problem solvers." Proceedings of the National Academies of Science101:16385-89.
Page, Scott, The Difference: How the Power of Diversity Creates Better Groups, Firms, Schools, and Societies (new ed.) (Princeton: Princeton University Press, 2008)

The paper demonstrates the "diversity trumps ability theorem," a formal model showing how groups of diverse problem-solvers outperform groups of the best problem solvers.  Page's forthcoming book is a highly accessible account of formal models and empirical results demonstrating how group diversity helps organizations solve problems and make better predictions.  The connection of Page's defense of diversity is indirect, and depends upon showing links between racial, ethnic, gender or class diversity of group members is linked to various forms of desirable cognitive diversity.  Page considers this issue, offering a cautious and qualified diversity defense of affirmative action, noting that identity diversities also bring conflicts to groups.

Rudman, L. A., R. D. Ashmore, and M. L. Gary. 2001. "Unlearning automatic biases: The malleability of implicit prejudice and stereotypes." Journal of Personality & Social Psychology 81.

Diversity in education can reduce racial prejudice and stereotypes.

Sklansky, David Alan, "Not Your Father's Police Department: Making Sense of the New Demographics of Law Enforcement" . Journal of Criminal Law and Criminology, Vol. 96, Spring 2006 Available at SSRN: http://ssrn.com/abstract=885446.

Argues that affirmative action in the hiring of police broke down a pathological, insular police subculture that has made police forces more open to positive reforms and better able to establish good relationships with the community.

That affirmative action increases professional services to disadvantaged populations

Cantor, JC; Miles, EL; Baker, LC; Barker, DC. "Physician service to the underserved: implications for affirmative action in medical education," Inquiry 33(2):167-80, 1996 Summer.
(From the abstract:) Using two large physician surveys, finds that minority and women physicians are much more likely to serve minority, poor, and Medicaid populations. Weaker, but significant association exists between physician and patient socioeconomic   background. Service patterns are sustained over time and are generally consistent with physician career preferences. Argues that ending affirmative action in medicine may imperil access to care. Results do not support affirmative action based on economic disadvantage instead of race, ethnicity, and sex.

Komaromy M. Grumbach K. Drake M. Vranizan K. Lurie N. Keane D. Bindman AB. "The role of black and Hispanic physicians in providing health care for underserved populations," New England Journal of Medicine 334(20):1305-10 (1996).

(From the abstract:) Analyzed data on physicians' practice locations and the racial and ethnic makeup and socioeconomic status of communities in California in 1990. Also surveyed 718 primary care physicians from 51 California communities in 1993 to examine the relation between the physicians' race or ethnic group and the characteristics of the patients they served. Found that communities with high proportions of black and Hispanic     residents were four times as likely as others to have a shortage of physicians, regardless of community income. Black physicians practiced in areas where the percentage of black residents was nearly five times as high, on average, as in areas where other physicians practiced. Hispanic physicians practiced in areas where the percentage of Hispanic residents was twice as high as in areas where other physicians practiced. After controlling for the racial and ethnic makeup of the community, black physicians cared for significantly more black patients (absolute difference, 25 percentage points; P < 0.001) and Hispanic physicians for significantly more Hispanic patients (absolute difference, 21     percentage points; P < 0.001) than did other physicians. Black physicians cared for more patients covered by Medicaid (P = 0.001) and Hispanic physicians for more uninsured patients (P = 0.03) than did other physicians. Concludes that Black and Hispanic physicians have a unique and important role in caring for poor, black, and Hispanic patients in California. Dismantling affirmative-action programs as is currently proposed, may     threaten health care for both poor people and members of minority groups.

That affirmative action corrects for biased criteria of merit and enables the identification of the talented among the disadvantaged, thereby adding them to the labor pool

Purdy, Laura. "In Defense of Hiring Apparently Less Qualified Women," Journal of Social Philosophy 15 (1984): 26-33.
Argues that women are often perceived to be less qualified than they are, so that affirmative action corrects for a sexist perceptual bias in evaluations of merit.

Davis, Michael. "Race as Merit," Mind 92 (1983): 347-367.

Argues that under certain circumstances, being black can count as a merit or qualification for office.  Affirmative action thus does not necessarily contradict the principle of merit or constitute reverse discrimination.

Other arguments of interest

*Moskos, Charles. "Success Story: Blacks in the Military," The Atlantic, May 1986.

Argues that the U.S. military's implementation of affirmative action policies on behalf of blacks is highly successful: it has resulted in a higher proportion of blacks in management positions than in any other sector of U.S. society.  Key points of its program are rigorous enforcement of integration and antidiscrimination principles and inclusion of race-relations skills as a dimension of merit in evaluating performance of officers.  Moskos went on to write a book with John Butler, All That We Can Be : Black Leadership and Racial Integration the Army Way (New York: Basic Books, 1997), which stressed as elements contributing to the success of military integration its uncompromising and uniform application of standards of merit for promotion, combined with intensive investment in skills and training of disadvantaged blacks so they can meet these standards.

Reskin, Barbara. The Realities of Affirmative Action in Employment (Washington, D.C.: American Sociological Association, 1998), pp. 32-37.

Useful survey of the empirical research on how affirmative action works in employment settings, its impact and outcomes.  Many empirical studies are covered in this brief work, which undermine some prominent criticisms of affirmative action--e.g., that it increases workplace inefficiency by hiring less qualified people, and that it puts psychological burdens on its recipients by making them feel undeserving.

Comprehensive Defenses of Affirmative Action

*Ezorsky, Gertrude. Racism and Justice: the Case for Affirmative Action. (Ithaca, N.Y.: Cornell University Press, 1991).
Outstanding source for undergraduate teaching.  Discusses institutional racism, with examples and case studies.  Focuses on race-conscious remedies for institutional racism in employment rather than educational contexts.  Responds to a full battery of criticisms of affirmative action.  Especially effective in response to tort model objections to compensatory affirmative action (cited below). Argues that the systematic but diffuse character of institutionalized racism makes the tort model of compensation inapplicable.  For example, recruitment by advertising to current (white) employees may effectively exclude nonwhites in general even though one cannot  identify specific individuals harmed by the policy, to whom individualized compensation could be offered. Argues that individual whites who bear the burdens of affirmative action policies can and ought to be compensated without dismantling affirmative action itself.  Contains excerpts of important Supreme Court cases relating to affirmative action policies.

Clayton, Susan and Crosby, Faye. Justice, Gender, and Affirmative Action (Ann Arbor: University of Michigan Press, 1992).

Unusual argument for affirmative action.  Argues that, while there is extensive evidence of sex discrimination in employment and that women know this, they are reluctant to perceive discrimination in their own case.  Thus, remedies for discrimination that rely on women to bringing forth complaints of discrimination on their own behalf will fail to rectify discriminatory conditions.  Argues that affirmative action should be billed as a means to eliminate difficult to perceive biases against equally or more qualified women, rather than as special helping efforts for less qualified women.  Most interesting for its theory and documentation of evidence for the claim that women are motivated to not perceive discrimination against themselves.

Rosenfeld, Michael. Affirmative Action and Justice. (New Haven: Yale University Press, 1991).

Argues that affirmative action is justified both morally and  Constitutionally as a means to ensure equality of opportunity bycompensating for the present effects of past discrimination.  Groups thatsuffer no present effects of discrimination are not entitled to affirmativeaction, nor may the remedy impose costs on others beyond taking away whatever competitive advantages they enjoy in virtue of the unjust  deprivations of others' opportunities.  Includes a comprehensiveassessment of various arguments for and against affirmative action and the rival conceptions of equality underlying them.

Gutmann, Amy. "Responding to Racial Injustice," in Gutmann and Appiah, Color Conscious: the Political Morality of Race. (Princeton: Princeton University Press, 1996).

Argues that color blindness is not a fundamental moral principle.  To the extent that people are deprived of full liberties and equal opportunities because of their racial status, they are entitled to remedies to correct this  problem.  Colorblind remedies such as class-based affirmative action, while they may be justified in themselves, are not an effective substitute  for color conscious remedies, either in principle or in practice.  Defends diversity rationales for affirmative action in higher education, and denies that the principle of merit creates entitlements on the part of the best  qualified.  Merit need be recognized only to the extent that all candidates for a position are minimally qualified.

Arguments Against Affirmative Action Policies

Arguments against affirmative action policies can be divided into 2 categories.  (1) First, arguments that oppose affirmative action policies on moral principle (considerations of justice).  (2) Second, arguments that oppose these policies on grounds of their bad consequences: that they are self-defeating, harmful, or inefficient.

Arguments on Moral Principle

That affirmative action is "reverse discrimination"

*Newton, Lisa. "Reverse Discrimination as Unjustified," Ethics 83 (1973): 308-12. Reprinted in Jeffrey Olen and Vincent Barry, eds., Applying Ethics (Belmont, Cal.: Wadsworth, 1989), pp. 311-315.
Argues that affirmative action violates the fundamental principle of equal protection of the laws ("colorblindness").  Turning the tables on previously favored groups is as unjust as the original discrimination.

That affirmative action violates the principle of merit

*Walzer, Michael. Spheres of Justice. (NewYork: Basic Books, 1983), pp. 143-154.
Argues that the meaning of offices requires hiring by merit.  The most meritorious candidate has a right to the position.  Race is a bona fide qualification for a job only in special circumstances.  In practice, race-based affirmative action will open opportunities to minorities and women at the expense of the least advantaged white men.

That affirmative action violates principles of compensatory justice

These arguments adopt a tort model of compensation for unjust discrimination and object to affirmative action policies for failure to secure compensation in line with this model:  for (a) benefiting individuals who have not suffered discrimination; (b) burdening individuals who have not engaged in discrimination; (c) failing to adjust the size and kind of compensation to the specific discriminatory harms each individual suffered.
*Gross, Barry. "The Case Against Reverse Discrimination," in James Sterba, ed., Morality in Practice, 4th. ed. (Belmont, Cal.: Wadsworth, 1994), 255-260.
Argues that affirmative action violates the tort model of just compensation, with special emphasis on the unfair burdens it places on those who have not engaged in discrimination.

That the use of race as a proxy for morally relevant variables is wrong

Richard Posner, "The DeFunis Case and the Constitutionality of Preferential Treatment of Racial Minorities," in Gabriel Chin, ed. Affirmative Action and the Constitution vol.1 (New York: Garland, 1998), pp. 249-280.
Classic argument against the use of race as a proxy in affirmative action contexts.  The use of race as a proxy for morally relevant variables (e.g. being a victim of dicrimination), Posner argues, is unjust for being both under- and over-inclusive.  Moreover, any argument that justifies the use of race as a proxy on efficiency grounds (saving the costs of individualized determination of merits) for the benefit of minority groups would equally well have to accept the use of race as a proxy in ways that disadvantage those very groups--e.g., an employers' use of minority group membership as a proxy for dispositions to criminality.  Posner argues for a per se legal rule forbidding racial preferences for any reason.  For teaching purposes, this article is very usefully paired with Judge Posner's opinion, Wittmer v. Peters, annotated below, which upholds a racial preference for black prison guards in a context where race appears to be relevant to the prison's ability to advance its mission.  The two positions are not strictly contradictory (in the first, Posner is arguing from his own premises, in the second, he acts as a judge bound by Supreme Court rulings that permit racial discrimination for compelling reasons), but they do reflect a substantive change in moral temperament-- toward a more empirical, pragmatically oriented form ofoffends virtually every value held dear in free speech jurisprudence reasoning in the latter case.

Arguments from Social Utility

That affirmative action harms its intended beneficiaries.

Steele, Shelby. The Content of our Character: A New Vision of Race in America. (New York: HarperPerennial, 1991).
Argues that affirmative action stigmatizes its intended beneficiaries by implying that they are less competent and can't compete as equals with others.
*Sowell, Thomas. "The Other Side of Affirmative Action,"
Jewish World Review, June 8, 1999
Argues that the main effect of abolishing affirmative action is to send black students from highly competitive schools where they are likely to fail to less selective schools where they are more likely to succeed, because their qualifications are more closely matched to those of their peers.
*Sowell, Thomas. "Dirty Secrets about Affirmative Action," Conservative Current, October 16, 1996
Argues that affirmative action programs cause white resentment toward blacks.  Also argues that relatively higher dropout rates of minorities from colleges is due to affirmative action policies that place students at more competitive colleges than they can handle.

That affirmative action is socially divisive and leads to a Balkanized society.

*Blackstone, William. "Reverse Discrimination and Compensatory Justice," Social Theory and Practice 3 (1975).
Among its many arguments that affirmative action programs have bad consequences, argues that the logic of compensatory affirmative action can place no limit on the number of groups claiming special privileges on account of a history of discrimination.  This produces incentives for individuals to identify more with aggrieved groups than with the nation as a whole, leading to fragmentation and increasingly divisive competition among different groups for their share of preferences.

That affirmative action is economically inefficient.

*Loury, Glenn. "How to Mend Affirmative Action," The Public Interest, Spring 1997.

Argues that, besides insulting middle-class blacks for implying that they are not competent enough to compete with whites on a level playing field, affirmative action reduces blacks' incentives to achieve and accumulate human capital by systematically lowering the standards of admission and employment expected of them.  This article is far from a litany of conservative criticisms of affirmative action, however.  For Loury also argues that "notwithstanding the establishment of a legal regime of equal opportunity, historically engendered economic differences between racial groups could well persist into the indefinite future. . . the pronounced  racial disparities to be observed in American cities are particularly problematic, since they are, at least in part, the product of an unjust history. . . .  Thus I would argue. . . that the government should undertake policies to mitigate the economic marginality of those languishing in the ghettos of America. This is not a reparations argument.  When the developmental prospects of an individual depend on the circumstances of those with whom he is socially affiliated, even a minimal commitment to equality of opportunity requires such a policy. In our divided society, and given our tragic past, this implies that public efforts intended to counter the effects of historical disadvantage among blacks are not only consistent with, but indeed are required by, widely embraced democratic ideals." One of the most nuanced and interesting papers available on affirmative action; not easy to classify as either "liberal" or "conservative."

That affirmative action is an ineffecient remedy for unconscious discrimination

Wax, Amy.  "Discrimination as Accident." 74 Indiana Law Journal 1129 (1999).

While conceding that unconscious racial bias is pervasive, argues that affirmative action is an inefficient remedy for it, because employers don't know how to correct for their unconscious biases, affirmative action won't help, and employees are the least cost avoiders of discrimination.  Hence, the costs of unconscious bias should be left where they fall.

That the diversity defense of affirmative action cannot bear the weight put on it

Jim Chen, "Diversity and Damnation," 43 UCLA L. REV. 1839 (1996)

Eugene Volokh, "Diversity, Race as Proxy, and Religion as Proxy," 43 UCLA L. REV. 2059 (1996).

Graglia, Lino. "Grutter and Gratz: Race Preference to Increase Racial Representation Held Patently Unconstitutional� Unless Done Subtly Enough in the Name of Pursuing Diversity� 78 TUL. L. REV. 2037 (2004).

Three blistering critiques of the diversity defense of affirmative action.  Chen argues that the diversity defense could only make sense in conjunction with First Amendment interests in a diversity of viewpoints, but that it "offends virtually every value held dear in free speech jurisprudence" by equating viewpoints with looks.  Volokh argues that the same diversity rationale for race would also apply to religion; since we reject discrimination on the basis of religion for diversity reasons, we should similarly reject discrimination on the basis of race for diversity reasons.   In  addition, the use of race as a proxy for attitudes is generally unjustified under the 14th Amendment, since it depends on racial stereotypes.  Graglia condemns diversity-based defenses of affirmative action in higher education for being dishonest and lacking in empirical support, as well as unconstitutional race discrimination.

That diversity is better left to private voluntary efforts than government management

Schuck, Peter. Diversity in America : Keeping Government at a Safe Distance (Belknap Press, 2003).

Comprehensive and nuanced account of the history and current conceptualization and practice of diversity in the U.S. Argues that, for the most part, the worthwhile forms of diversity can and will be achieved through private sector voluntary efforts. Criticizes most but not all state-sponsored diversity efforts, especially by courts, as muddled, meddling failures.

Comprehensive Critiques of Affirmative Action

Glazer, Nathan.  Affirmative Discrimination.  (New York: Basic books, 1975).
Classic early statement against affirmative action.  Argues that, with the success of the civil rights movement in changing racial attitudes and legally abolishing discrimination, African-Americans can be expected to follow the pattern of assimmilation and upward mobility of white immigrant groups, without needing special preferences.  Adoption of affirmative action preferences therefore represents a gratuitous assault on the colorblind principle, threatening a return to the Balkanizing identity politics of the past.  Glazer has since changed his assessment of affirmative action, in light of the failure of his optimistic prediction to be realized after 20 years.  In We are All Multiculturalists Now (Cambridge, Mass.: Harvard University Press, 1997), Glazer reluctantly concedes the necessity for affirmative action because persistent segregation and discrimination continue to prevent blacks from realizing equal opportunity.  Glazer found particularly persuasive the evidence on segregation presented by Massey and Denton in American Apartheid, annotated above.  Pairing Glazer's earlier and later views would make an excellent instructional tool--a rare and vivid demonstration of how empirical evidence can actually change someone's mind on a passionately held moral issue. 

Sowell, Thomas. Preferential Policies: An International Perspective. (New York: William Morrow, 1990).

Argues that affirmative action policies promote incompetence and resentment against its intended beneficiaries.  Cites data suggesting that blacks admitted to elite institutions under affirmative action policies cannot compete effectively there and would do better if they attended the lower-ranked schools for which they are qualified.  Unusual for taking an international perspective on the issues.

Cohen, Carl. Naked Racial Preference. (Boston: Madison Books, 1995).
Cohen, Carl and Sterba, James. Affirmative Action and Racial Preference : A Debate (Oxford: Oxford University Press, 2003).

Cohen offers a comprehensive critique of affirmative action as unjust "reverse racism." The author is the University of Michigan professor most active in opposing the U-M's own affirmative action policies.  The second, more recent volume updates Cohen's argument, and is paired with a defense of affirmative action by James Sterba.

Thernstrom, Stephan and Thernstrom, Abigail. America in Black and White: One Nation, Indivisible (Cambridge, Mass.: Harvard University Press, 1998)

Comprehensive review of black progress in the U.S.  Argues that white racism is largely a thing of the past, and that blacks have made great progress since the civil rights era, largely due to market forces.  Remaining inequalities are due to problems such as low black educational attainment, high crime rates, and black family structures, that affirmative action programs cannot remediate.  Comprehensive critique of all race-based programs, not just affirmative action in employment and admissions, but such measures as race-conscious election districting.

*Pojman, Louis. "The Moral Status of Affirmative Action," in Julie McDonald, ed., Contemporary Moral Issues in a Just Society (Belmont, CA: Wadsworth, 1998), pp. 297-315.

Comprehensive critique of affirmative action programs.  Contains useful history of affirmative action in the U.S., and summarizes, with replies, arguments in favor of these policies.  Pojman's critique is one of the most widely reproduced in textbooks and anthologies, and is probably the single most useful paper for undergraduate teaching that offers an overall review of the arguments against affirmative action. See also Pojman's "The Case against Affirmative Action," International Journal for Applied Philosophy, available online.

Affirmative Action in Education

Affirmative action in educational settings raises special issues not necessarily encountered in other areas, such as employment.  (1) The rationale for AA includes not just compensation for past and continuing discrimination, but consideration of the educational value of diversity. (2) The Supreme Court has specifically recognized universities as having a First Amendment right to free speech and hence academic freedom, under which admissions and hiring policies are included.  Thus, questions about the legality of affirmative action policies in education must be considered in view of First Amendment as well as the 14th Amendment and civil rights laws. (3) Merit-based arguments against affirmative action are weaker, the earlier in the "pipeline" affirmative action policies are applied. (a) Few students have a choice over schools; their educational attainment is largely a function of the resources the state has chosen to devote to them. This in turn is a function of place of residence, which, given pervasive housing segregation, is a function of race.  Conventional criteria of "merit" for admissions therefore do not measure purely individual factors (talent and determination), but also reflect many dimensions of class, race, and gender privilege.  (b) The best rationale for awarding opportunities on the basis of merit concern the efficiency advantages of assigning the most talented people to perform productive tasks.  But students are not employees, they are in school to learn more than to produce.  (c) Students do produce an educational environment, however, and so are selected in part for what they can contribute to that environment and hence to the education of their fellow students.  However, at this point diversity itself has been defended as a dimension of merit--that is, being able to bring to the educational environment various perspective shaped (not defined) by having lived in substantially different circumstances from the majority of students constitutes part of  students' merit for admissions purposes.

The Educational Value of Diversity (Multiculturalism) in the Curriculum

Nussbaum, Martha. Cultivating humanity : a classical defense of reform in liberal education (Cambridge, Mass.: Harvard University Press, 1997).
Articulates a rationale for multiculturalism, including educational reforms (expanding the canon, relatively new disciplines of women's studies, African-American studies, gay and lesbian studies, etc.) in terms of an ideal of the liberal arts as advancing a cosmopolitan community, in which people can learn from others in all walks of life.  This book represents a comprehensive reply, based on personal observation of multicultural education in a vast range of colleges and universities, to such critics of "political correctness" as Dinesh D'Souza and Allan Bloom (see below).

Levine, Lawrence. The Opening of the American Mind. (Boston: Beacon Press, 1996).

 This book by a distinguished U.S. historian, examines the history of the U.S. college curriculum, demonstrating its constant flux in canons, ideals, and presuppositions.  Defends multicultural educational reforms against critics, exposing the critics' misrepresentations of the reforms and what preceded them.

*Lauter, Paul. "Race and Gender in the Shaping of the American Literary Canon: a Case Study from the Twenties, " in Canons and Contexts. (Oxford: Oxford University Press, 1991).

 Case study of the roles of sexism and racism in eliminating from the canon of American literature works written by women and African-Americans.

D'Souza, Dinesh. Illiberal Education : The Politics of Race and Sex on Campus. (New York: Vintage, 1992).

Attacks multiculturalism and affirmative action in universities for promoting racial divisiveness and undermining freedom of thought.  One of the leading works in the movement against "political correctness" in higher education.

Bloom, Allan. The Closing of the American Mind. (New York : Simon and Schuster, 1987).

One of the leading works in the movement against "political correctness" in higher education.  Argues that recent educational trends in universities are destroying the liberal arts and undermining Western civilization.

The Educational Value of Diversity among Students and Educators

*Brown, James Robert. "Affirmative Action and Epistemology," in Robert Post and Michael Rogin, eds. Race and Representation: Affirmative Action (New York: Zone Books, 1998), pp. 333-337.
Argues that all inquirers are affected by unconscious biases; in many areas of inquiry these biases are influenced by the social position (race, gender, etc.) of the inquirer.  Research communities that lack diversity along these dimensions are ill-equipped to detect their biases.  Diversity of researchers improves theories by opening them up to a more rigorous critical review from more perspectives.

*Butler, Judith. "An Affirmative View,"  in Robert Post and Michael Rogin, eds. Race and Representation: Affirmative Action (New York: Zone Books, 1998), pp. 155-173.

Criticizes the Regents of the University of California decision to abolish consideration of race, ethnicity, and gender in admissions and hiring for its failure to consider and respond to the educational value of diversity.  Argues that diversity is not to be defended on the assumption that the meaning or value of an individual's contribution to inquiry can be reduced to or predicted by their social identity.  Rather, discourse in a diverse community is valuable in providing inquirers with opportunities to change and reinterpret the significance of their identities through exchange with others who are differently positioned.  Surprisingly accessible statement by a postmodernist author not usually so easy to read.

*McGowen, Miranda Oshige. "Diversity of What?"  in Robert Post and Michael Rogin, eds. Race and Representation: Affirmative Action (New York: Zone Books, 1998), pp. 237-250.

Argues that the racial and ethnic categories universities use to promote diversity fail to reflect educationally relevant dimensions of  diversity.  They are not good proxies for diverse points of view; thus affirmative action policies as actually practiced by colleges and universities can't be easily justified with the diversity rationale.  Makes an excellent skeptical companion piece to the others in this section, for instructors interested in focusing on diversity arguments for affirmative action.

Blum, Lawrence. "Philosophy and the Values of a Multicultural Community," Teaching Philosophy (1991): 127-134.

Argues that there are three distinct values, each
essential to a multicultural community such as a college: (1) opposition to racism (opposition to racial domination, and to attitudes that support it by denying the moral equality of all human beings), (2) multiculturalism (respect for one's own and others' cultures), and (3) community (a sense of connection across racial and ethnic lines). Argues that these are distinct, though seldom sufficiently distinguished, values, and explores their convergences and possible divergences.

"The Compelling Need for Diversity in Higher Education."  Expert Testimony of Witnesses for Defendants in Gratz v. Bollinger, Grutter v. Bollinger

The most comprehensive empirically based defense of the diversity rationale for affirmative action available on the Web, developed by the University of Michigan for its defense in two legal challenges to its policies.  Effectively illustrates both the strengths and the weaknesses of the diversity defense--judge for yourself whether the evidence of expert witness Professor Pat Gurin changes your mind on the merits of diversity.  Sympathizers note that she finds positive effects of diversity on an extraordinarily wide range of educational outcomes.  Skeptics observe that the outcomes are measured by students' self-report, raising the suspicion that students who are liberal about race relations just think more highly of themselves.  Don't overlook the important testimony of the other expert witnesses.

Affirmative Action in Faculty Hiring

Cahn, Steven (ed.). Affirmative Action and the University. (Philadelphia: Temple University Press, 1993).

The best single and comprehensive source of arguments for and against affirmative action in faculty hiring currently available.

Outcome Studies on Affirmative Action

Positive Studies

Bowen, William and Bok, Derek. The Shape of the River: Long-Term Consequences of Considering Race in College and University Admissions. (Princeton: Princeton University Press, 1998).
By far the most sophisticated, empirically grounded, comprehensive study of the outcomes of affirmative action programs in American universities.  Based on a huge database, unlike most other commentaries, which are based mainly on anecdotal evidence.  Packed with charts and tables.  The definitive work thus far on outcome assessment of affirmative action in higher education.  Must be consulted by anyone offering an opinion on the matter.  The most important finding is that the "mismatch" hypothesis is false:  black students who attend a more selective school do better (with respect to graduation rates, attainment of advanced degrees, income, satisfaction with college experience) than their academically equivalent peers who attend a less selective school (i.e., the schools they would attend if affirmative action were abolished). This study is good on educational outcomes, and post-graduation outcomes for the direct beneficiaries of affirmative action, but considerably weaker on impacts of affirmative action on blacks beyond those directly targeted. Too busy to read this giant book? Read my review, "From Normative to Empirical Sociology in the Affirmative Action Debate: Bowen and Bok’s The Shape of the River," Journal of Legal Education 50 (2000): 284-305. Available on request--see my contact information at the bottom of this webpage. For a harsher review, see Stephan Thernstrom and Abigail Thernstrom, "Reflections on The Shape of the River"UCLA Law Review 46 (1999): 1583-1631.

Zweigenhaft, Richard and Domhoff, G. William. Blacks in the White Establishment? A Study of Race and Class in America. (New Haven: Yale University Press, 1991).

Study of the challenges faced by graduates of A Better Chance, a program that has brought black students from economically impoverished backgrounds to elite U.S. prep schools.  Among its many findings:  although ABC students started off behind their peers in academic preparation, they graduated at roughly the same rates and by their senior year had a median class rank comparable to their non-ABC peers.  Pages 145-158 contain an excellent discussion, with international comparisons, of how stigmatization can lead to "oppositional identity" in which stigmatized students are disaffected from school.  Elite prep schools, by contrast, give their students a "superiority complex" that aids their academic achievement.

 Davidson, RC and Lewis, EL. "Affirmative action and other special consideration admissions at the University of California, Davis, School of Medicine," Journal of the American Medical Association(JAMA) 278(14):1153-8, 1997, Oct 8. [published erratum appears in JAMA 1998 Feb 18;279(7): 510].

(From the abstract:) Twenty-year, retrospective, matched-cohort study of affirmative action admissions at U-C Davis Medical School.  Found that graduation and failure rates of special admissions and regular admissions students were comparable, as were career patterns after graduation.  Concludes that criteria other than undergraduate grade point average and Medical College Admission Test scores can be used in predicting success in  medical school. An admissions process that allows for ethnicity and other special characteristics to be used heavily in admission decisions yields powerful effects on the diversity of the student population and shows no evidence of diluting the quality of the graduates.

Cantor, JC; Miles, EL; Baker, LC; Barker, DC. "Physician service to the underserved: implications for affirmative action in medical education," Inquiry 33(2):167-80, 1996 Summer.

(From the abstract:) Using two large physician surveys, finds that minority and women physicians are much more likely to serve minority, poor, and Medicaid populations. Weaker, but significant association exists between physician and patient socioeconomic   background. Service patterns are sustained over time and are generally consistent with physician career preferences. Argues that ending affirmative action in medicine may imperil access to care. Results do not support affirmative action based on economic disadvantage instead of race, ethnicity, and sex.

Komaromy M. Grumbach K. Drake M. Vranizan K. Lurie N. Keane D. Bindman AB. "The role of black and Hispanic physicians in providing health care for underserved populations,"
New England Journal of Medicine 334(20):1305-10, 1996
May 16.

(From the abstract:) Analyzed data on physicians' practice locations and the racial and ethnic makeup and socioeconomic status of communities in California in 1990. Also surveyed 718 primary care physicians from 51 California communities in 1993 to examine the relation between the physicians' race or ethnic group and the characteristics of the patients they served. Found that communities with high proportions of black and Hispanic     residents were four times as likely as others to have a shortage of physicians, regardless of community income. Black physicians practiced in areas where the percentage of black residents was nearly five times as high, on average, as in areas where other physicians practiced. Hispanic physicians practiced in areas where the percentage of Hispanic residents was twice as high as in areas where other physicians practiced. After controlling for the racial and ethnic makeup of the community, black physicians cared for significantly more black patients (absolute difference, 25 percentage points; P < 0.001) and Hispanic physicians for significantly more Hispanic patients (absolute difference, 21     percentage points; P < 0.001) than did other physicians. Black physicians cared for more patients covered by Medicaid (P = 0.001) and Hispanic physicians for more uninsured patients (P = 0.03) than did other physicians. Concludes that Black and Hispanic physicians have a unique and important role in caring for poor, black, and Hispanic patients in California. Dismantling affirmative-action programs as is currently proposed, may threaten health care for both poor people and members of minority groups.

Holzer, Harry and Neumark, David, "Assessing Affirmative Action" National Bureau of Economic Research Working Paper No.W7323

The full text of this paper can be obtained through the link, which is on the Social Science Research Network website. This literature review of economic studies of affirmative action finds no significant evidence for the claim that affirmative action impairs efficiency.

Holzer, Harry and Neumark, David, "Equal Employment Opportunity and Affirmative Action," Public Policy Institute of California Working Paper No. 2002-07.

Finds that antidiscrimination laws improved the economic condition of blacks in the 1960s, and that affirmative action has had a positive impact on employment for blacks and women. Finds no evidence that affirmative action has resulted in firms hiring women with lower qualifications than men. Does find evidence that affirmative action has resulted in firms hiring racial minorities with lower qualifications than other employees, but fails only weak evidence that this lag in entering qualifications had an impact on minority performance on the job. Finds that minority-owned businesses that receive contracts under affirmative action have weaker performance, but that these deficiencies can be corrected through technical assistance and better vetting.

Holzer, Harry and Neumark, David, "What Does Affirmative Action Do?"Industrial and Labor Relations Review, 53 (2000): 240-271.

The full text of this paper, when it was a working paper, can be obtained through this link on the SSRN website. (From the abstract) Authors "find that Affirmative Action increases the number of recruitment and screening practices used by employers, raises employers' willingness to hire stigmatized applicants, increases the number of minority or female applicants as well as employees, and increases employers' tendencies to provide training and formally evaluate employees. When Affirmative Action is used in recruiting, it generally does not lead to lower credentials or performance of women and minorities hired. When it is also used in hiring, it yields minority employees whose credentials are somewhat weaker, though performance generally is not. Overall, the more intensive search, evaluation, and training that accompany Affirmative Action appear to offset any tendencies of the policy to lead to hiring of less-qualified or less-productive women and minorities.

Research Supporting the University Affirmative Action Programs, University of Michigan

Website with full-text links to empirical research supporting affirmative action in higher education, including many expert reports and amicus briefs filed on behalf of the University of Michigan in its lawsuits, Grutter v. Bollinger and Gratz v. Bollinger. See, in particular, the defense of Patricia Gurin's controversial research in "The Compelling Need for Diversity in Higher Education.", showing a wide range of positive educational outcomes of affirmative action, on the basis of student surveys.

Taylor, Bron. Affirmative Action at Work: Law, Politics, and Ethics (Pittsburgh: University of Pittsburgh Press, 1991).

Outcome study of an aggressive affirmative action policy practiced at a large workplace. Responds to a range of objections to affirmative action. Argues on the basis of evidence gathered, including interviews with employees, that affirmative action is morally permissible.

Weisskopf, Thomas. Affirmative Action in the United States and India: A Comparative Perspective. (Routledge, 2004).

Unusual international comparison study offers extensive empirical evidence that affirmative action in both the US and India has helped create a more racially integrated elite, and, in virtue of this fact, a more accountable and legitimate elite. At the same time, the policies have imposed some ethnic conflict, especially in India.

Mixed Studies

Krieger, Linda, "Civil Rights Perestroika: Intergroup Relations After Affirmative Action" California Law Review 86 (1998): 1251-1333.

Surveys the literature, finding theoretical and empirical support for the claim that affirmative action exacerbates racial tensions. However, the paper also provides theoretical and empirical support for the claim that, absent affirmative action, conventional tools to combat discrimination, such as individual claims under antidiscrimination laws, are not up to the task of preventing or remedying discrimination and racial segregation.

Bailey, Elizabeth; Rask, Kevin., "Are We Role Models?: A Multinomial Mixed Logit Model of Major Choice in an Undergraduate Institution," Colgate Economics Discussion Paper #95-07.

Finds evidence that minority and female faculty do exert a role model effect on minority and female students, attracting them to majors where the former have greater representation.

Negative Studies

Lott, John R., "Does a Helping Hand Put Others At Risk?: Affirmative Action, Police Departments, and Crime" Economic Inquiry 38 (2000): 239-77.

The full text of this paper can be obtained through this link on the SSRN website. Finds that crime rates rise when affirmative action leads to the hiring of less qualified minority police officers. Does not find such a relationship with the hiring of female police officers.

Rothman, Stanley; Lipset, Seymour Martin & Nevitte, Neil, "Racial Diversity Reconsidered," The Public Interest Spring 2003.

The full text of this paper is available at The Public Interest website, although you must search for it once you are there. Finds, on the basis of survey research techniques designed to avoid "politically correct" responses, that higher racial diversity on campuses is correlated with higher reports of racial discrimination, lower perceived work ethic among the students, and less satisfaction with the educational experience among students. Finds strong opposition to affirmative action among students. Concludes that racial diversity on college campuses does not have the benefits claimed on its behalf. This article provides a useful counterpoint to Patricia Gurin's expert report on behalf of the University of Michigan, cited above, which reports, on the basis of survey data, wide-ranging educational benefits of racial diversity. A more technical presentation of the same results was published in the International Journal of Public Opinion Research.

Cole, Stephen and Barber, Elinor, Increasing Faculty Diversity: The Occupational Choices of High-Achieving Minority Students (Harvard University Press, 2003).

Empirical study finding that minority students are deterred from pursuing Ph.D.'s, and hence pursuing academic jobs, because affirmative action at the undergraduate level channels these students into more selective schools, where they get lower grades relative to their white peers. This depresses their motivation to pursue graduate study, perpetuating the "pipeline problem" that continues to keep faculty diversity below the level of student diversity in higher education. Also rejects the role model theory that minority professors encourage minority students to pursue academic careers. Cole's findings are somewhat in tension with those of Bowen and Bok, who find that affirmative action produces a big boost in the number of black students who pursue advanced degrees, notwithstanding its negative impact on black students' GPAs. It is possible that this gain is generated by black students' pursuit of professional degrees (M.D., J.D., M.B.A.) which offsets a reduction in pursuit of Ph.D.'s. Cole's study is focused on the supply side of minority aspirants to academic positions. It is an open question whether the positive supply effect of moving students to less selective colleges would be offset by the depressed demand for these students on the part of Ph.D. programs. Such programs tend to favor applicants from more selective schools, both because the more selective schools tend to offer more rigorous curricula, more research opportunities with faculty, and hence better preparation for academic careers, and because authors of letters of recommendation from more selective schools are more likely to be known to, and taken seriously by, faculty on Ph.D. program admissions committees.

Sander, Richard. “A Systemic Analysis of Affirmative Action in American Law Schools.” Stanford Law Review 57 (2004).

Argues that affirmative action hurts black law students by placing them in highly-ranked law schools where they can't compete.  Major recent argument for the "mismatch" hypothesis.  Sander's article is vigorously criticized by numerous scholars in the Stanford Law Review (2005).

Sowell, Thomas. Preferential Policies: an International Perspective. (New York : William Morrow and Co., 1990)
Sowell, Thomas. Affirmative Action around the World: An Empirical Study (New Haven: Yale University Press, 2005).

Two of the few studies of affirmative action in an international context, by a major critic of affirmative action. The evidence is more of an anecdotal than rigorous nature, however.

Alternatives to Affirmative Action

Many opponents of race-based affirmative action believe its abolition will lead to a color-blind, meritocratic society in which nonwhites face no significant race-based disadvantages. Others acknowledge that this is not true and that without replacing it with some alternative, U.S. society will be more segregated and blacks especially more marginalized than before.  Hence, many seek an alternative.  Class-based affirmative action appeals to those who believe that justice must be color-blind but not class-blind, that blacks suffer from no unjust disadvantages that can't be traced to class, and who are not particularly moved by merit-based objections to affirmative action.  Other alternatives are considered below.

Class-based affirmative action

*Kahlenberg, Richard. "Class, not Race," New Republic, April 3, 1995, pp. 21-26.
The single best very brief defense of class-based affirmative action available.

*Kinsley, Michael. "The Spoils of Victimhood," New Yorker, March 27, 1995, p. 66ff.

Brief attack on class-based affirmative action.

Kahlenberg, Richard. The remedy : class, race, and affirmative action. New York: Basic Books, 1996.

The most authoritative, comprehensive defense of class-based affirmative action as a substitute for race-based affirmative action.  Argues that affirmative action is justified only as a remedy for poverty, not cultural differences between racial groups.  Compensatory arguments for race-based affirmative action wrongly attribute all black disadvantage to discrimination, when black cultural differences may be to blame.  Diversity rationales for race-based affirmative action may backfire by increasing stigmatization rather than promoting tolerance.

Carnevale, Anthonyh P. and Rose, Stephen J., "Socioeconomic Status, Race/Ethnicity and Selective College Admissions," The Century Foundation, New York, March 2003.

Important empirical study demonstrating the feasibility of aggressive class-based affirmative action policies that would yield selective student bodies in which 38% of students are below the median socioeconomic status. Highly significant is their simulation of graduation outcomes, were class-based affirmative action to be aggressively practiced by selective schools. Contrary to claims that students in the bottom 40% of the economic ladder would not be able to handle the challenging curricula at selective schools, due to the weakness of their high schools, C&R's simulation predicts that aggregate graduation rates would actually increase if schools admitted lower-class students who were otherwise academically qualified, instead of using the mismash of antimeritocratic criteria (e.g., legacy admissions) currently in use.

Malamud, Deborah. "Affirmative Action, Diversity, and the Black Middle Class," University of Colorado Law Review 68 (1997): 939-999.

Argues that diversity defenses of affirmative action in education cannot stand alone; they must be supplemented by justice rationales.  Class-based affirmative action cannot compensate for the distinctively race-based disadvantages suffered by the black middle class.  Meaningful levels of racial integration therefore cannot be achieved without race-based affirmative action.  Especially valuable for extensive documentation of the socioeconomic disadvantages the black middle class suffers because of race.

Malamud, Deborah. "Assessing Class-Based Affirmative Action," Journal of Legal Education 47 (1997): 452-471.

Argues that class-based affirmative action cannot significantly increase the opportunities of the poor without significantly reducing admissions standards at selective schools. In practice, it will advance the prospects of some lower middle class applicants over others.  Discusses difficulties of constructing color-blind criteria of class-based disadvantage within the middle class.  In practice, class-based affirmative action will expand opportunties to lower middle class whites at the expense of middle class blacks, who suffer from race-specific socioeconomic disadvantages not recognized by the policy.  This will have the effect of enabling whites to more effectively transmit their middle-class status to their children, while depriving blacks in the middle-class from doing the same.  Race-based affirmative action offers the best prospects of enabling blacks to establish an intergenerational presence in the middle class, at least cost to the other goals of higher education.

Geographical Affirmative Action ("Percentage Plans")

Montejano, David. "Maintaining Diversity at the University of Texas,"  in Robert Post and Michael Rogin, eds. Race and Representation: Affirmative Action (New York: Zone Books, 1998), pp. 359-369.

In response to the Hopwood decision, which prohibited race-based affirmative action at the University of Texas, the Texas state legislature acted to limit the threat it posed to resegregate higher education  by mandating that the top 10% of each high school class be automatically admitted to the Texas university of their choice.  This geographical standard prevents segregation by taking advantage of the fact that high schools themselves are highly segregated, due to the concentration of black and Hispanic populations in particular residential districts.  This paper offers a preliminary assessment of this proposal.

 Fryer, Roland G., Loury, Glenn C. and Yuret, Tolga, "Color-Blind Affirmative Action" (November 2003). NBER Working Paper No. W10103. Available at SSRN: http://ssrn.com/abstract=468790

Finds that color-blind affirmative action policies compromise merit far more than color-conscious affirmative action policies. "The short-run efficiency losses of implementing color-blind affirmative action (in our sample) are four to five times as high as color-conscious affirmative action."

Horn, Catherine and Flores, Stella. "Percent Plans in College Admissions:
A Comparative Analysis of Three States' Experiences". Harvard Civil Rights Project

One of the now numerous studies documenting that "top x% plans", requiring the admission of the top x% of any in-state high school graduates to state university, have failed to live up to their billing of improving racial diversity on selective college campuses. To the extent that they have achieved any diversity, they are reliant on underlying race-conscious policies, such as vigorous recruitment from minority high schools and massively increased financial aid. Cites to other studies making similar findings can be found at the University of Michigan's website on Research Supporting Affirmative Action in Higher Education.

Focus on the "truly disadvantaged"

Wilson, William Julius. The Truly Disadvantaged. (Chicago: University of Chicago Press, 1987.
Wilson, William Julius. When Work Disappears. (New York: Vintage Books, 1997).

Criticizes affirmative action for failing to help the most disadvantaged and for increasing racial divisiveness, thus posing obstacles to cross-racial coalition building. Massive public works projects, designed to bring jobs to the inner cities where jobs have disappeared, are a better, color-blind alternative.

Owen Fiss, A Way Out: America's Ghettos and the Legacy of Racism (Princeton: Princeton University Press, 2003).

While not specifically criticizing affirmative action, this book takes up Wilson's challenge to focus on the most disadvantaged, those who are not directly benefitted by affirmative action. Defends aggressive state measures to enable poor blacks to move out of the inner cities (which provide poor job and educational opportunities, and prevent their residents from accumulating savings or avoiding victimization from crime) and achieve racial integration in more successful suburban neighborhoods. Contains numerous critical responses to his proposal and replies by Fiss. See especially the sharply worded critique by Jim Sleeper, author of Liberal Racism. A fresh, provocative book.

Closing the "Merit Gap" between advantaged and disadvantaged groups

Jencks, Christopher and Phillips, Meredith. "America's Next Achievement Test: Closing the Black-White Test Score Gap," The American Prospect 40 (September-October 1998): 44-53.
Argues that, in light of the rollback of affirmative action programs in higher education, the best chance for advancing black prospects is to close the gap between blacks and whites in performance on standardized tests.  Provides data on the persistent but declining gap, and suggests strategies for closing the gap.

Jencks, Christopher and Phillips, Meredith, eds. The Black-White Test Score Gap (Washington, D.C.: Brookings, 1998).

Highly important collection of articles on the causes and potential remedies of the black-white educational achievement gap.

More Vigorous Enforcement of Individualized Antidiscrimination Law

Eastland, Terry. Ending Affirmative Action: the Case for Colorblind Justice (New York: Basic Books, 1996).
Reviews a wide range of affirmative action programs in employment, college admissions, and contracting, arguing that they all betray the moral requirement that all be treated equally, regardless of race. Argues that discrimination should be addressed only by individualized remedies, and that more resources should be devoted to enforcement of antidiscrimination law if racism persists after the abolition of affirmative action.


Steele, Shelby. The Content of Our Character : A New Vision of Race in America. (New York: HarperPerennial, 1991).
Argues that black reliance on affirmative action reinforces black dependency on whites.  Blacks should shed their victim's mentality and civil rights demands for white help and focus instead on self-improvement and self-help.  A contemporary revival of Booker T. Washington's philosophy.

Loury, Glenn. One by One from the Inside Out : Essays and Reviews on Race and Responsibility in America (New York: Free Press, 1995).

Shares with Steele's work the interest in reviving Booker T. Washington's individualist self-help and personal responsibility philosophy.  Although he criticizes affirmative action policies, Loury denies that his individualistic philosophy absolves the wider U.S. society of responsibility for ensuring equal opportunity for poor blacks.

Letting Free Markets Solve the Problem

Epstein, Richard. Forbidden Grounds : The Case Against Employment Discrimination Laws. (Cambridge, Mass.: Harvard University Press, 1992).

Argues that antidiscrimination laws are inefficient and fail to help their intended beneficiaries: by raising the cost of firing members of  protected groups, they increase the incentives of employers to not hire them in the first place.  Abolition of such laws need not lead to disaster for blacks: much discrimination is inefficient from the employer's perspective and, in any event, abolition would also eliminate legal challenges to affirmative action policies.

Sunstein, Cass. "Why Markets Don't Stop Discrimination," in Free Markets and Social Justice. (Oxford: Oxford University Press, 1997), pp. 151-166.

Concise, accessible account of several reasons why markets don't stop discrimination: it can be an efficient response to employee and customer "tastes" for discrimination; statistical discrimination can be a rational (self-interested) alternative to bearing the costs of obtaining individualized information about job candidates; such discrimination discourages victims from investing in human capital (they get lower returns), thereby creating a vicious circle; discrimination motivates and provides spurious evidence for blaming the victim.

 Informal Promotion of Integration, Stressing Intermarriage and Cross-Racial Friendship

Lind, Michael. The Next American Nation; The New Nationalism and the Fourth American Revolution (New York: Free Press, 1995).

Advocates a color-blind society, with proposals to reduce class-based and affiliation barriers to racial integration.  Among these proposals is the promotion of interracial marriages and adoptions.

DeMott, Benjamin. The Trouble with Friendship: Why Americans Can't Think Straight about Race. (New Haven: Yale University Press, 1995).

Criticizes the tendency of white Americans to think that racial divisions can be overcome by interracial frienship and personal rejection of racial antipathy.  Argues that the primary obstacles blacks face concern the dismantling of a caste system, not making friends with whites.

Legal Cases about Affirmative Action

This section contains five parts:  (1) Important civil rights cases not specifically about affirmative action, but that articulate principles and arguments often cited for or against affirmative action; (2) affirmative action, phase I: the key Supreme Court cases that established the constitutionality of race-conscious affirmative action; (3) affirmative action, phase II:  various affirmative action cases, including those based on gender and veterans' status, and others consolidating and extending the scope of permissible affirmative action; (4) affirmative action, phase III: the key Supreme Court cases that have cut back on the scope of permissible affirmative action policies or that articulate principles constraining their shape; (5) affirmative action, phase IV: key cases regarding affirmative action that have been decided or are being considered by federal courts below the Supreme Court.  Decisions by Federal Appeals Courts are not binding outside the circuits in which they were heard.

Key Civil Rights Cases Relevant to Affirmative Action

Plessy v. Ferguson, 163 U.S. 537 (1896)

Notorious decision upholding the constitutionality, under the 13th and 14th Amendments, of state laws requiring racially segregated public facilities (in this case, of Louisiana's railroad cars), provided they are "separate but equal".  This decision was overruled by Brown v. Board of Education, 347 U.S. 483 (1954).  The case remains relevant because of  Justice Harlan's vigorous dissent, now taken by many to represent the correct Constitutional analysis, not only of this case, but of the constitutional status of race-conscious laws.  Harlan's declaration that "our Constitution is color-blind" (559) is often cited in opposition to the constitutionality of affirmative action programs.  Less noticed is his declaration, immediately prededing this sentence, that "in view of the Constitution. . . [t]here is no caste here," which suggests different implications (if color-conscious remedies are needed to dismantle caste barriers).  Another interesting feature of the case is its stress on analysis of the 13th Amendment (prohibiting slavery) rather than the 14th Amendment (requiring states to provide equal protection of the laws).  The 13th Amendment does not merely prohibit slavery, but state imposition of the "badges of servitude."  Harlan exposed the disingenuousness of the Court's pretense that racial segregation imposed equal burdens on whites and blacks, arguing that its manifest intent was to stigmatize blacks.  Yet, Harlan hardly disavowed racism as a social philosophy, declaring that "every true man has pride of race" (554), and that "The white race deems itself to be the dominant race in this country.  And so it is in prestige, in achievements, in education, in wealth and in power.  So, I doubt not, it will continue to be for all time if it remains true to its great heritage. . . ." (559).

Korematsu v. United States, 323 U.S. 214 (1944)

Notorious decision upholding the constitutionality of the internment of Japanese-Americans during Word War II, on vague and unsubstantiated suspicions that they might be involved in espionage and sabotage on behalf of Japan.  This case has never been overruled, and indeed is still cited as establishing the principle that "all legal restrictions which curtail the civil rights of a single racial group are immediately suspect" and  must pass the test of "rigid [i.e. strict] scrutiny"(216).  The Court's use of this standard did not reflect the later Court's concern with means (requiring narrow tailoring), focusing only on the overriding importance of the military end in view, and deferring to the judgment of the military and of Congress that Japanese-American internment was necessary.  Don't miss the strong dissents by Justices Roberts, Murphy, and Jackson. Hirabayashi v. United States, 320 U.S. 81 (1942) established the key precedent for Korematsu.  It declared that "[d]istinctions between citizens solely because of their ancestry" are "odious to a free people whose institutions are founded upon the doctrine of equality." (100)

Shelley v. Kraemer, 334 U.S. 1 (1948)

Prohibited states from enforcing private agreements to exclude people from buying real estate or renting housing on account of their race.  Notable for its unusually expansive conception of state action.  Since the Civil Rights Cases, 109 U.S. 3 (1883), the 14th Amendment has been held to prohibit state action only, not the discriminatory actions of private individuals.  (The Civil Rights Cases overturned the Civil Rights Act of 1876, by which Congress prohibited segregation of privately owned public accommodations, as not falling within the scope of Congress' power to enforce the 14th Amendment.  The Civil Rights Act of 1964 achieved the same end as the 1876 act by appealing to Congress' power to regulate interstate commerce.)  Shelly v. Kraemer took the unusual step of arguing that, although private individuals could still make restrictive covenants, it would violate the Equal Protection clause of the 14th Amendment for states to enforce them.

Brown v. Board of Education, 347 U.S. 483 (1954)
Brown v. Board of Education, 349 U.S. 294 (1955)

Perhaps the most important Supreme Court case of the 20th c., declared state-imposed racial segregation of schools unconstitutional under the equal protection clause of the 14th Amendment, thereby overturning
Plessy v. Ferguson, 163 U.S. 537 (1896).  "Separate educational facilities are inherently unequal." (495).  Although its holding is now a fixed point of constitutional adjudication, some have criticized its reliance on social scientific evidence that segregation harms the self-esteem of black children. Shouldn't racial segregation be condemned as inherently unequal, even if black children showed unusual psychological resilience in the face of stigmatization?  The 1955 decision required public schools to desegregate "with all deliberate speed" (301).  In practice, the order to be "deliberate" was used by the states to slow down and obstruct desegregation orders, leading to a protracted, decades-long struggle to enforce Brown.

Griggs v. Duke Power Co., 401 U.S. 424 (1971)

A key case for understanding the Civil Rights Act of 1964, prohibiting discrimination in employment on account of race, color ethnicity, national origin, religion and sex.  Held that the Civil Rights Act of 1964 prohibits not only intentional racial discrimination, but hiring and employment policies that have a differential impact by race that (a) perpetuates the effects of past discrimination and (b) is not justified by business necessity.  Black plaintiffs in Griggs challenged Duke Power Company's requirement that applicants for its better-paying jobs pass a general test of academic ability.  The test was not valid (performance on the test did not predict peformance on the job), but black job applicants were much more likely to fail it than white applicants were, due to the history of racially segregated schools, which deprived black applicants of an adequate education.  The Court agreed that the 1964 Civil Rights Act required private employers to remove arbitrary obstacles to black advancement, even if those obstacles were not put in place with the intention to discriminate against them. Griggsrevolutionized the enforcement of the Civil Rights Act of 1964, by shifting focus away from intent and toward a disparate impact standard.  It established that a company's failure to employ a workforce whose racial composition reflected the racial composition of the local, qualified labor pool constituted prima facie evidence of a violation of the Act.  This shift set the stage for outcome-sensitive affirmative action policies.  Should be read in conjunction with Washington v. Davis, 426 U.S. 229 (1976).

Washington v. Davis, 426 U.S. 229 (1976)

Held that the Due Process Clause of the 5th Amendment (holding the Federal Government to the same equal protection standards imposed on the states by the 14th Amendment) prohibits only intentional racial discrimination, and does not require the government to correct for the unintended differential racial impact of its policies.  A fascinating companion case to Griggs v. Duke Power Co., 401 U.S. 424 (1971), because the fact pattern presented is nearly identical to that in Griggs, but the outcome was different.  Black plaintiffs sued the Washington D.C. police department for using a reading test to screen applicants for police jobs, arguing that the use of the test disproportionately disadvantaged black applicants, who did not score as well on it, and that the test was not validated.  The Court declined to hold the Federal Government to the same disparate impact standard under the 5th Amendment as private employers are held under the 1964 Civil Rights Act.

Village of Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252 (1977)

Following up on Washington v. Davis' requirement that discriminatory intent be proven before a violation of Equal Protection is found, held that local governments are free to adopt zoning regulations (e.g., prohibition of multi-family dwellings) with disproportionate racial impact as long as the specific history of such regulations does not reveal a discriminatory motive.  The chief difficulty with the evidentiary standard articulated in Arlington Heights for proving racist intent is that it assumes a requirement to show specific intent with respect to the particular case in question.  It thus neglects the sorts of evidence for the causal impact of race on zoning regulations that can only be developed via statistical analysis of many cases across different local government units.  In addition, it fails to inquire into deeper motivations behind constitutionally permitted motives--for example, whether a legitimate concern to adopt certain zoning regulations for the protection of property values reflects an underlying belief that the regulations protect property values precisely by keeping out unwanted racial groups.  This opinion effectively gives local governments carte blanche to exclude racial groups by means of nonracial proxies, so long as they keep quiet about their motivations, enact the regulations before facing a specific threat of integration, and apply the regulations consistently thereafter, regardless of race.

Affirmative Action, Phase I: Key cases establishing its legality
Race conscious affirmative action is subject to review under Title VII of the Civil Rights Act of 1964 if practiced by private or public employers and unions; Title VI if practiced by state or private recipients of federal funds; and the Equal Protection clause of the 14th Amendment if practiced by government units.  Bakke held that the requirements of Title VI are the same as those for the 14th Amendment, so in practice adjudication follows two tracks, one for Title VII, the other for the 14th Amendment.  The Court's central Title VII decision, Weber, generated a series of subsequent decisions that  put voluntary race-based affirmative action by private employers and unions on the firmest legal ground.  By contrast, the central cases establishing the constitutionality of state-run race-conscious affirmative action--Bakke and Fullilove--failed to achieve a majority representing the opinion of the Court, and crucially left open the question of the appropriate standard of review (strict vs. intermediate scrutiny) for the "benign" (legitimately motivated) use of racial classifications by the state.
Regents of the University of California v. Bakke, 438 U.S. 265 (1978)
 (RealAudio version of oral argument)
The key case establishing the Constitutionality (and compatibility with Title VI) of race-conscious admissions by universities.  The Court found a majority for its holdings--permitting race-based admissions, rejecting UCal's particular admissions system--but could not agree on a rationale.  Justice Powell's famous opinion argues that universities may practice race-conscious affirmative action for purposes of enhancing educational diversity, in recognition of their First Amendment right to academic freedom in judging what qualities of a student body would most enhance education.  However, the following constraints must be observed:  (a) racial quotas are not allowed; (b) all students must be evaluated according to common standards, by a common admissions committee; (c) race may not operate as an overriding factor, but only as a "plus"on a par with "a range of factors a university properly may consider in attaining the goal of a heterogeneous student body."  Four other justices joined Powell in agreeing that universities may use race-conscious affirmative action, but none joined the part of his opinion endorsing his "diversity" rationale.  Instead, they argued that universities may practice affirmative action to remedy societal (private) discrimination.  Powell disagreed:  the university's remedial uses of race must be confined to constitutional or statutory violations of equal protection and antidiscrimination law.  In a less noticed aspect of his opinion, Powell allowed that delivery of medical services to underserved populations is a compelling state interest that could justify race-based affirmative action, provided the use of race is necessary to advance that interest.  However, he rejected UCal Davis' argument to this effect as merely speculative, not supported by evidence. Thus, Bakke leaves this line of argument open to future evidentiary developments.
United Steelworkers of America, AFL-CIO-CLC v. Weber, 443
U.S. 193 (1979)
The key case establishing the permissibility under Title VII of a voluntary race-conscious affirmative action program instituted by an employer or union designed to remedy past discrimination in employment.  Upheld a temporary reservation for black employees of half the training opportunities for skilled steel jobs as a means to undo racial segregation of these positions.  The case illustrates how far the Court may go in adopting a purposive interpretation of a law at variance with its plain meaning and a considerable part of its legislative history, in light of evidence that Congress' purposes could not have been achieved with the means it expressly contemplated at the time of enactment.  Title VII prohibits, in plain language, an employer or union's acting "to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual's race" (42 U.S.C. 2000e-2), and the legislative history reveals proponents of Title VII vigorously denying that the law would require employers to implement racial quotas.  Brennan's opinion, relying on a tortured reading of the legal text, is valuable for one insight into the legislative history:  that Title VII had 2 purposes:  (1) to end employment discrimination, (2) to remedy the consequences of past discrimination so as to enable the integration of minorities into the economy.  The second purpose may sometimes require race-based remedies.  Blackmun's concurring opinion succinctly states the practical problem at the core of Title VII: a literal reading of the law would put employers who had engaged in racial discrimination in an impossible bind, subject to being dragged into court for violating Title VII whether it fails to remedy its discrimination, or voluntarily remedies it.  But it is absurd to interpret a law in such a way that it prevents violators from taking voluntary action to bring themselves into compliance with it. White Respondent Weber's reading of Title VII would require employers, against their interest, to specifically identify all victims of their past discrimination (thus exposing their liability for backpay), a task that in any event is impossible under the Griggs disparate impact standard, which recognizes that it is not always possible to identify those who would have been hired but for an employment practice (e.g., recruiting new employees by relying on their connections to current employees) that perpetuates racial exclusion.  Respondent Weber's reading of Title VII, forbidding admission to training for skilled positions a greater % of blacks than prevails in the qualified labor pool, would also unjustly "lock in" the effects of prior job segregation (given that the low % of blacks in the pool is a reflection of prior discrimination).  With this decision, the Court thus authorized employers not only to implement a voluntary affirmative action plan to remedy past, unadmitted if "arguable," violations of Title VII, but (1) to redress discrimination that lies beyond the scope of Title VII (e.g., for pre-Act discrimination), (2) in ways that benefit blacks who may not have been victims of that discrimination.  Don't miss Rehnquist's blistering dissent, which offers a powerful reading of Title VII's plain meaning and legislative intent, although it does not address the practical problem of voluntary compliance identified by Blackmun and ignores the evidence of broader remedial purpose evinced in Brennan's opinion.
Fullilove v. Klutznick, 448 U.S. 448 (1980)
The key case establishing the Constitutionality of race-conscious affirmative action in government contracting with private businesses.  Held that the Federal Government may set aside a percentage of contracting funds for minority-owned businesses, for the purpose of ensuring that federal funds are not used in such a way as to perpetuate the racially exclusionary effects of prior discrimination against minority businesses, provided (1) the set-aside does not continue longer than the effects of discrimination;  (2) the set-aside can be waived if sufficiently qualified minority contractors cannot be found, or if they charge a higher price than can be explained by their attempts to cover costs inflated by the present effects of prior disadvantage and discrimination; (3) a procedure is available to prevent minority businesses that are not disadvantaged from claiming the set-aside; (4) the set-aside does not place too great a burden on innocent nonminority businesses.  Private parties (prime contractors) can be required to subcontract 10% of their business to minority businesses even if they had not been guilty of discrimination themselves. (A line of decisions following Griggs had already established that under the Commerce clause,  Congress may prohibit business practices that perpetuate the effects of discrimination that, because it took place prior to the Civil Rights Act, was not illegal. Fullilove reasoned that Congress has the same power over private parties under its authority to spend for the general welfare, and over states under its 14th Amendment enforcement authority. Fullilove thus permits Congress to regulate state action under Griggs-style differential impact analysis, even though Washington v. Davis does not directly grant private parties such a cause of action under the Equal Protection clause of the 14th Amendment.)  Powell, concurring with the Court's judgment, held that the set-aside must and did meet the standards of strict scrutiny, but no other justices joined him in that view.  Thus, to the extent (not yet determined by the Court) that Fullilove's analysis depends on a less than strict scrutiny of Federal racial classifications, it has been overruled by Adarand.  Check out Stevens' vigorous dissent, rejecting the use of race for backward-looking (reparations-based) remedies to discrimination.  Stevens was later to become the Court's greatest advocate of a lax standard of review for forward-looking (instrumental) uses of race, as in his dissent in Wygant, accepting the role-model justification for race-based affirmative action.
Affirmative Action, Phase II: Consolidation and Extension
All of these cases except for Metro Broadcasting uphold affirmative action in employment.  Johnson and Mass. v. Feeney deal with affirmative action based on gender and veterans' status, which are subject to a less stringent standard of review than race-based affirmative action.  However, at the time these cases were decided, the Court had not yet determined that race-based affirmative action programs operated by the state are subject to strict scrutiny.
Personnel Administrator of Massachusetts v. Feeney, 442 U.S. 256 (1979)
Held that an affirmative action program that gave an absolute lifetime preference to U.S. veterans of the armed forces, regardless of the inferiority of their qualifications,  for employment in the Massachusetts civil service did not violate the state's duty to offer equal protection to women, even though it effectively barred women, no matter how highly qualified, from access to better civil service jobs.  (Petitioner Feeney was passed over for lower-scoring veterans for numerous jobs, including one for which she got the second highest examination score, another for which she got the third highest score.)  This case instantiates the following principles: (1)  the disparate gender impact of a state policy, no matter how extreme, of itself raises no constitutional questions, as long as that disparate impact is not purposeful (see Washington v. Davis); (2) discrimination among citizens in the allocation of benefits other than fundamental constitutional rights, on grounds other than race, ethnicity, sex, or religion, are accorded mere "rationality review"--a standard of review so lax almost any preferential treatment, no matter how extreme, can satisfy it; (3) if a preference based on a nonsuspect classification is constitutional, the degree of the preference, no matter how great, cannot make it unconstitutional; (4) states are free to discriminate on the basis of a facially gender-neutral classification even if another government unit (in this case, the Defense Department) has purposefully prevented women from joining it.  (At the time Mass. v. Feeney was litigated, the Pentagon's ceiling on female enlistment in the Armed Forces resulted in a class of veterans only 2% female.)  Don't overlook Marshall's eloquent dissent, which demonstrates how Supreme Court precedents on the 14th Amendment could have been taken in a different direction.
Firefighters v. Cleveland,  478 U.S. 501 (1986)
Upheld the right of employers and unions to enter into a voluntary consent decree that provides a race-conscious remedy for prior employment discrimination, even if the remedy benefits individuals not identified as victims of that discrimination.  Follow-through on Weber, decisively affirming Blackmun's concurring opinion in that case, stressing the need to permit employers to remedy their discriminatory action by voluntary means.  Rejects the view that Section 706(g) of Title VII, which precludes a district court from entering an order requiring the hiring or promotion of an individual who was refused employment or promotion for any reason other than discrimination, prevents employers' voluntary adoption of a remedy for discrimination that has the effect of hiring or promoting individuals who had not been discriminated against.
Sheet Metal Workers v. EEOC, 478 U.S. 421 (1986)
Upheld an EEOC imposed racial membership quota on the Sheet Metal Workers' union, as a narrowly tailored remedy to stubborn and egregious racial discrimination practiced by the union, in defiance of court order under Title VII.  While Title VII prohibits racial quotas for the purpose of racial balancing, it does not prohibit quotas as a remedy for discrimination.  Nor does it limit remedies to actual victims of prior discrimination when a history of egregious discrimination shows that mere injunctive relief will not stop the violations, imposition of temporary race-based affirmative action, including a quota,  may be necessary to stop the illegal behavior.  Rejects the view that section Section 706(g) of Title VII prevents a court-ordered imposition of such a remedy, interpreting its provision to mean only that the court may not order the employment, promotion, or admission to a union of someone who would have been refused this even in the absence of discrimination. Sheet Metal Workers therefore empowers courts to impose race-conscious remedies in the same way that Firefighters v. Cleveland empowers employers and unions to voluntarily remediate their discrimination through consent decrees.
United States v. Paradise, 480 U.S. 149 (1987)
Upheld the constitutionality under the 14th Amendment of a court-ordered temporary racial promotion quota on a government unit (the Alabama Department of Public Safety) as a remedy for its "pervasive, systematic, and obstinate discriminatory exclusion of blacks" in defiance of multiple previous court judgments.  As in Weber, approved an "accelerated" promotion quota (in which the ratio of blacks to whites promoted exceeds the ratio of qualified blacks to qualified whites in the workforce, until the % of blacks in higher positions equals the % of qualified blacks in the workforce).  Four justices argued that the quota survived "strict scrutiny."  This case does for the 14th Amendment what Sheet Metal Workers v. EEOC does for Title VII.  More than that, it articulates some considerations that fall under the "narrow tailoring" requirement for racial classifications:  (1) necessity (lack of race-neutral alternatives to the same end); (2) flexibility (goal or quota waived if there is a lack of qualified minorities); (3) limited duration (preference is temporary, with a forseeable stopping-point); (4) means-end fit (reasonable relationship between the goal/quota and numerical representation of minorities in the qualified labor pool); (5) burdens on innocent third parties are not too great.  (Other cases have insisted that racial preferences not be (6) overinclusive (help those not disadvantaged by discrimination) or (7) underinclusive (fail to help disadvantaged nonminorities)
Johnson v. Transportation Agency, 480 U.S. 616 (1987)
Upheld the legality under Title VII of a gender-based affirmative action plan adopted by a state actor to remedy the traditional gender segregation of transportation agency jobs.  The plan did not establish quotas, but allowed a woman's sex to count as a "plus" in applying for traditionally male jobs for which she was qualified, and established as a goal the representation of women in traditionally male job categories in proportion to their representation in the local labor force.  Followed through on Weber's contention that an employer need not admit to prior discrimination to justify an affirmative action plan; a statistical disparity between the % of women in the job category and in the local labor pool is enough to justify the adoption of a plan. Johnson shows how the logic of affirmative action, as it develops, pushes very hard on the distinction between remedying the effects of discrimination and straightforward race- and gender- proportional representation.  Johnson distanced itself from the latter purpose by distinguishing (1) long-term (not controlling) from short-term goals of a program (where the long-term goal may take proportional representation as a benchmark); (2) goals to be attained rather than maintained (attainment signifies the end of discrimination, maintenance signifies an interest in proportional representation for its own sake).  Scalia correctly observes in dissent that this case holds, without admitting it, that an employer may adopt an affirmative action program intended to overcome "societal discrimination" (i.e., general social attitudes discouraging women to pursue certain occupations) rather than just its own discrimination.  Scalia intends this as a criticism, arguing that the antidiscrimination requirements of Title VII cannot be less stringent than those in the 14th Amendment (which, Bakke and Wygant suggest, forbid a state from using racial classifications to remedy any discrimination other than its own).  If Scalia's equivalence claim is accepted, Johnson could be viewed in the other direction as a precedent for state uses of suspect classifications to remedy societal discrimination.
Metro Broadcasting, Inc. v. Federal Communications
Commission, 497 U.S. 547 (1990)

Upheld the constitutionality of the FCC's policies establishing a preference for awarding broadcast licenses to minority owners of radio and TV stations.  Held that the minority preference is justified in light of the state's important First Amendment interest in enabling the expression of diverse viewpoints to the public.  This is the only Supreme Court case other than Bakke upholding racial preferences on "diversity" rather than remedial grounds.  A very odd decision, since it applied mere "intermediate scrutiny" to a racial classification even though it postdated Croson's move toward strict scrutiny.  O'Connor's dissent expresses important objections to the very idea of linking race to viewpoint diversity, arguing that it illegitimately trades on racial stereotypes and ultimately conflicts with First Amendment requirement that the state remain viewpoint neutral.  To the extent that Metro Broadcasting relies on intermediate scrutiny, it has been overruled by Adarand.  Few legal scholars believe that the FCC's program would be upheld under strict scrutiny.

Affirmative Action, Phase III: Key cases constraining affirmative action programs

Firefighters v. Stotts, 467 U.S. 561 (1984)
A precursor to Wygant, limiting remedial action available under Title VII.  Held that a court may not override a bona fide seniority system with a race-based system for determining layoffs just because the employer was under a consent decree covering hiring and promotion (but not mentioning layoffs).
Wygant v. Jackson Board of Education, 476 U.S. 267 (1986)
Held: (1) Race-based affirmative action, even when valid for hiring, training, and promotion, may not be practiced for layoffs, because the burdens placed on innocent whites are unfairly concentrated on particular individuals.  (2) To justify a race-based affirmative action employment program, the employer must have a "strong basis in evidence" that remedial action is necessary.  (3) Rejected a "role model" theory for justifying the hiring of black public school teachers in proportion to the percentage of black students in a school, as a remedy for "societal discrimination." (The Jackson Board's theory appeared to be that black students needed black teachers to help them cope with societal discrimination against the students.) (4) Powell's decision is also widely regarded as holding that government units may use racial classifications only to remedy their own discrimination, not discrimination in society or by other government units.  However, his reasoning against "societal discrimination" as a ground for state action seems to be addressed more to the peculiarities of the role model theory and to the vagueness and indeterminateness of a raw appeal to "societal discrimination," without precise identification of the specific discriminatory actions being remedied, than to the source of discrimination in state or private action.
Richmond v. J. A. Croson Co., 488 U.S. 469 (1989)
Held that state and local government race-based affirmative action programs are subject to strict scrutiny: they must be shown (1) to be pursuing a compelling government purpose and (2) the use of race must be narrowly tailored: (a) neither under- nor over- inclusive, and (2) lacking race-neutral alternatives that could achieve the same end.  Although commonly reported as a great blow to affirmative action programs, in requiring the highest level of scrutiny for state and local race-based affirmative action programs, O'Connor's opinion for the Court in fact opened up a remedial justification for affirmative action seemingly foreclosed by Powell's opinions in Bakke and Wygant, which suggested that the state may use racial criteria only to remedy its own discrimination, and not the discrimination of private parties.  O'Connor argued that the state has a compelling interest in not being a "passive participant" in private discrimination.  Race-based set-asides in contracting could therefore be justified if sufficient evidence were produced that race-neutral contracting awards would amount to state complicity in or perpetuation of the effects of the racially exclusionary practices of contractors.  Issues of evidence loomed large in this decision, which overturned the City of Richmond's 30% contracting set-aside for minority owned businesses on account of the city's failure to amass sufficient evidence of discrimination in contracting requiring a race-based remedy.  However, the evidentiary issues have also been widely misinterpreted, as supposedly reflecting the Court's skepticism about the continued existence of racial discrimination.  The point of requiring states to amass evidence of discrimination, and to tailor their program to empirical evidence about local conditions, is that this is evidence that the state's purpose in establishing a set-aside for minority subcontractors is to avoid passive participation in the racially exclusionary practices of contractors (a legitimate and compelling purpose), rather than just to establish a racial spoils system (an unconstitutional purpose).
Adarand Constructors, Inc. v. Pena, 515 U.S. 200 (1995)
Held that Federal race-based affirmative action programs, no less than state and local programs, must satisfy the requirements of strict scrutiny. Croson had not settled this matter, since much of that opinion had stressed how the 14th Amendment empowered the Federal government while constraining the states. O'Connor, writing for the Court, justified this position on the basis of three principles.  First, skepticism: given the dismal history of race-based state action, the motivations of the state in using a racial classification must be closely scrutinized by the courts, and its mere assertion that its motives are benign cannot be taken at face value.  Second, consistency:  the same level of scrutiny must be applied to all race-based state policies, whatever the race of those affected by that policy.  Third, congruence:  equal protection analysis under the Fifth Amendment (which applies to the Federal government) is the same as under the Fourteenth Amendment (which applies to the states).  To the extent (not determined by the Court in this decision) that Fullilove and Metro Broadcasting relied on a less than strict scrutiny in upholding the race-based policies they reviewed, they are overruled.

Latest Legal Cases on Affirmative Action: Lower Court Decisions

Weber and subsequent Title VII cases building on this precedent have put private race- and gender- conscious affirmative action plans in employment on a fairly secure legal footing; thus, lower courts are not currently deciding cases that are likely to unsettle legal doctrine or established practices in this area.  Litigation is more active for 14th Amendment cases.  (a) Numerous affirmative action plans by state employers have been litigated; results have been mixed, with no clear trend for or against such plans.  Most litigation in this area focuses on assorted ambiguities in narrow tailoring requirements. (b) Since Croson, state and local set-asides for contracting have been vigorously challenged.  Results here are also mixed, but with a decided trend against set-asides.  Most decisions in this area have focused on evidentiary requirements; set-asides survive only if supported by elaborate and expensive studies demonstrating a cause for remedial action.  The key question the Supreme Court will eventually need to answer is how much and what kind of evidence of discrimination by private contractors is sufficient to warrant a public contracting set-aside.  The major new development in this area is a lower-court decision on the famous Adarand case, upholding a racial preference in state contracting.  (c)  Because Bakke is the only Supreme Court case dealing with affirmative action in education, and did not contain a majority opinion, doctrine in this area is the most unsettled.  Litigation is very active, and several highly significant cases have recently been or are being litigated.  This section therefore focuses on the major recent lower-court cases dealing with public education, with an aside to the unusual Wittmer v. Peters and the important Adarand decision.

Wittmer v. Peters, 87 F.3d 916 (7th Cir. 1996)

Unusual opinion by Judge Richard Posner, upholding a nonremedial racial preference in promotion for corrections officers, on the ground that the state has shown  that minority officers are needed to achieve law enforcement objectives.  The state provided evidence that black inmates were unlikely to play the "correctional game of brutal drill sergeant and brutalized recruit" unless some of the black corrections officers held positions of authority.  The Supreme Court refused to review this case on appeal, cert. denied 117 S. Ct. 949 (1997).

Adarand Constructors, Inc. v. Rodney Slater, 2000 Colo. J. C.A.R. 5574  (10th Cir., 2000)

The latest disposition of the famous Adarand case, which, when reviewed by the Supreme Court (as Adarand v. Pena) firmly established strict scrutiny as the standard for evaluating all government affirmative action programs.  The Supreme Court remanded the case back to the district court, instructing it to judge it by the strict scrutiny standard.  The district court ruled that the state's incentive to hire minority subcontractors failed strict scrutiny in Adarand Constructors, Inc. v. Pena, 965 F. Supp. 1556 (D. Colo. 1997).  In this eighth round of adjudication, the 10th Circuit Court of Appeals reversed.  It stressed the fact that the Supreme Court's Adarand decision insists that some affirmative action programs can pass strict scrutiny.  It found that Congress has a compelling interest in redressing the effects of current and past racial discrimination and that it had a strong basis in evidence to conclude that such discrimination occurred in the construction industry.  The Court's summary of evidence on the extent of discrimination in the construction industry is a worthy reference on this issue.  It also found that the design of Congress' current program, as recently revised to meet strict scrutiny standards--were narrowly tailored to the compelling purpose.  This opinion stands as a exemplar of how a race-based affirmative action program in contracting can meet strict scrutiny.

Podberesky v. Kirwan, 38 F.3d 147 (4th Cir. 1994)

Invalidated the University of Maryland's scholarship program reserved for African-Americans.  (1) Applying Croson's evidentiary demands, denied that UM had sufficient evidence that current factors discouraging black enrollment and achievement were caused by UM's own past discrimination:  (a) UM's bad reputation in the black community, being tied solely to knowledge of UM's historic discrimination, cannot justify a racial preference because such knowledge will always be available but remedies must be temporary.  (b) UM's alleged hostile environment, due to the hostility of white students, is a form of societal discrimination not specifically attributable to UM's past discrimination. (2) UM's statistical underrepresentation of blacks and lower retention and graduation rates, [even if, in the face of weak evidence, the cause is stipulated to be UM's own past discrimination], cannot justify the scholarship program, which fails narrow tailoring requirements: (a) it is overinclusive, in attracting high achieving blacks, who were not victims of UM's past discrimination, and in being awarded to non-Maryland blacks, when the program is alleged to be needed to remedy the underrepresentation of Maryland blacks at UM; (b) it relied on the role-model theory rejected in Wygant; (c) the benchmark of underrepresentation was determined against an arbitrarily inflated reference pool that failed to factor out nonrace-based causes of underrepresentation, and that hence failed to constrain the extent of remediation to the extent of the effects of past discrimination; (d) it failed to consider (unspecified) race-neutral alternatives. [Sorry, I could not find a link to the full text of this opinion; the link is to an extended summary.] 

Piscataway Township Board of Education v. Taxman, 91 F. 3d 1547 (3d Cir. 1996)

Invalidated a race-conscious affirmative action plan for faculty employment decisions, adopted for diversity rather than remedial purposes.   The Court viewed the case solely within the terms of Title VII, holding, on its reading of Weber and Johnson, that it permits race-based employment decisions only for remedial purposes.  Notwithstanding Bakke's claim that schools have a compelling interest in diversity, Title VII forbids schools from pursuing that nonremedial interest through racially discriminatory employment (as opposed to admissions) policies.  Even if diversity were admitted as a valid purpose, (1) the state provided no benchmark of adequate diversity to measure progress and termination of the program, and thus acted arbitrarily and (2) placed an undue burden on white petitioner Taxman through its layoff provision.  The dissent disputes the claim that Weber and Johnson restrict valid affirmative action plans to remedial purposes, holding that neither case determined the outer bounds of permissible aa plans. This case was set to go to the Supreme Court, cert. granted, 117 S. Ct. 2506 (1997), until it was settled by a financial intervention by civil rights groups, worried that its fact pattern was not a favorable one under which to review the general principles of affirmative action in education.  Cert dismissed, 118 S. Ct. 595 (1997).  The case had an unfavorable fact pattern for the resolution of a diversity claim, because it involved a layoff decision (see Wygant), and because the % of blacks who were teachers exceeded their % in the population.  (On the other hand, the affirmative action plan allowed a racial preference only among equally qualified candidates with equal seniority.)

Hopwood v. State of Texas materials
, including 78 F.3d 932 (5th Cir., 1996)

Invalidated the University of Texas' law school's race-based admissions program, reversing the district court's judgment that UT had compelling interests in using race-based admissions, Hopwood v. Texas, 861 F. Supp. 551 (W.D. Tex.1994). (1) Audaciously, the Court claims that Powell's opinion in Bakke, permitting racial diversity as a compelling state interest in educational contexts, is not controlling, because no other justice joined him in that part of his opinion; only remedial justifications for racial classifications are allowed.  Equating race with skin color, the Court found not even a rational basis for connecting such a superficial characteristic to educational outcomes.  It denied that race may be used as a proxy for socially relevant characteristics, regarding this as an assertion of harmful, stigmatizing stereotype. (2) The Court rejected UT's remedial claims, arguing that (a) the law school may only remedy its own discrimination, not discrimination in society or even in the Texas school system at large, or even in UT at large, because it is not in a position to measure the discrimination of other agents; (b) Following Podberesky v. Kirwan, rejected UT's claims of bad reputation among and hostile environment to minorities as justifying remediation.  Judge Weiner, concurring, assumed that an educational interest in diversity could justify a race-conscious admissions policy, but held that UT's policy was not narrowly tailored to achieve this aim: it was underinclusive, in that it did not include goals for all racial/ethnic groups that could contribute to diversity; it operated too much like a quota; and it assured only "facial diversity," not "true diversity" (suggesting that the quest for racial diversity is a quest for skin color diversity, which is irrelevant to education).

Wessman v. Gittens, 160 F.3d 790 (1st Cir. 1998)

Invalidated Boston Latin School's race-conscious admissions program.  Half the places in the school were allocated in rank order of test scores in a competitive examination; the other half were allocated by proportional race/ethnic representation among those who scored in the top 50% of all applicants.  Without rejecting Bakke's claim that diversity is a compelling state interest, the Court found that BLS's admissions program violated Bakke's strictures:  (1) race and ethnicity were the only admissions factors other than test scores, while the pursuit of true diversity allows racial/ethnic status to count as but one factor among other factors contributing to diversity; (2) race-neutral admissions on the basis of test scores alone would yield a combined Black and Hispanic student population of 15-20%, why does "diversity" require more?; (3) BLS's use of proportional representation in selection shows that its purpose was "racial balancing"--impermissible under the 14th Amendment.  The Court also rejected BLS's remedial claims, observing that the Boston school district, although once under a desegregation order, had been declared unified and that it had failed to provide sufficient evidence that the currently observed racial academic achievement gap was a vestige of the district's own past discrimination.  It rejected anecdotal evidence as insufficient to support claims about systemic causes of disadvantage, and statistical evidence gathered in another district as irrelevant to the Boston school district.  Finally, it found BLS's proportional representation admissions not narrowly tailored for remedial purposes: causally ineffective, in not eliminating the purported cause of disadvantage (low teacher expectations in elementary school); and overinclusive, in admitting minorities from private schools and nondisadvantaged white and Asian students.  The opinion relies heavily on its reading of Croson's evidentiary requirements.  The dissent articulates a laxer standard of those evidentiary requirements, holding that the state need only make out a prima facie case of discrimination, and arguing that it had.

Hunter v. Regents of the University of California, 190 F.3d 1061 (9th Cir. 1999)

Upheld a race-conscious admissions program to an elementary school run by the UCLA Graduate School of Education and Information Studies on unusual instrumental grounds.  The school was set up to as a site for research and training teachers to cope with the educational needs of California's increasingly diverse student population.  Students are therefore selected on the basis of their suitability for research interests.  Applying strict scrutiny, the Court ruled that the school served the compelling state interest of improving the quality of education in urban public schools, and that racial and ethnic criteria were narrowly tailored to serve that interest.  UCLA submitted extensive evidence that ensuring a racially and ethnically diverse student population was necessary to advance their goal, because such diversity is needed to capture educationally relevant differences in students' cultures, language proficiency, learning styles, and important data on intergroup interaction and conflict.  Holding that researchers have First Amendment academic freedom interests in defining their research needs, the Court refused to second-guess academic judgments that race and ethnicity were indispensible selection criteria for research subjects.  The Supreme Court declined to hear this case, cert. denied 121 S. Ct. 186 (2000).

Johnson v. Regents of the University of Georgia, 106 F. Supp. 2d 1362 (11th Cir., 2000)

Rejected the University of Georgia's affirmative action program, on the ground that "the 'diversity' interest is so inherently formless and malleable that no plan can be narrowly tailored to fit it."  Found that the University of Georgia had not provided compelling evidence that racial diversity advanced important educational goals, and followed O'Connor's opinion in Metro Broadcasting in casting doubt on any connection between racial diversity and the "robust exchange of ideas," upon which Powell based his diversity defense in Bakke. Followed Hopwood in also finding  that Powell's diversity argument is not binding precedent.  An interesting feature of the case is that the University of Georgia gave an admissions preference to men (an extra .25 points, compared to .50 points for being non-white).  The court rejected the gender preference on the same grounds as the racial preference, viz. that neither racial balancing nor gender balancing were legitimate state purposes.

Johnson v. Regents of the University of Georgia, 263 F.3d 1234 (11th Cir., 2001)

Affirms the district court ruling above, rejecting the University of Georgia's affirmative action policy, but on different grounds.  The Court of Appeals declined to rule on the validity of the diversity defense, although its review of the legal precedents suggests skepticism about the viability of such a defense.  Instead, it chose to reject the University of Georgia's policy for its failure to meet narrow tailoring requirements.  A mechanical, fixed increment of points added to an application simply due to the applicant's race is (1) of arbitrary weight, (2) effectively screens out applications before taking into consideration any other dimensions of diversity that the applicant may contribute to a student body, (3) fails to treat applicants as individuals, and (4) fails to consider race-neutral alternatives to the same purpose of diversity.  This opinion has potential implications for the upcoming case in the 6th circuit, Gratz v. Bollinger.  Although the 6th circuit, in Grutter v. Bollinger, has upheld the University's Law School admissions policy, which does not add a fixed number of points to an application on account of the applicant's race, it has not yet ruled on the university's undergraduate case (Gratz), which employs a procedure not unlike that rejected in Johnson. 

Katuria Smith v. University of Washington Law School, 233 F.3d 1188 (9th Cir., 2000)

Upheld the constitutionality of the University of Washington's Law School admissions program, on the ground (contra Hopwood) that Justice Powell's opinion in Bakke remains good law.  The case has no direct practical effect for affirmative action, as the citizens of Washington had already prohibited it by referendum in proposition I-200.  The legal importance of the decision lies in its argument that Powell's opinion is controlling because his diversity argument is the narrowest ground on which the holding of Bakke can be rationalized.  This follows the legal rule that "when a fragmented Court decides a case and no single rationale explaining the result enjoys the assent of five Justices, 'the holding of the Court may be viewed as that position taken by those Members who concurred in the judgments on the narrowest grounds.'" Marks v. United States, 430 U.S. 188 (1977). The 9th circuit's application of the Marks rule to Bakke is disputed by Judge Friedman in Grutter v. Bollinger, below.

Gratz and Hamacher v. Bollinger, 122 F. Supp. 2d 811 (E.D. Mich. 2000)

A major district court decision, upholding the University of Michigan's current affirmative action policy for undergraduate admissions.  Judge Duggan declared the University's earlier policy unconstitutional for being functionally equivalent to a quota.  The larger significance of the case is as a model of how a diversity defense of affirmative action could survive strict scrutiny.  The distinguishing character of this case is the extensive presentation of evidence and expert testimony supporting the link between racial diversity and educational interests, in "The Compelling Need for Diversity in Higher Education."  None of the other educational affirmative action cases, except for Hunter, have presented more than conclusory testimony to the educational importance of racial diversity, or explained the significance of race to diversity interests.  Rejects the Hopwood Court's reasoning that the Supreme Court has effectively eliminated diversity as a compelling interest in educational contexts, and upholds diversity as a compelling interest, in view of the unrebutted evidence advanced by defendants.  Contains a useful discussion of how the University's present policy meets the narrow tailoring constraints required by strict scrutiny, in contrast with its previous policy.  However, in a later opinion, Gratz and Hamacher v. Bollinger, (Feb. 26, 2001), Judge Duggan rejected the arguments of student intervenors on behalf of affirmative action, that the University's affirmative action program was justified as redressing the effects of racial discrimination. Here again, the quality of the evidence presented was crucial to the disposition of the case.  Judge Duggan argued that the intervenors had not shown that the University was practicing affirmative action for remedial (as opposed to diversity) purposes, and that the current shape of its program fit diversity interests better than remedial ones.  In addition, intervenors failed to provide persuasive evidence of past or current discrimination by or in the University, and failed to demonstrate that racial preferences were a necessary remedy to the disparate impact of the use of standardized tests in admissions.

Grutter v. Bollinger et al, 97-CV-75928-DT (E.D. Mich., 2001)

Finds that the University of Michigan Law School's affirmative action policy violates equal protection.  Grutter is a parallel case to Gratz, decided by Judge Bernard Friedman.  The key argument in the case is Judge Friedman's rejection of diversity as a compelling interest, on the ground that Marks does not designate Powell's opinion in Bakke as controlling.  In addition, Judge Friedman ruled that the Law School's affirmative action program failed the narrow tailoring requirements of strict scrutiny--(1) its alleged diversity interest was too amorphous to determine what numbers would meet it, (2) the policy  had no time limits, (3) was functionally indistinguishable from a quota in weighting race too heavily in admissions,  (4) did not provide principled grounds for selecting particular racial groups for favorable treatment, and (5) failed to consider race-neutral alternatives to achieving racial diversity, such as abolishing the use of standardized tests.  Also rejected intervenors' arguments that the Law School's policy was justified as a remedy for past and current discrimination.  The University relied on the same evidence on the educational value of diversity as it did in Gratz.  Gratz and Grutter together therefore offer a model of how differently disposed judges can rule in opposite ways on essentially the same evidence.  Judge Friedman's order that the Law School cease using race as a factor in admissions was stayed by the 6th Circuit Court of Appeals in Grutter v. Bollinger
, 2001 FED App. 0103P (6th Cir., 2001).

Grutter v. Bollinger et al, 288 F.3d 732 (6th Circuit, 2002)

Upholds the University of Michgan Law School's affirmative action policy.  This is the most important educational affirmative action case decided since Bakke, because it is virtually certain to be reviewed by the Supreme Court.  Grutter, in affirming a race-based affirmative action admissions policy, conflicts with Hopwood, which categorically rejects affirmative action in education.  When two federal circuits issue contradictory rulings on such an important legal matter, the Supreme Court has a compelling reason to set the record straight.  In this bitterly contested 5-4 opinion, the majority upheld the Law School's policy on the following reasoning:  Powell's Bakke decision, upholding diversity as a compelling interest, is good law, and the Law School's admissions policy is indistinguishable from the "Harvard Plan" that Powell held up as a paradigm of an admissions policy satisfying the narrow tailoring constraints for using racial preferences in educational contexts.  The cautious strategy of the court is therefore to persuade the Supreme Court not to open the question of Bakke's authority, but to accept the conventional view that Powell's decision is controlling and precedent should be followed.  The strategy appeals to the Court's conservatism in refraining from overturning prior cases on which many people have come to rely, and which have been the basis for establishing entrenched practices--see Planned Parenthood v. Casey 505 U.S. 833, 855 (refraining from overturning Roe v. Wade, the case that established womens' right to abortion).  This stategy could fail if the Court decides to revisit the merits of affirmative action de novo, as the dissent urges (arguing that Marks provides no grounds for thinking that Powell's opinion, or the opinion of the other four justices who joined him in the judgment upholding the use of race, is controlling).

   Listen to the Oral Arguments before the Supreme Court, Gratz v. Bollinger, Grutter v. Bollinger, April 1, 2003Or read the Complete Transcript of the Oral Arguments.

The questions from the Supreme Court Justices suggest that the justices will vote as predicted, with O'Connor holding the swing vote and keeping her cards close to her chest. The big surprise in the arguments was the close attention paid by the Court to the U.S. Military Academy's briefs in support of affirmative action. Another surprise was the lack of attention paid to what Bakke stands for, suggesting, perhaps, that the Court intends to review the case de novo, rather than relying on precedent. The wrangling in the arguments over the meaning of "quotas" is mostly silly, given that the popular definition of "quota" (any numerical goals at all, however qualified) has no relation to the legal definition (which only prohibits inflexible targets that cannot be waived if a lack of qualified minorities is available to fill them).

   Grutter v. Bollinger, et. al (slip opinion), 539 U.S. -----_(2003)

In the most imporant Supreme Court decision in 25 years (since Bakke was decided), the Supreme Court upheld the University of Michigan Law School's affirmative action program on the ground that it was narrowly tailored to meet the compelling state interest of realizing the educational value of diversity in the context of public and private schools. O'Connor, as widely predicted, held the swing vote in this 5-4 decision. And, as the oral arguments indicated, the Court dispensed with the mind-numbing task of trying to figure out the precedent set under the Marks test by Bakke's notoriously fragmented opinion, instead deciding to evaluate the merits of the diversity defense de novo. Don't be deceived by the uniformity of the "diversity" rhetoric flowing through the majority opinion, however. In fact, O'Connor's decision appeals to several additional rationales distinct from the educational argument: (1) the Military Academy's argument that the legitimacy and hence effectiveness of its officer corps would be diminished if it were not racially representative; (2) an integrationist argument, that "effective participation by members of all racial and ethnic groups in the civic life of our Nation is essential if the dream of one Nation,indivisible,is to be realized"; (3) a more general argument that the legitimacy of the U.S. elite depends on its "visible"openness to all races ("the path to leadership [must] be visibly open to talented and qualified individuals of every race and ethnicity"). Consistent with precedent, the opinion draws a clear distinction, muddled in public discourse, between quotas (prohibited) and goals (permitted). It also clarifies the "race-neutral alternatives" test for narrow tailoring, allowing that alternatives need only be considered, not implemented, if there is good faith reason to believe implementation would fail to realize the school's diversity objectives or compromise its interest in maintaining academic standards. This endorses Sam Issacharoff's justification for affirmative action, cited above, that it is the only way to achieve the dual goals of racial diversity and academic excellence. The opinion raises some puzzles: (1) O'Connor grants deference to the Law School's representation of its own benign motives (p. 17: the Law School's "good faith" is "presumed" "absent a showing to the contrary"), in apparant contravention of what O'Connor once defined as the main point of strict scrutiny of racial classifications--namely, to probe the state defendant's professions of benign motivation with skepticism (Adarand). (This point is noted by Thomas in his partial dissent.). (2) What is the significance of the Court's stated "expectation" that affirmative action will no longer be needed to bring about diversity in 25 years? Thomas, partially concurring, takes this to signify an absolute termination date. This, however, would be inconsistent with the rationale of strict scrutiny, which is that racial means must be permitted by the Court as long as the compelling interest remains and racial means are needed to satisfy it. I interpret it, rather, as a signal that the Court doesn't trust educational institutions to terminate programs that have outlived their usefulness. The Court is setting up a date with future affirmative action litigants, 25 years hence, to revisit the issue. Justice Scalia's intemperate dissent adds nothing of interest to the debate, and plainly contradicts both Supreme Court precedent and the original intent of the Framers of the 14th Am. in declaring a categorical prohibition of state-sponsored racial discrimination. Focus instead on Justice Thomas' much more interesting partial dissent, opening with the forceful words of Frederick Douglass. Thomas' interpretation of Douglass' insistence that white attempts at benevolence rather than justice toward blacks do them harm (by expressing and promoting patronizing attitudes, stigmatization, and policies that hook blacks into underachieving dependency, claiming benefits on the ground of victimization and pity rather than achievement) go to the core of Thomas' moral objections to affirmative action. (They are not, however, Constitutional objections, since unintended racially harmful effects are not unconstitutional; under Washington v. Davis, only invidious racially discriminatory purposes are.) Note in particular Thomas' claims on behalf of the beneficial effects of de facto racially segregated Historically Black Colleges on black students' achievement. He both takes them seriously as a ground for preserving the racially specific mission of the HBC's (see his opinion in Fordice), and denies that they could justify racial discrimination--thereby exposing the double-edged sword of relying on "diversity" as a 1st Amendment rationale for affirmative action: what if racial homogeneity were appealed to as a 1st Amendment ground for keeping out minority groups? In contrast with Thomas' ringing and forceful critique of affirmative action, the dissents of Rehnquist and Kennedy focus on legal minutia.

   Gratz et al v. Bollinger et al, (slip opinion) 529 U.S. ----- (2003)

This companion case to Grutter v. Bollinger, above, rejected by a 6-3 vote the University of Michigan College of Literature, Science, and the Arts affirmative action admissions program, on grounds of failure to narrowly tailor its use of racial means to the state's compelling interest of diversity. The Court's main objection to the U-M's undergraduate policy is that its mechanical "point system," which automatically grants every black, Hispanic, and Native American applicant 20 points in admission, is too crude and mechanical to satisfy the requirement that each applicant's contribution to diversity must be subject to individualized review. From a legal point of view, this case is less significant than Grutter. However, it will certainly require many schools across the country to revise their admissions systems, and increase the costs of admissions programs that directly seek racial diversity in their student bodies.

Parents Involved in Community Schools vs. Seattle School District No. 1 No. 05–908 (2007)

Overturns a voluntary school integration program, arguing that the school district's interest in the educational value of diversity cannot justify a program that focuses on race alone.

Fisher v. University of Texas, 631 F.3d 213 (5th Cir. 2011) and Case Materials at UT.

After the 5th circuit overturned University of Texas's affirmative action program, the Texas state legislature instituted a "top 10% plan", admitting the top 10% of each Texas high school graduating class to UT.  This yielded a somewhat racially diverse class by race-neutral means, by taking advantage of de facto racial segregation in Texas high schools.  In practice, the diversity yield for Latino students was much better than for blacks, because Latino students are more heavily segregated than blacks in Texas public high schools.  After Grutter v. Bollinger upheld the constitutionality of race-based affirmative action, UT modified its admissions policy to by adding a race-conscious affirmative action policy for students who fall short of the top 10%, to improve its diversity.  This time, the district court and the 5th Circuit upheld UT's policy.  However, this decision has been appealed to the Supreme Court, which heard oral arguments on Oct. 12, 2012 and will render a decision in 2013. 

Sources on the Web

         Diversity Web: an interactive resource hub for higher education

Association of American Colleges and Universities website offering information about and support for a wide range of diversity practices in U.S. higher education.

Results about Affirmative Action

Useful links to sites presenting pros, cons, and history of affirmative action from Resultsabout.com.

Research Supporting Affirmative Action in Higher Education, University of Michigan

Links to expert testimony and amicus briefs on behalf of the University of Michigan in Grutter v. Bollinger and Gratz v. Bollinger as well as links to much other research.

Understanding Prejudice

Contains links to thousands of other resources related to prejudice; surveys, interactive demonstrations and exercises, a directory of experts on prejudice and discrimination, and much more. How sexist are you? Take the Ambivalent Sexism inventory. Do you harbor unconscious biases in favor or against various race, age, etc. groups? Take the Implicit Association Test. Get teaching tips on how to present controversial subject matter to students.

Racism and Psychology

From the American Psychological Association, basic discussions of racism, stereotpying and discrimination from a psychological point of view.

Racism--Global Issues

From Globalissues.org, useful resource on manifestations of racism around the world.

The Geography of Race in the United States

Website by Elizabeth Anderson, the author of this webpage, and Jeff Jones. Contains census maps of major U.S. metro areas, displaying high degrees of racial segregation, explaining causes and consequences of segregation, information on disenfranchisement and voting rights, etc. A principal feature of this website is the ability of the browser to "flash" among different census maps of the same territory with a flick of the mouse, enabling a vivid demonstration of the segregation of different racial groups and the high correlation of segregation with income.

Debating Racial Preference

Excellent website by Curtis Crawford, pairing arguments for and against affirmative action, and discussing major legal cases on affirmative action. Crawford, an opponent of affirmative action, sometimes provides the rebuttals to arguments in favor.

Affirmative Action Review: Report to the President

In 1995, President Clinton ordered a comprehensive review of federal affirmative action programs, under the slogan "mend it, don't end it."  This is the report that resulted from his order.  Useful for the numerous citations to empirical research on the need for and impact of affirmative action programs.
United States Commission on Civil Rights
The USCCR is an independent investigating commission, with no enforcement power, that investigates alleged civil rights violations around the country.  Check out their extensive publications on all aspects of civil rights, including work for Americans with Disabilities and reports on problems facing overlooked minority groups, such as Arab-Americans.
The Glass Ceiling Commission
The Glass Ceiling Commission has issued numerous reports on invisible barriers to the advancement of women and other disadvantaged groups in employment.  Search in the comprehensive collection of glass ceiling reports archived by the Cornell University School of Industrial and Labor Relations.

The Affirmative Action and Diversity Project: A Web Page for Research
Bibliographic reference on race, racism, gender, sexism, and affirmative action.  Huge list of citations with many links to brief articles.  Stresses the variety of opinions.  Extremely well organized by theme; comprehensive in scope.
Laurence Hinman's Ethics Updates
 Excellent bibliographic references in his pages on race and ethnicity, gender and sexism; race, ethnicity and moral theory, and sexual orientation.  Loads of links to other internet resources.
Interesting site for corporate news, reports, and advocacy on affirmative action.  Many useful links, especially to breaking news.
Asian-Nation: The Landscape of Asian-America
Covers issues affecting Asian-Americans, including stereotyping and affirmative action.
Harris Sussman's Diversity Page
Perspectives on diversity from someone who runs diversity workshops for businesses and nonprofit institutions.  Useful for seeing how diversity is promoted in the business sector.  Links to many other diversity sites, alas without annotation to guide the browser.


Contains references to sources on diversity, affirmative action education, etc., that are recommended by scholars and experts in their fields.  The scope of the Infography ranges far beyond affirmative action issues, to numerous fields of general interest.

Center for Equal Opportunity
Think-tank, headed by Linda Chavez, offering numerous papers from a conservative point of view on affirmative action, immigration issues, etc.  Promotes an assimmilationist ideal for immigrants, and argues that Latino immigrants by and large aspire towards and meet that ideal.

Toward a Fair Michigan

The Michigan Civil Rights Initiative, a proposed Michigan constitutional amendment to ban affirmative action, passed Nov. 2006.  While formally independent from the campaign for the MCRI, and officially dedicated to promoting fair debate on its merits, this website is run by supporters of MCRI, as its content reveals.  The site now looks ahead to what should be done in a post-affirmative action state.

Michigan United

Website of opponents of the Michigan Civil Rights Initiative and defenders of affirmative action in Michigan.  Considers what can be done to promote diversity in the wake of the passage of MCRI, which banned affirmative action as practiced by government bodies in the state.


Up-to-date resources on affirmative action and numerous other civil rights issues.

You are visitor  since November 25, 1998.
This page last modified on May 16, 2013.
Please send comments to eandersn@umich.edu

Elizabeth S. Anderson
John Rawls Collegiate Professor of Philosophy and Women's Studies
Arthur F. Thurnau Professor
Department of Philosophy
University of Michigan
2239 Angell Hall 1003
435 S. State St.
Ann Arbor, MI 48109-1003