Review History

The Construction of Racial Identity in the United States

About: P. Schor, Compter et classer : histoire des recensements américains, EHESS.

by Daniel Sabbagh , 17 December 2010
with the support of FMSH

Télécharger l'article : PDF

The history of the American census is also that of the institutionalization of racial and ethnic categories. In a detailed study of the practices of classifying the U.S. population from 1790 to 1940, the historian Paul Schor demonstrates the instability of the categories produced and the way in which forms of classification of blacks were extended to other categories.

Reviewed: Paul Schor, Compter et classer: histoire des recensements américains, Paris, Presses de l’École des Hautes Études en Sciences Sociales, collection “En temps et lieux,” 2009. 383 p., 22€.

The happy few who had the opportunity to plunge into the thesis of the historian Paul Schor, Senior Lecturer at the Université Paris Diderot (Paris 7), have been expecting that the publication of the corresponding book would place this remarkable study within the purview of a wider audience. This is now achieved and their patience amply rewarded: captivating from start to finish and endowed with an abundant iconography, Compter et classer will indeed become the standard reference work on the history of ethnic and racial categories in the United States census from 1790 – date of the first exhaustive enumeration of the population carried out by the federal government – until 1940 [1].The subject is important for at least two reasons. First, the census has plausibly contributed to the institutionalization of race as a social category in the American context, even its persistence in social awareness and as a language referent, notwithstanding the disappearance of its historic matrix (slavery) and the marginalization of its theoretical foundation (racism as a pseudo-scientific ideology with rationalizing value). Second, it is probable that the census has also played a role in the internalization by individuals of certain norms of classification within the existing categories, as witnessed by the absence of major upheavals due to the shifts to self-classification in 1970, and to the possibility of declaring oneself belonging to a plurality of “racial” groups introduced in 2000 [2]. And so the subject deserved a study as complete as this one, in which the census is minutely examined in all its dimensions: as an instrument for proportioning legislative representatives and fiscal contributions among the states [3] – the initial raison d’être for its introduction (according to the 1790 Constitution) was “an enumeration and description of the inhabitants, distinguishing the sexes and color of free persons”, but also a “site for assigning collective identities to individuals” (p. 20) inscribed within a hierarchy that was widely perceived as such, and in fact a determining factor in the “production of a national community” (p. 11). As such, and thanks to its documentary richness, the book has no equivalent on this side of the Atlantic [4]. Rather than summarize the twenty chapters one by one, I would like to specify where I think its added value lies, before nevertheless indicating two of its limitations.

The first distinctive trait of Paul Schor’s approach consists of centering the analysis as much as possible not on racial taxonomies and related discourses, but on the practices of classification used by the census agents charged with assigning individuals to one or another of the various reference groups, without Schor yielding to the temptation to attribute to these practices the degree of coherence of any racist ideology being elaborated at the time or subsequently. Thus Schor demonstrates that in a certain number of cases, the creativity inherent in operations that were not totally subject to codification beforehand was ultimately translated into the introduction into the nomenclature of new categories corresponding to unanticipated responses present in rather large numbers in the forms filled out during previous censuses (“Chinese” in 1870, “Japanese” in 1890, and no doubt “Mexican” in 1930). This was a form of adaptation to “data” on the ground, itself issuing from the necessary interactions for administering the questionnaire. In this respect, extending the work of Ariela Gross (2008), Walter Johnson (2000), and John Tehranian (2000) – also focused on the concrete dimension of the classification process and how doubtful cases were handled, but solely in the legal arena – Schor is similarly led to stress the uncertain and unstable dimension of racial attributions that were partially determined by the “social and residential context” (p. 220) and by the “performance” or “self-presentation” (p. 93) of the individuals in question. The constructivist approach puts the accent on the porous nature of boundaries and the uncertainty of categorizations (Yanow (2003)); far from circumscribing its pertinence to the modern period, this approach may be profitably applied to what sometimes passes as the golden age of racial essentialism.

Second, following the recommendations of Rogers Brubaker and his collaborators (Brubaker, Loveman and Stamatov (2004)), and modeled on other recent books using heteroclite materials (Hattam (2007)) or centered on immigration policies (Ngai (2005)), Compter et classer has the great merit of gathering into the same analytical framework categories traditionally described in the United States as arising from “race” on the one hand, and from “ethnicity” on the other. While in France, where the delegitimizing of racism has entailed a disqualification of race as a term claiming to be descriptive, so that “ethnic” often functions as a euphemistic substitute for “racial” in public discourse, in the U.S. context the two terms are by no means synonymous. Not only do they refer to different groups – the former principally to European immigrants of an origin other than Anglo-Saxon and their descendants, the latter to black, Amerindian, and Asiatic minorities as well as to the white majority – but this difference also corresponds to a disjunction between the relevant historiographic fields that are liable to occult them. On the contrary, the advantage of Schor’s de-compartmentalizing is that it highlights two very significant factors: first, the breadth of the chronological gap between the premises of the racialization and ethnicization of American public statistics [5] (the precocity of the former being all the more striking because at the time the census had no ambition whatever to “draw up an inventory of society” (p. 33)) [6]; and second (and especially), the degree to which racialism came to constitute the conceptual matrix of ethnicity. More precisely, it is only thanks to this overall perspective of population categories that the system thereby formed becomes perceptible, a system founded upon the prime division between Blacks and Whites. The writer shows very convincingly the transmutation of the one-drop rule (prescribing the assigning to the black race of any individual having merely one black ancestor) into a principle of hypodescendence [7] that was progressively extended, more or less strictly, to the ensemble of strata and ramifications of the classificatory arrangement, from “racial” minorities other than blacks and then to groups defined on an “ethnic” basis. The inclusion underforeign stock” (a statistical entity composed of immigrants and their descendants) of persons born in the United States of whom one parent was born abroad testifies to this extension.

Third, in addition to these insights into the nature of the articulation between “race” and “ethnicity” that correlates the one-drop rule with the principle of hypodescendence, Schor’s analysis is remarkably precise and nuanced. On the one hand, he shows how “race” is only a maximalist translation of an absolutely asymmetrical conception of racial “metabolism,” which reduces white identity to an object of alteration that is infinitely vulnerable, and black identity to an agent of corruption that is perfectly effective, an asymmetry that was reflected more or less directly in many concrete procedures. Thus during the census of 1850 (the first to separately identify each individual), the census agents had received instructions to indicate a person’s color only if they thought they did not belong to the white population, making its status as pole of reference evident and fully assumed. Similarly, although the category of “mulatto” had also been introduced in 1850 in the hope of corroborating the hypothesis of the lowest longevity of members of this group – an anticipated proof of the noxious effects of interbreeding [8] – and was maintained (with rare exceptions) until 1920 [9], the available documentation about the modes of presenting the data indicates that the Census Bureau always considered mulattos as a sub-category of blacks and not as a collectivity occupying an intermediary position between the two extremes of the American “racial order” (King & Smith (2005)). However, the writer constantly stresses that the census agents possessed neither the authority nor the means necessary to conduct the type of genealogical investigations theoretically required by applying the one-drop rule in cases of doubt about the “race” of the surveyed person after the initial visual examination, hence the importance of the informal interactions mentioned above. He also shows that extending the scope of the principle of hypodescendence to racial minorities other than blacks even indicated its incompleteness, since in the case of Indians (until the end of the period under consideration), census instructions explicitly prescribed departing from the rules when the person in question was “considered to be […] white in the community in which he or she lives” (p. 241).

The one-drop rule and the related principle of hypodescendence constituted the main distinctive trait of U.S. racial configuration, whose singularity as the “product of a national history” (p. 227) is revealed by the particularly original study carried out in 1930 Census in various peripheral territories that had been recently acquired, where other assigning norms were current. Thus in Puerto Rico, the author shows how the suppression of the “mulatto category led to an increasing number of persons being identified as white by the census agents, in contrast to the complete absorption of mulattos within the black population that was forecast and observed on the continent. [10] In the Virgin Islands (the small Caribbean archipelago ceded by Denmark in 1917), the prospect of the elimination of the “mixed” category (the functional equivalent of mulatto) even aroused protests on such a scale that the governor ultimately asked for and obtained its retention (as a derogatory term). Principally determined by motivations of a symbolic and not material order, this resistance to the imposition of rules of U.S. racial categorization offers a striking contrast with the more utilitarian orientation of the rare opposition noted in the rest of American territory, such as the objection to the racialization of Mexicans in 1930, which was due to its manifest discriminatory repercussions. As the book confirms, such opposition never challenged the legitimacy of the very principle of the racial classification of persons by government agencies, which is particularly remarkable in the eyes of a French observer.

As accomplished as it is, Paul Schor’s book calls for some reservations of secondary importance that I now mention briefly. First, its internal balance leaves something to be desired. Although the author states that the study as a whole was “conceived as an implicit comparison with our day” (p. 338), I admit being somewhat disappointed by a foreshortened epilogue that tells the story of the evolutions observed over the six subsequent decades (1940-2000) in less than six pages. Sometimes difficult to follow by an informed reader, this swift summary could have been the basis of a fuller chapter that would have been able to reduce the bulk assigned to the fifth part, in which certain developments – notably the trajectory of Charles Hall, one of the black employees of the Census Bureau who was ultimately promoted to the rank of “Specialist in Negro Statistics” (chapter 18) – seem anecdotal and reflect the understandable concern to exploit untapped sources whose intrinsic interest is not quite convincing. In the event of a translation – which is desirable for all the aforementioned reasons – the author is encouraged to remedy this imbalance.

Moreover, while we cannot reproach a historical book on the United States of America for not taking a position on the current French controversy over “ethnic statistics,” it is hard not to notice the absence of any indication of the possible political implications of a study that is devoted to demonstrating the irreducibly national character of the American system of ethno-racial classification and that stresses the difficulties in exporting it [11]. The book in fact includes many developments that might inform the debates in France, starting with the detailed study of the process by which the recording of the country of birth of the subjects and those of their parents (the latter presented today in France as a possible approximate substitute for the missing data about minorities in the grip of the most flagrant discriminations) [12] finally opened the way to introducing the issue bearing most directly on ethnicity as community of ancestry. Did such a process, perceptible starting in the 8th census of 1860 through the decision to survey U.S. inhabitants “born in the former kingdom of Poland” in order to delimit the boundaries of the reference group constituted in the host country by a Polish community apprehended in its ethno-linguistic dimension (even when the Polish state had disappeared) result from cultural and institutional factors that were specifically American, or was the dynamic at work partially autonomous and potentially effective in other contexts? That matter is left up in the air. No doubt it is difficult to answer solely by historical methods, although this book on its own subject offers the best illustration of what it can do.

First Published in Translated by Susan Emanuel with the support of Foundation Maison des Sciences de l’Homme.

by Daniel Sabbagh, 17 December 2010

Further reading

  • Anderson, Margo (1988). The American Census: A Social History. New Haven: Yale

University Press.

  • Bénichou , Sarah, Ioanna Kohler & Daniel Sabbagh (2009). Égalité de traitement dansl’emploi: les enseignements de l’expérience américaine dans la lutte contre les discriminations fondées sur l’origine, report of the program “Égalité des chances” by the French American Foundation, mars:
  • Bourdieu, Pierre et Loïc Wacquant (1998). « Sur les ruses de la raison impérialiste ». Actes de la recherche en sciences sociales 121-122: 109-118.
  • Brubaker, Rogers, Mara Loveman et Peter Stamatov (2004). « Ethnicity as Cognition ».

Theory and Society 33 (1), 2004: 31-64.

  • Dominguez, Virginia (1998). « Exporting U.S. Concepts of Race: Are There Limits to the

U.S. Model?”. Social Research 65 (2): 369-399.

  • Ford, Christopher (1994). « Administering Identity: The Determination of “Race” in Race Conscious Law ». California Law Review 82 (5): 1231-1285.
  • Gross, Ariela (2008). What Blood Won’t Tell: A History of Race on Trial in America. Cambridge (Mass.): Harvard University Press.
  • Harris, Marvin (1964). Patterns of Race in the Americas. New York: Norton.
  • Hattam, Victoria. (2007). In the Shadow of Race: Jews, Latinos, and Immigrant Politics in the

United States. Chicago: University of Chicago Press.

  • Hochschild, Jennifer et Brenna Marea Powell (2008). « Racial Reorganization and the United States Census 1850–1930: Mulattoes, Half-Breeds, Mixed Parentage, Hindoos, and the Mexican Race ». Studies in American Political Development 22: 59-96.
  • Johnson, Walter (2000). « The Slave Trader, the White Slave, and the Politics of Racial

Determination in the 1850s ». Journal of American History 87 (1): 13-38.

  • King, Desmond et Rogers M. Smith (2005). « Racial Orders in American Political

Development ». American Political Science Review 99 (1): 75-92.

  • Loveman, Mara et Jeronimo O. Muniz (2007). « How Puerto Rico Became White: Boundary

Dynamics and Intercensus Racial Reclassification ». American Sociological Review 76 (2): 915-939.

  • Ngai, Mae M. (2005). Impossible Subjects: Illegal Aliens and the Making of Modern America.

Princeton: Princeton University Press.

  • Nobles, Melissa (2000). Shades of Citizenship: Race and the Census in Modern Politics.

Stanford: Stanford University Press.

  • Perlmann, Joel, and Mary Waters (eds) (2002). The New Race Question: How the Census

Counts Multiracial Individuals. New York: Russell Sage Foundation.

  • Simon, Patrick et Patrick Weil (2008). « Sortir de l’inaction contre les discriminations »,

Libération, 4 juin.

  • Skerry, Peter (2000). Counting on the Census? Race, Group Identity, and the Evasion of

Politics. Washington, D.C.: Brookings Institution.

  • Tehranian, John (2000). « Performing Whiteness: Naturalization Litigation and the

Construction of Racial Identity ». Yale Law Journal 109 (4): 817-848.

  • Yanow, Dvora (2003). Constructing « Race » and « Ethnicity » in America: Category-Making

in Public Policy and Administration. New York: M.E. Sharpe.

To quote this article :

Daniel Sabbagh, « The Construction of Racial Identity in the United States », Books and Ideas , 17 December 2010. ISSN : 2105-3030. URL :

Nota Bene:

If you want to discuss this essay further, you can send a proposal to the editorial team (redaction at We will get back to you as soon as possible.


[1The periodization is justified by both the existence of a large number of excellent studies devoted to the system of ethno-racial classification in the post-segregationist era (Ford (1994); Skerry (2000); Perlmann & Waters (2002)) and, concerning the 1940s and 1950s, by the tendency to depoliticize the related stakes and the muting of their polemical dimension, notably due to the restriction on immigration and the greater attention given to economic statistics in the context of the New Deal and its sequels.

[2An option then used by only 2.6% of those surveyed.

[3And up until 1913, date of the establishment of a direct federal tax on revenue thanks to the Sixteenth Amendment.

[4Neither the excellent “social history of the American census” by Margo Anderson (1988) – whose purpose is much more general – nor the more recent and more specialized works by political scientists Melissa Nobles (2000) and Jennifer Hochschild & Brenna Marea Powell (2008) reserve as detailed a treatment of the elaboration and evolution of ethnic and racial categories, as this subject requires.

[5In the census the distinction established according to “color” or “race” (1790) anticipates by thirty years that made between Americans and foreigners (1820), by sixty years the recording of the country of birth of inhabitants (1850), by ninety years the country of birth of their parents (1880), and by 120 years the introduction of a question on the maternal language of residents (1910).

[6To take only one example, the question about profession was introduced only in 1820.

[7The principle of hypodescendence prescribed the assignment of a child from a mixed couple to the subordinate group rather than to the dominant group. On this point, see the pioneering work by the anthropologist Marvin Harris (1964).

[8This was in a context where mulattos appeared as the residue of the “sins committed by white men with black women, slaves rather than free, frequently in the form of an adulterous relation obtained under constraint”. Thus the “predicting [of] their extinction” was also “wishing to erase the visible trace of these relations that were purportedly censured” (p. 86).

[9This was the 1900 Census.

[10For the Puerto Rican case, see the excellent article by Loveman and Muniz (2007), whose conclusions are convergent in this respect.

[11On this point, see Dominguez (1998) and (in a more polemical vein) Bourdieu & Wacquant (1998).

[12Simon & Weil (2008); Bénichou, Kohler & Sabbagh (2009); COMEDD (2010).

[13Presidency of COMEDD was given to François Héran, who is therefore the principal author of the report.

Our partners

© - Any replication forbidden without the explicit consent of the editors. - Mentions légales - webdesign : Abel Poucet