Race and ethnicity — the social construction of difference

Intuition Beginner

Race feels obvious. You can see it on people's faces, the story goes, and it tells you something real about who they are. Sociologists push back hard on that feeling. Race is real — but it is real as a social force, not as a biological essence.

Think of a passport, a census form, or a school registration. Each one asks you to tick a box: Black, White, Asian, Hispanic, and so on. Those boxes are not discovered in nature. They are made by governments, changed over time, and drawn differently from one country to the next. The categories do work — they sort people, distribute resources, and grant or deny rights — but the work they do is social and political, not biological.

Geneticists have known for decades that there is more genetic variation within any racial group than between groups. Two people grouped as "Black" or "Asian" may differ from each other more than either differs from someone grouped as "White." If race were a biological fact, the boxes would line up with genes. They do not.

Yet race still shapes lives. It affects who gets a mortgage, who is stopped by police, who is hired, and how long a person lives. The puzzle this unit solves is the central one: how can something not written in biology be so powerful in society? The answer is that societies build race into their institutions, and built structures hit hard whether or not they are natural.

Visual Beginner

The table maps the core distinctions this unit develops.

Term	What it points to	One-line contrast
Race	A socially built category sorting people by supposed physical ancestry	Not a fixed biological group
Ethnicity	Shared culture, language, history, or origin	Cultural, not the same as race
Prejudice	A rigid attitude or belief about a group	Lives in the head
Discrimination	Action that treats a group unfairly	Lives in behaviour
Institutional racism	Racial bias built into how organisations work	Not about individual intent

A simple way to hold the distinction: ethnicity is about who you are through heritage and culture, while race is about how society sorts you using a small set of invented boxes.

Worked example Beginner

Residential segregation is one place where a socially built category produces measurable numbers. Sociologists measure it with the index of dissimilarity. The index answers a concrete question: what share of one group would have to move house so that every neighbourhood matches the city-wide mix?

Take a small city with two neighbourhoods and two groups, each 1,000 people city-wide. Neighbourhood A has 800 White residents and 200 Black residents. Neighbourhood B has 200 White residents and 800 Black residents.

Work out each group's share in each place. In A, the White share is 800 out of 1,000, or 0.8; the Black share is 0.2. In B it flips: White share 0.2, Black share 0.8.

The gap in A is 0.8 minus 0.2, which is 0.6. The gap in B is also 0.6. Add the two gaps to get 1.2, and take half. The result is 0.6.

That number, 0.6, is the index of dissimilarity. It means about 60 percent of either group would need to move to a different neighbourhood for the two groups to be evenly mixed. Real cities run far higher. In recent decades the Black-White index for Detroit and Milwaukee has sat near 0.80. Sociologists call values above 0.60 hypersegregation: separation so extreme that chance or income alone cannot explain it. The boxes were invented — but the streets built from them are measured, material, and stark.

Check your understanding Beginner

Formal definition Intermediate+

Race is a system of social classification that sorts people into putatively biological categories on the basis of supposed shared ancestry and selected physical markers (skin colour, hair texture, facial features). The categories are produced and reproduced by states, law, science, and everyday practice; they vary across societies and shift across time; and they are made to carry unequal value, even though they do not correspond to discrete biological populations.

Ethnicity is a classification by shared culture — language, religion, history, national origin, customs — that is typically self-identified and based on belonging rather than on imposed physical typing. The same population may be one race and several ethnicities (for example, people racially classified as Black may be Ethiopian, Jamaican, or African American), which shows the two schemes cut on different axes.

Prejudice is an attitude: a rigid, usually unfavourable generalisation about a category of people. Stereotypes are the specific beliefs — often schematic, exaggerated, and resistant to counter-evidence — that fill in that attitude. Prejudice and stereotypes are cognitive and affective states; they need not produce action.

Discrimination is action. It is unequal treatment of people on the basis of their membership in a category. Sociologists distinguish three levels. Individual discrimination is one person's biased act — an insult, a refusal to serve. Institutional discrimination is biased outcomes produced by the routine operation of organisations — hiring tests, school funding formulas, lending algorithms — whether or not any actor intends bias. Systemic (or structural) racism is the cumulative, mutually reinforcing effect of institutional discrimination across the housing market, the labour market, education, healthcare, and criminal justice, durable across generations ^{[Bonilla-Silva 1997]}.

Intersectionality, introduced by Kimberlé Crenshaw in 1989, is the analytical claim that systems of subordination organised around race, gender, class, sexuality, and other axes interact to produce experiences of advantage and disadvantage that cannot be recovered by examining any single axis on its own ^{[Crenshaw 1989]}. Intersectionality is a claim about interaction, not simple addition: the combination of two axes yields a qualitatively distinct position, not the sum of two separate burdens.

A final, essential distinction runs through the unit. Positive claims describe what is the case — that a racial wealth gap exists, that Black-White segregation measures 0.80 in some metros. Normative claims state what ought to be the case — that equality is desirable, that segregation is unjust. Sociology's descriptive findings do not by themselves settle normative questions, but they constrain them: any defensible normative position must reckon with the measured structure of racial stratification.

Comparative framework Intermediate+

The study of race is a contested field. Reading the major traditions against one another exposes what each captures and what each misses, and shows that several live questions have no settled answer.

Du Bois: double consciousness and the color line

W. E. B. Du Bois wrote in 1903 that "the problem of the Twentieth Century is the problem of the color line" and gave the field its first sustained analysis of racialized consciousness ^{[Du Bois 1903]}. His concept of double consciousness describes the experience of always seeing oneself through the eyes of a society that regards one as a problem — a "two-ness" of being both American and Black, with the two identities held in tension. Du Bois also anticipated intersectionality by decades, attending to the simultaneous operation of race, gender, and class in ways the discipline would not recover until the late twentieth century.

Park and the race-relations cycle

Robert Park's race-relations cycle — contact, competition, accommodation, assimilation — proposed that immigrant and minority groups move through predictable stages toward absorption into a dominant culture, a process he treated as largely irreversible ^{[Park 1950]}. Park's framework captures real dynamics of immigrant incorporation, but it has been criticised for treating assimilation as natural and inevitable, for centering the European immigrant experience, and for underplaying the coercive force of racial domination that blocks assimilation for groups marked as non-white.

Omi and Winant: racial formation

Omi and Winant's racial formation theory argues that race is a matter of social structure and cultural representation simultaneously ^{[Omi-Winant 1986]}. Races are formed through racial projects — efforts to organise and distribute resources along racial lines while simultaneously creating or reproducing racial categories and meanings. The theory's strength is that it explains why race is neither illusion nor essence: it is a real structure produced by historical projects that can be contested and transformed. Racial formation is the dominant contemporary framework in the sociology of race precisely because it avoids both the biological-realist trap and the "race is just an idea" trap.

Bonilla-Silva: racialized social systems

Bonilla-Silva's racialized social systems theory pushes further, arguing that racism is not primarily a matter of prejudice in individual hearts but a structural feature of societies "organized around race" ^{[Bonilla-Silva 1997]}. In a racialized social system, racial categories are built into the allocation of privileges and penalties; the interests they create make racial inequality self-maintaining even in the absence of bigoted actors. This reframes the unit of analysis from attitude to structure.

A first contested question: class or race?

A standing dispute concerns the relative weight of class and race in producing inequality. William Julius Wilson's The Declining Significance of Race (1978) argued that for some outcomes — inner-city employment, poverty — class had overtaken race as the operative cleavage, a claim meant to redirect policy toward universal economic programs. Wilson's critics — Bonilla-Silva, Oliver and Shapiro, and others — contend that ostensibly class-based outcomes remain racially structured: the racial wealth gap cannot be explained by income, education, or family structure once asset history is taken into account. The empirical synthesis most consistent with the evidence is that race and class are co-constitutive: racial exclusion shapes the class position into which families are sorted, and class position conditions how racial disadvantage is experienced. The dispute is positive (what causes what), but each side carries a normative subtext about what policy should prioritise, and those subtexts should be marked as such rather than smuggled in as neutral description.

A second contested question: assimilation or pluralism?

Whether minority groups should assimilate to a dominant culture or maintain distinct identities is a normative question that sociology can describe but not settle. Classical assimilation theory (Park; Milton Gordon 1964) treats convergence toward the mainstream as the expected endpoint. Segmented assimilation (Portes and Zhou 1993) argues that contemporary second-generation immigrants assimilate into different segments of a stratified society — sometimes into the middle class, sometimes into an underclass — depending on context of reception, skin colour, and human capital. Pluralist and multicultural frameworks (Horace Kallen, later postcolonial writers) hold that maintaining distinctive communities is both a fact and a value. These positions are not just empirical predictions; they embed judgements about whether cultural convergence is desirable. Sociology can map the conditions under which each path occurs, but it cannot, from data alone, declare one outcome the right one.

Primary research and commentary

The literature on race mixes two kinds of work, and a careful reader keeps them distinct. Primary research — Lewontin's 1972 apportionment of human genetic diversity; Massey and Denton's American Apartheid (1993) documenting segregation indices; the American Anthropological Association's 1998 statement on race; audit studies sending matched résumés to employers — generates evidence. Commentary and theory — Du Bois, Park, Omi and Winant, Bonilla-Silva, Crenshaw — interpret that evidence and frame what it means. Both are necessary, but they should not be cited interchangeably: a finding about segregation levels is not the same as an argument about why segregation persists.

Bridge. The comparative frameworks above build toward the structural account of racial stratification developed in the stratification unit 30.04.01, and this is exactly where the unit on institutions 30.05.01 will take up the concrete mechanisms — schools, labour markets, housing, criminal justice — through which racial projects are enforced; the central insight — that race is a structure produced by projects rather than a thing found in bodies — appears again in the Master-tier analysis of racial formation, hypersegregation, and intersectionality, where the foundational reason racial inequality is self-maintaining is made precise.

Exercises Intermediate+

Advanced analysis Master

The measurement of segregation: beyond dissimilarity

The index of dissimilarity captures evenness, but residential segregation has several dimensions that no single index exhausts. Massey and Denton (1988) identified five: evenness (dissimilarity, the Gini index), exposure/isolation (the probability of contact between groups), concentration (the physical amount of space occupied), centralisation (location relative to the city centre), and clustering (the degree to which minority neighbourhoods adjoin one another). A group can be high on one dimension and moderate on another; hypersegregation denotes a group that is extreme on at least four. Black Americans in a set of large US metropolitan areas have historically met the hypersegregation threshold — a level of spatial isolation experienced by no other racial group in any comparable democracy. This is the empirical core of Massey and Denton's American Apartheid (1993), and it is primary research: measured indices from census data, not commentary on them.

Racial formation and the racial state

Omi and Winant's racial formation theory gives the structural account its mechanism. A racial project is simultaneously an interpretation, representation, or explanation of racial dynamics and an effort to organise and distribute resources along racial lines. The racial state is not neutral with respect to these projects; it is itself a site where racial categories are codified — in the census, in immigration and naturalisation law, in blood-quantum rules for tribal recognition, in police classification — and where the distribution of rights and resources is enforced. Racial formation thus locates the construction of race neither in individual prejudice (too thin) nor in a reified "structure" floating above actors (too vague), but in the historically specific projects through which categories and distributions are produced together. The theory is the field's contemporary common ground because it accommodates both cultural meaning and material allocation without reducing one to the other.

Bonilla-Silva: structure as the default

Bonilla-Silva's racialized social systems argument completes the structural turn by inverting the burden of explanation ^{[Bonilla-Silva 1997]}. In a racialized system, racial inequality is the expected, default outcome of normal institutional operation; what requires explanation is not the persistence of racial inequality but its reduction. This reframing has a methodological consequence: the analyst asks how institutions are organised around race, rather than searching for prejudiced individuals to blame for disparities. It also has a normative edge — the theory is committed to the view that racial hierarchy is unjust — but that commitment is separable from the positive claim that racial hierarchy is structurally reproduced. Marking the two apart keeps the descriptive apparatus usable by scholars who share the diagnosis of mechanism without endorsing every normative conclusion drawn from it.

Du Bois, the color line, and the recovery of a founder

Du Bois's 1903 analysis of double consciousness and the color line was marginalised within a discipline structured by the very racial hierarchy it described; the recovery of Du Bois as a founding theorist is itself a finding about the sociology of knowledge ^{[Du Bois 1903]}. Double consciousness names the internal cost of a society that forces the racially subordinated to view themselves through a hostile gaze — a phenomenological complement to the structural accounts above. Du Bois's insistence that the "color line" was global, linking racial domination within the United States to colonialism abroad, anticipated the postcolonial and world-systems analyses that the discipline would not develop for another half-century. His work is the primary source from which the contemporary theoretical apparatus partly descends; citing it as such distinguishes origination from later commentary.

Intersectionality as structural analysis

Crenshaw's 1989 intervention was a legal critique — anti-discrimination doctrine's single-axis structure could not represent the discrimination faced by Black women — but its sociological generalisation treats intersectionality as a claim about how systems of subordination co-produce social positions ^{[Crenshaw 1989]}. Patricia Hill Collins's matrix of domination extends it: race, gender, class, sexuality, and nation are not separate systems that happen to overlap but interconnected structures that constitute one another, operating simultaneously at the level of personal biography, of community, and of social institutions. The methodological point is that an analysis that holds all axes but one constant — "controlling for class," "controlling for gender" — can misstate the phenomenon, because the position of interest is produced by their joint operation. Intersectionality thus generalises the structural turn from a single axis to the field of axes together.

Assimilation, ethnicity, and their critics

The classical assimilation tradition, running from Park through Gordon, treated convergence toward a dominant culture as the normal endpoint of intergroup contact. Its post-1965 successor, segmented assimilation (Portes and Zhou 1993; Zhou 1997), keeps the focus on second-generation trajectories but divides the mainstream into segments — middle-class, underclass, and selective-acculturation paths — whose determinants include context of reception, phenotype, and the strength of the ethnic enclave. Ethnicity theory (Yancey, Erickson, and Julian 1976) goes further, arguing that ethnicity itself is situationally evoked: ethnic boundaries are drawn differently in different institutional settings and may be dormant until mobilised. Whiteness studies (Roediger 1991; Ignatiev 1995) applies the construction insight reflexively, treating "whiteness" not as a default but as a historically produced racial project — Irish, Italian, and Jewish immigrants were made white over time, which shows race responding to interest and politics rather than to biology.

Race and health: weathering and fundamental causes

Arline Geronimus's weathering hypothesis documents that Black Americans, and Black women in particular, experience accelerated physiological ageing — earlier onset of hypertension, cardiovascular disease, and maternal morbidity — at every income level relative to White Americans. The mechanism is the cumulative stress of navigating a racialized society, not genetics or individual behaviour. Link and Phelan's fundamental cause theory explains why racial health gaps survive every specific epidemiological transition: class and race confer flexible resources (money, knowledge, power, networks) that exploit whatever health-relevant advantage the current environment offers, so that when one risk factor is neutralised a new one takes its place. Weathering is primary research (measured physiological outcomes); fundamental-cause theory is the interpretive frame that organises it.

Whiteness, colour-blind racism, and the contemporary frame

Bonilla-Silva's later Racism Without Racists (2003) argues that the dominant contemporary ideology in the United States is colour-blind racism: a framework that explains racial inequality through apparently non-racial terms (cultural difference, market outcomes, "natural" preferences) and so preserves racial hierarchy while disavowing it. The argument is contested — some scholars hold that overt prejudice remains the operative force; others that racial attitudes have genuinely liberalised while structural mechanisms persist — and the dispute is partly empirical (what explains the measured disparities) and partly normative (what the state should do about them). Presenting both positions, and marking where the empirical claim shades into a normative one, is the discipline's standing obligation on this question.

Synthesis. Putting these together — segregation indices, racial formation, racialized social systems, double consciousness, the matrix of domination, segmented assimilation, weathering — the central insight is that race is a structure produced by projects and reproduced by institutions, and this is exactly why it is durable without being natural: the foundational reason racial inequality persists across generations is that the mechanisms are systemic, not attitudinal, so the inequality is maintained even as explicit prejudice declines. The structural turn generalises from single institutions to the whole field of axes (race, gender, class, sexuality), is dual to the cultural account in that meanings and material allocations are produced together through racial projects, and builds toward a unified theory of stratification in which the bridge is the recognition that constructed categories can be the most powerful structures of all, an insight that appears again in every downstream analysis of institutions 30.05.01, deviance and social control 30.06.01, and global inequality 30.07.03 pending.

Connections across the curriculum Master

This unit deepens the race axis introduced in the stratification overview and feeds several later units.

The prerequisite 30.04.01 established race as one of five axes of stratification alongside class, gender, caste, and global position, and introduced intersectionality; the present unit formalises that axis — prejudice versus discrimination, individual versus institutional, the structural default — and supplies the theoretical traditions (Du Bois, Park, Omi-Winant, Bonilla-Silva) that the stratification unit could only name.

The culture unit 30.02.01 treats ethnocentrism, stereotypes, and cultural relativism as mechanisms of meaning-making; the present unit shows how those cultural mechanisms are yoked to the distribution of resources through racial projects, so that "cultural difference" becomes, in Bonilla-Silva's analysis, one of the resources colour-blind ideology uses to explain away structural inequality.

The anthropology unit 31.01.01 carries the biological critique of race — the four-fields consensus that human variation is clinal and does not sort into discrete races, formalised in the American Anthropological Association's 1998 statement; sociology takes that biological finding as given and asks the different question of how socially constructed categories nonetheless produce real stratification, so the two disciplines are complementary rather than redundant.

The socialization unit 30.03.01 explains how racial identities are learned — how children come to know themselves and others as raced — which is the micro-level process through which the racial projects described here are internalised and reproduced across generations; double consciousness is the phenomenological reading of that process from the standpoint of the subordinated.

The deviance and social control unit 30.06.01 takes up mass incarceration as a racializing institution: the criminal justice system converts arrest and conviction into permanent civic disability (voting, housing, employment) concentrated among Black Americans, which is the paradigmatic case of a racial project that distributes penalties along racial lines while operating through formally neutral law.

Historical and philosophical context Master

The modern concept of race took shape in the eighteenth and nineteenth centuries alongside European colonialism and the transatlantic slave trade. The racial taxonomies of Linnaeus (1735) and Blumenbach (1795) classified humans into a small number of varieties and ranked them, providing a putatively scientific gloss for the extraction of African labour and the dispossession of Indigenous peoples. Race, on this account, was made — by naturalists, colonists, and jurists — to justify a redistribution of resources already underway; the category was a racial project before the term existed.

W. E. B. Du Bois's The Souls of Black Folk (1903) opened the sociological analysis of race from the standpoint of the subordinated ^{[Du Bois 1903]}. The concepts of double consciousness and the color line located racial domination simultaneously in subjective experience and in global structure, and Du Bois's later work (Black Reconstruction, 1935) traced how the ideology of race had been used to split the working class along racial lines — an analysis of racialized social systems a half-century before the term.

Robert Park and the Chicago School formalised the study of race relations in the early twentieth century, proposing the contact-competition-accommodation-assimilation cycle to describe immigrant incorporation ^{[Park 1950]}. The framework shaped a generation of research but was criticised from the 1960s onward for treating European immigrant assimilation as the model and for understating the coercive, caste-like character of Black-white relations in the United States — a critique sharpened by the civil rights movement and by the renewed reception of Du Bois.

The structural turn of the late twentieth century rebuilt the field. Omi and Winant's Racial Formation in the United States (1986; second edition 1994) argued that race is constructed through historically specific racial projects linking meaning and resource distribution, and supplied the framework that now organises mainstream sociology of race ^{[Omi-Winant 1986]}. Bonilla-Silva's 1997 article "Rethinking Racism" argued that racism should be analysed as a property of racialized social systems rather than of individuals, inverting the explanatory burden ^{[Bonilla-Silva 1997]}. Crenshaw's 1989 article "Demarginalizing the Intersection of Race and Sex" introduced intersectionality as a critique of single-axis anti-discrimination doctrine and became one of the most cited ideas in the social sciences ^{[Crenshaw 1989]}.

The standing philosophical question is whether, and how, racial categories ought to be retained in policy and law. Eliminativists hold that the state should stop classifying by race; retentionists hold that classification is necessary to detect and repair structural racism; the dispute is normative and the empirical record constrains but does not settle it. Du Bois, Omi-Winant, Bonilla-Silva, and Crenshaw are all, in different registers, retentionist in their analytics; colour-blind eliminativism is the chief alternative in the contemporary literature, and each position should be presented as a position rather than as a neutral description of what race "really" is.

Bibliography Master

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

2. Bonilla-Silva, E. (2003). Racism Without Racists: Color-Blind Racism and the Persistence of Racial Inequality in the United States. Lanham, MD: Rowman & Littlefield.

3. Collins, P. H. (1990). Black Feminist Thought: Knowledge, Consciousness, and the Politics of Empowerment. Boston: Unwin Hyman.

4. Crenshaw, K. (1989). "Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique of Antidiscrimination Doctrine, Feminist Theory and Antiracist Politics." University of Chicago Legal Forum, 1989(1), 139–167.

5. Du Bois, W. E. B. (1903). The Souls of Black Folk. Chicago: A. C. McClurg.

6. Du Bois, W. E. B. (1935). Black Reconstruction in America, 1860–1880. New York: Harcourt, Brace.

7. Geronimus, A. T. (1992). "The Weathering Hypothesis and the Health of African-American Women and Infants." American Journal of Public Health, 82(11), 1573–1578.

8. Gordon, M. M. (1964). Assimilation in American Life: The Role of Race, Religion, and National Origins. New York: Oxford University Press.

9. Lewontin, R. C. (1972). "The Apportionment of Human Diversity." Evolutionary Biology, 6, 381–398.

10. Link, B. G., & Phelan, J. (1995). "Social Conditions as Fundamental Causes of Disease." Journal of Health and Social Behavior, 35 (Extra Issue), 80–94.

11. Massey, D. S., & Denton, N. A. (1988). "The Dimensions of Residential Segregation." Social Forces, 67(2), 281–315.

12. Massey, D. S., & Denton, N. A. (1993). American Apartheid: Segregation and the Making of the Underclass. Cambridge, MA: Harvard University Press.

13. Omi, M., & Winant, H. (1986). Racial Formation in the United States: From the 1960s to the 1980s. New York: Routledge.

14. Park, R. E. (1950). Race and Culture. Glencoe, IL: Free Press.

15. Portes, A., & Zhou, M. (1993). "The New Second Generation: Segmented Assimilation and Its Variants." Annals of the American Academy of Political and Social Science, 530, 74–96.

16. Roediger, D. R. (1991). The Wages of Whiteness: Race and the Making of the American Working Class. London: Verso.

17. Wilson, W. J. (1978). The Declining Significance of Race: Blacks and Changing American Institutions. Chicago: University of Chicago Press.

18. Yancey, W. L., Erickson, E. P., & Juliani, R. N. (1976). "Emergent Ethnicity: A Review and Reformulation." American Sociological Review, 41(3), 391–403.