Terms of Belonging: Are Models of Membership Self-Fulfilling Prophecies?

Georgetown Immigration Law Journal , 13, 1 (Fall 1998): 1-24.

TERMS OF BELONGING: ARE MODELS OF MEMBERSHIP

SELF-FULFILLING PROPHECIES?

T. ALEXANDER ALEINIKOFF AND RUBÉN G. RUMBAUT*

“If men define situations as real, they are real in their consequences.” —W. I. Thomas (1928)1

“We respond not only to the objective features of a situation, but also, and at times primarily, to the meaning this situation has for us . . . Certain kinds of definitions of the situation—we focus on the important class of public prophecies, beliefs, and expectations—become an integral part of the situation and thus affect subsequent developments . . . It is the social or public self-fulfilling prophecy that goes far toward explaining the dynamics of ethnic and racial conflict in the America of today.”

—Robert K. Merton, THE SELF-FULFILLING PROPHECY (1948)2

I. INTRODUCTION It is traditional, at least in the United States, to see immigrant adaptation as a straight-line process in both political and social terms: non-members of a nation-state residing outside state borders gain entry, take up residence, and (usually) seek and (frequently) obtain full membership in the state. As a narrative of legal status, the acquisition of full membership is the transition from alien to citizen, from stranger to rights-holder, from foreigner to governor. As a narrative of social belonging, the story is one of (more or less effective) integration or assimilation over time; a transition from out-group to in-group which is expected to be completed within a span of two or three generations. These versions of the national narrative, to be sure, fit the historical experience of some ethnic and racial groups more than that of others, and there are egregious exceptions to the normative tale. Different groups’ frames of remembrance and retellings of their past—their definitions of the situation—tell

* Paper prepared for the Social Science Research Council’s conference on “Immigrant Political Incorporation in Europe and the United States: Public Debates and Social Science,” May 15-17, 1998, Berlin, Germany. 1 WILLIAM I. THOMAS & DOROTHY SWAINE THOMAS, THE CHILD IN AMERICA 572 (1928). 2 ROBERT K. MERTON, THE SELF-FULFILLING PROPHECY (1948), reprinted in ON SOCIAL STRUCTURE AND SCIENCE 183, 184-86 (Piotr Sztompka ed., 1996).

2 GEORGETOWN IMMIGRATION LAW JOURNAL [Vol 13:0000

much about who is included and excluded in the national narrative, and hence about the society’s terms of belonging. We write as a lawyer and a sociologist interested in the relationship of our disciplines’ perspectives on these narratives. Of course, the legal and social stories need not follow the same script. Non-citizens regularly carry on daily life in the United States much as citizens, irrespective of status (even illegal status); and citizenship is not necessarily a marker of commitment, loyalty, or socialization. But there are clearly linkages between the political and the social realms. We seek here to explore those linkages by examining United States models of membership. We will argue that the political model of membership that had prevailed in the United States in the latter half of this century may be undergoing change, and we identify the prospects and perceptions of assimilation as one of the causes. Simply put, we argue that the way people are invited or welcomed to become members of the society influences their joining behavior which, in turn, influences how the society invites others to join it. It is in this self-reflexive story of belonging and incorporation that law and sociology meet.

II. MEMBERSHIP AND ASSIMILATION: SOME HYPOTHESES At first glance, the relationship of models of membership and assimilation may appear straightforward. A welcoming polity that is open to immigration permits immigrants to obtain citizenship rather easily, and it asks that immigrants undergo little change in the process of becoming citizens. This polity is likely to foster an immigrant population that seeks to belong. The receptive attitude among immigrants would then provide support for continuation of the receptive attitude of the polity. From this perspective, an inclusionary model of membership promotes assimilation which confirms the inclusionary model. The logic of this process would also demonstrate the logic of its converse: an unwelcoming polity that makes entry and attainment of citizenship difficult and asks that would-be members change dramatically to conform to the norms of their new home is likely to produce an immigrant population less willing to choose to conform, and such behavior may then affirm for the polity its wisdom in viewing immigrants as persons in need of reform. Here, an exclusionary model of membership hinders assimilation which confirms the exclusionary model. But life is not so simple. Consider some mundane but well-known examples. First, some inclusionary practices yield relatively weak affiliations: associations that may be joined voluntarily—the PTA, a church, a union—may

1999] TERMS OF BELONGING 3

inculcate little loyalty from members. By making membership open to many (or all), by providing little in the way of benefits, and by not asking much of new members, these associations may attain relatively low levels of commitment. Second, some exclusionary practices may yield strong affiliations: membership in an exclusive club—one that is very selective, demands a great deal from its members, but provides substantial benefits—may promote a strong esprit de corps among the chosen few. Thus, hypothetically, a nation-state that allows relatively little immigration and makes citizenship difficult, but promises significant privileges to citizens which are not granted to non-citizens, might promote a strong desire to belong and a willingness of newcomers to undergo significant change. (Perhaps the analogy to some forms of religious conversion is not too far off the mark.) It is, in short, difficult to predict the relationship of membership models and assimilation in the abstract. The outcomes will depend on a range of historical and ideological variables that are likely to vary state by state. We therefore turn our attention to the American experience to see which, if any, of several alternatives provide an adequate description.

III. MEMBERSHIP AMERICAN STYLE The United States conceives of itself as a “nation of immigrants.” But it is, in fact, largely a “nation of citizens.” While the percentage of foreign-born persons in the United States population is about 10%3 and immigration constitutes more than one-third of annual population growth, the vast majority of persons residing in the United States are citizens who attained citizenship at birth. The rule of jus soli (citizenship of the place where one is born) is the fundamental United States membership principle.4 For native-born citizens, acculturation or “cultural assimilation,” as that term is commonly used, is not generally an issue. Nearly all native-born Americans (and for that matter also immigrants who came to the United States as young children) speak English as a first language, and they are fully socialized into the dominant culture. However, such processes are “segmented” and take

3 Dianne Schmidley & Herman A. Alvarado, CENSUS BUREAU, U.S. DEP’T OF COMMERCE, CURRENT POPULATION REPORTS, THE FOREIGN BORN POPULATION IN THE UNITED STATES: MARCH 1997 (UPDATE). 4 Of course, this was not the rule initially applied to some native-born groups. See Elk v. Wilkins, 112 U.S. 94 (1884) (holding that a child born within Indian tribe is not a citizen at birth); Dred Scott v. Sandford, 60 U.S. 393 (1856) (holding that native-born free blacks are not citizens). But see United States v. Wong Kim Ark, 169 U.S. 649 (1898) (holding that children born to Chinese immigrant in United States are citizens at birth, despite the ineligibility of their parents for naturalization).


different forms depending on the sector of American society to which one is “assimilating.” More precisely, segmented assimilation processes are adaptations that take place within specifiable opportunity structures and through the influence of differential associations, reference groups, experiences and attachments, especially in primary social relationships stratified by race, religion, region, and class.5 Thus, although immigration provides only a relatively small portion of new members, it raises most of the questions about assimilation. Indeed, by “membership model” we mean a polity’s understanding and treatment of members vis-à-vis non-members.6 The “nation of immigrants” slogan must be placed alongside a history of race-based immigration and citizenship policies.7 The first naturalization statute, enacted in 1790, permitted only “white” persons to naturalize.8 In the aftermath of the Civil War, the statute was broadened to include “persons of African nativity, or African descent,” but still excluded immigrants who were neither white nor black.9 In addition, the original native inhabitants of the continent—doubly misnomered “American Indians”—were presumed to be “loyal to their tribes” and not granted United States citizenship until 1924.10 It was not until 1952 that the United States’ naturalization law became fully race-neutral.11 The first major federal immigration statutes, adopted in the 1880s and 1890s,

5 See Rubén G. Rumbaut, Assimilation and Its Discontents: Between Rhetoric and Reality, 31 INT’L MIGRATION REV. 923 (1997). 6 To be sure, persons may be nominal citizens but not full members, either because they are denied basic membership rights (such as voting) or suffer discrimination. Thus, a complete description of membership models would examine differential treatment among classes of citizens. However, since that discussion would not invoke concepts of assimilation, we leave it for another day. 7 United States citizenship and immigration policies have also included, from the start, incidents of gender discrimination. The first citizenship statute permitted the transmission of citizenship to children born overseas to United States fathers but not to United States mothers. This policy was not changed until 1934. See Wauchope v. United States Dep’t of State, 985 F.2d 1407 (9th Cir. 1993). And in the early decades of the twentieth century federal law provided that a United States citizen woman who married a foreigner lost her United States citizenship for the duration of the marriage. The statute was upheld in MacKenzie v. Hare, 239 U.S. 299 (1915). See generally CANDICE LEWIS BREDBENNER, A NATIONALITY OF HER OWN (1998). Justice Ruth Bader Ginsburg details at length gender discrimination in United States citizenship law in her dissenting opinion in Miller v. Albright, 118 S.Ct. 1428, 1449 (1998) (Ginsburg, J., dissenting). 8 See IAN F. HANEY LÓPEZ, WHITE BY LAW: THE LEGAL CONSTRUCTION OF RACE 42 (1996) (citing Act of Mar. 26, 1790, ch. 3, 1 Stat. 103). 9 Id. at 43-44, (citing Act of July 14, 1870, ch. 255, § 7, 16 Stat. 254). 10 Id. at 41 (citing Act of June 2, 1924, ch. 233, 43 Stat. 253). 11 See id. at 46 (citing Immigration and Nationality Act of 1952, ch. 2, § 311, 66 Stat. 239 (codified as amended at 8 U.S.C. 1422 (1988)).


expressly excluded Chinese laborers from entering the United States,12 and a similar “Gentleman’s Agreement” with Japan in 1907-08 largely barred Japanese immigration.13 The “National Origins Quota” system, in place from the 1920s until 1965, effectively limited immigration from Asia and southern and eastern Europe on explicit national origin and racial grounds.14 This race-based model of membership and citizenship lasted longer than the race-neutral model that has replaced it. The latter is largely a product of the post-World War II civil rights era.15 The current United States model, which has taken form over the past four decades, is more inclusionary. Its basic elements include:

• birthright citizenship; • relatively easy (non-race based) naturalization rules; • general parity between citizens and aliens in terms of access to

social benefits; • constitutional restrictions on state discrimination against aliens (other

than in the political sphere); • few restrictions on work opportunities for permanent resident aliens; • exit and re-entry rights for permanent resident aliens (provided they

do not stay outside the United States in excess of one year); • protection for immigrants under most federal labor, health, and

safety regulations; • recognition that immigrants possess most constitutional rights

afforded to citizens (such as the right of free speech and religion, and rights provided to criminal defendants);

• protection of immigrants under federal anti-discrimination laws prohibiting discrimination on grounds of national origin, race and citizenship status.

A decade ago, Peter Schuck pithily summarized this inclusionary model by noting that United States citizenship was “notably easy to obtain, difficult to lose, and confer[red] few legal or economic advantages over the status of

12 Id. at 37-38 (citing Chinese Exclusion Act, ch. 126, 22 Stat. 58 (1882); Act of July 9, 1884, ch. 220, 23 Stat. 115; Act of May 5, 1892, ch. 60, 27 Stat. 25; Act of April 29, 1902, ch. 641, 32 Stat. 176; Act of April 27, 1904, ch. 1630, 33 Stat. 428). 13 See Kevin R. Johnson, Race, the Immigration Laws, and Domestic Race Relations: A “Magic Mirror” into the Heart of Darkness, 73 IND. L.J. 1111, 1121 (1998). 14 See HANEY LÓPEZ, supra note 8, at 37-38. 15 It is not a coincidence that the statute repealing the “National Origin Quota” system was enacted in 1965, the heyday of civil rights legislation.


permanent resident alien.”16 To a large degree, this model still dominates, and it has been accompanied by strong expectations of assimilation. But there appears to have been a shift in the wind in recent years, with storm clouds on the horizon and already a non-trivial amount of rain.

IV. MEMBERSHIP MODELS AND “ASSIMILATION ANXIETY” The wind comes from the West. California is home to 40% of the undocumented population in the United States.17 Its population is about one quarter foreign-born,18 and it is the destination of over 20% of all permanent resident aliens each year.19 More aliens who legalized under the amnesty programs of the 1986 immigration law live in California than in any other state.20 Indeed, fully one-third of the “immigrant stock” population of the United States resides in California, with the concentrations being denser still within metropolitan areas.21 Three of the six top primary metropolitan areas (or “PMSAs”) in the United States, ranked based on the size of their immigrant-stock populations, are in California. In Los Angeles County in 1997, 62% of the area’s 9.5 million people were of immigrant stock, as were 54% of Orange County’s 2.8 million residents, and 43% of San Diego’s population of 2.7 million.22

16 Peter H. Schuck, Membership in the Liberal Polity: The Devaluation of American Citizenship, 3 GEO. IMMIGR. L.J. 1, 1 (1989). 17 Roughly two million out of five million. See I.N.S., U.S. DEP’T OF JUSTICE, 1995 STAT . Y.B. 183. 18 Rubén G. Rumbaut, Origins and Destinies: Immigration to the United States Since World War II, 9 SOCIOLOGICAL FORUM 583, 601 (1994). 19 See I.N.S., U.S. DEP’T OF JUSTICE, Immigrants Admitted by State of Intended Residence Fiscal Years 1987-95, 1995 STAT . Y.B. 65 (Mar. 1997). 20 I.N.S., U.S. DEP’T OF JUSTICE, IMMIGRATION REFORM AND CONTROL ACT : REPORT ON THE LEGALIZED ALIEN POPULATION 18 (Mar. 1992). 21 The “immigrant stock” of the United States today numbers about 55 million people—that is, persons who are either immigrants (26.8 million) or U.S.-born children of immigrants (27.8 million). That figure—one fifth of the national total—does not include 2.8 million others residing in the 50 states who were born (as were their parents) in Puerto Rico or other United States territories, nor the even larger number who reside in Puerto Rico and the other territories. If today’s “immigrant stock” formed a country, it would rank in the top 10% in the world in population size—about twice the size of Canada, and roughly the size of the United Kingdom, France, or Italy. See Rubén G. Rumbaut, Transformations: The Post-Immigrant Generation in an Age of Diversity 2 (Mar. 21, 1998) (unpublished manuscript on file with the author). 22 Id. The other three largest PMSAs in the size of their “immigrant stock” population are New York, Chicago, and Miami. See id. at Table 1.


In 1994, the electorate of California adopted Proposition 187, the so-called “Save Our State” initiative, by a wide margin.23 The purpose of Proposition 187 was to rid California of undocumented aliens. It declared their ineligibility for state-funded benefit programs, sought to deny undocumented children access to public schools, and required state medical and educational facilities to verify the immigration status of their client populations by telling persons suspected of being unlawfully in the United States to either seek legal status or “go home.”24 Directed at undocumented aliens, the benefits provisions of the Proposition were largely consistent with the prevailing membership model—except for the denial of education to school-aged children, which was directly contrary to a 1982 decision of the Supreme Court.25 But the State’s attempt to effectively adopt its own immigration enforcement strategy (based on its perception of the federal government’s failure to control the borders) was a new intervention in the immigration debate.26 A federal court enjoined implementation of the Proposition primarily on the ground that it interfered with federal regulation of immigration.27 Proposition 187 made clear that the long-standing toleration of (and indeed demand for) undocumented workers in California (and elsewhere) did not apply to family members who needed medical care, attended schools, and sought other social benefits. That is, there was a clear financial aspect to the support for the Proposition. But it also reflected a deeper set of concerns about immigration in general. “Saving the State” meant not only preserving tax dollars for citizens and lawful residents; it also meant preserving California in racial and cultural terms.28 It was regularly noted that shortly after the year 2000 California would become a “majority minority” state—that persons of Latino, African-American and Asian descent would comprise a majority of the state’s

23 Paul Feldman & Patrick J. McDonnell, Prop. 187 Backers Elated—Challenges Imminent, L.A. TIMES, Nov. 9, 1994, at A1. 24 For a description of Proposition 187’s sections, see League of United Latin American Citizens v. Wilson, 908 F. Supp. 755, 764-65 (C.D. Cal. 1995), amended by 997 F. Supp. 1244 (C.D. Cal. 1997). 25 See Plyler v. Doe, 457 U.S. 202 (1982). 26 States controlled entry and exit in the mid-nineteenth century until such laws were struck down by the Supreme Court. See Gerald L. Neuman, The Lost Century of American Immigration Law (1776-1875), 93 COLUM. L. REV. 1833 (1993). 27 See League of United Latin American Citizens v. Wilson, 908 F. Supp. 755 (C.D. Cal. 1995), amended by 997 F. Supp. 1244 (C.D. Cal. 1997). 28 Johnson, supra note 13, at 1134-37.


population.29 Immigration was recognized as a crucial component in that demographic story.30 Demography may not be destiny, but to many Californians, it represents drift and disunity. A state without a white majority is perceived as a multicultural state, a state where the project of assimilation has failed. There is no a priori reason why a “majority minority” state could not be culturally united; the “melting” of white ethnics in the first half of this century demonstrates the possibilities of E Pluribus Unum. But many Californians have apparently lost faith in the melting pot, and putatively are increasingly concerned about the prospects for assimilation of the large number of newcomers. It is hard not to suspect race and racial privilege as the motivating factors in white majority perceptions. These perceptions are not limited to the West Coast. A public opinion poll conducted in the summer of 1997 asked a representative number of Americans the following question:

How concerned are you that the growing number and nationalities of immigrants will . . . lead to increased conflict between racial and ethnic groups: very concerned, somewhat concerned, not too concerned, not at all concerned?

Nearly two-thirds of the respondents answered “very concerned” or “somewhat concerned.” These pessimistic views were shared by Whites (65% of whom were within these two categories), Blacks (61%) and Hispanics (59%).31 The conclusion that seems hard to resist is that immigration is

29 Dorothy E. Roberts, Who May Give Birth to Citizens? , in IMMIGRANTS OUT! 205, 209 (Juan F. Perea ed., 1997). 30 Kevin R. Johnson, An Essay on Immigration Politics, Popular Democracy, and California's Proposition 187: The Political Relevance and Legal Irrelevance of Race, 70 WASH. L. REV. 629, 650-51 (1995); see Roberts, supra note 29, at 205-19. 31 Columbus Day Survey: Looking For America, Question 049, Public Opinion Online, Aug. 1997, available in LEXIS, Market Library, RPOLL file. Also high on the “concern” scale was the view that increasing immigration will “overburden the welfare system and raise taxes;” 79% were very concerned or somewhat concerned (including 67% of Hispanics recorded as being “very concerned” and 12% as “somewhat concerned”). See id. at Question 052. Less concern was expressed over whites becoming a minority: when asked the question “How concerned are you that [increased numbers and nationalities of immigrants] will lead to whites of European descent no longer being in the majority,” only 36% were very concerned or somewhat concerned. See id. at Question 050. This result may simply reflect the (correct) view that whites will in fact remain the majority in the United States as a whole for many years to come.


perceived as contributing to multiculturalism and adding fuel to the fires of cultural conflict. It should not be surprising that concerns about the viability of the assimilation project will produce demands that immigration be reduced.32 However, the point we wish to stress here is that such concerns appear to be influencing the underlying model of membership as well. We see evidence of this in several recent policy initiatives. First is the suggestion that the fundamental membership principle—jus soli citizenship—be changed so that the children of undocumented aliens not be deemed citizens at birth in the United States. Until recently, it had always been assumed that the Fourteenth Amendment’s definition of birthright citizenship (that “all persons born . . . in the United States . . . are citizens of the United States”33) applied to the children born to aliens unlawfully in the United States. But a book written by two Yale University professors in 1985 has opened up public debate on the matter. In Citizenship Without Consent, Peter Schuck and Rogers Smith argued that the Fourteenth Amendment could be read as encompassing a principle of “consent” that endowed Congress with the power to decide whether or not the children of undocumented aliens born in the United States should be citizens.34 Although the authors’ interpretation of the Fourteenth Amendment and reading of constitutional history have been roundly criticized,35 the thesis of the book has occasioned a congressional hearing on the topic and a number of statutory proposals.36 Second, the 1996 welfare law constituted a major shift in long-standing premises about social membership in the United States. Since the creation of the Great Society welfare programs, permanent resident aliens had been included in most social benefits programs. In 1996 Congress acted to exclude immigrants from federally funded programs and authorized states to exclude them from state-funded programs.37 Most harshly, the statute terminated aid to

32 See PETER BRIMELOW , ALIEN NATION (1995). 33 U.S. CONST . amend. XIV, § 1. 34 See generally PETER H. SCHUCK & ROGERS M. SMITH, CITIZENSHIP WITHOUT CONSENT (1985). 35 See Gerald L. Neuman, Back to Dred Scott?, 24 SAN DIEGO L. REV. 485 (1987) (book review); David A. Martin , Membership and Consent: Abstract or Organic?, 11 YALE J. INT’L L. 278 (1985) (book review). 36 See, e.g., Societal and Legal Issues Surrounding Children Born in the United States to Illegal Alien Parents, Joint Hearing Before the Subcomm. on Immigration and Claims and Subcomm. on the Constitution, House Comm. on the Judiciary, 104th Cong., 1st Sess. 103 (1995) (Statement of Prof. Gerald L. Neuman, Columbia University Law School, discussing eight bills). 37 Balanced Budget Act of 1997, Pub. L. No. 105-33, § 5301, 111 Stat 251, 597-98. The constitutionality of these measures will have to be resolved by the courts. The Supreme


elderly and disabled immigrants then on the rolls—an action subsequently reversed in part.38 The motivation was financial, but the justificatory rhetoric was membership-based: those in support of the measure argued that Congress had the authority—if not the duty—to offer benefits first and foremost to full members (citizens).39 Third, the 1996 immigration legislation dramatically restructured procedures for determining the admissibility and deportability of immigrants. Proceedings were streamlined, judicial review of administrative action was severely limited, and relief from removal was significantly restricted.40 Much of the congressional action was predicated on the desire to speed the removal of aliens convicted of criminal offenses and also to prevent dilatory practices of lawyers. But the toughness of the laws ought also to be seen as part of a redefinition of social membership. Arguably, long-term resident aliens function virtually as citizens in the American social and economic system. Yet the immigration laws—particularly as amended by the 1996 legislation—treat them, in effect, as considerably less than full members: there is no “statute of limitations” on deportability grounds, and the 1996 laws rendered various avenues of relief from removal far more difficult to obtain. Finally, controversy over dual nationality appears to be just over the horizon. An increasing number of foreign states have adopted rules permitting their nationals to retain citizenship despite naturalization elsewhere—and despite the United States’ oath of naturalization, requiring renunciation of all “allegiance and fidelity to any foreign prince, potentate, state or sovereignty.”41 In 1997 Mexico amended its constitution to this effect, and the changes became effective in March, 1998. These amendments have raised in some American minds the specter of divided loyalty and dual voting by naturalized Mexican-American citizens.42 We find it likely that the trend toward increasing recognition of dual

Court has recognized broad congressional authority to treat citizens and aliens differently. See Mathews v. Diaz, 426 U.S. 67 (1976). However, the Court has generally struck down state laws that deny benefits to immigrants. See Graham v. Richardson, 403 U.S. 365 (1971). Whether Congress has the power to authorize states to discriminate against aliens remains to be seen. 38 Balanced Budget Act of 1997, § 5301; Agricultural Research, Extension, and Education Reform Act of 1998, Pub. L. No. 105-185, § 504, § 506, 112 Stat. 523, 578-79. 39 See Aleinikoff, The Tightening Circle of Membership, in IMMIGRANTS OUT!, supra note 29, at 324. 40 See generally Antiterrorism and Effective Death Penalty Act of 1996, Pub. L. No. 104-132, 110 Stat. 1214; Illegal Immigration Reform and Immigrant Responsibility Act of 1996, Pub. L. No. 104-208, 110 Stat. 3009. 41 Immigration and Nationality Act § 337(a), 8 U.S.C. § 1448(a) (1994). 42 See, e.g., Mark Fritz, Pledging Multiple Allegiances: A Global Blurring of Boundaries Challenges Notions of Nationality, L.A. TIMES, Apr. 6, 1998, at A3; Sam


nationality will strengthen demands that American citizenship be unitary and meaningful—demands that can be seen to be met by drawing new distinctions between citizens and aliens. Together, these elements indicate significant pressure on the inclusionary model of membership. The United States seems to be in the process of redefining its model of membership as more closely linked to citizenship, and that redefinition of the situation seems driven, to a significant degree, by “assimilation anxiety,” to use a phrase now in vogue in public debates.43

V. ASSIMILATION: LANGUAGE AND IDENTITY We have argued that assimilation fears may be causing a rethinking of American membership models. In the section, we examine whether those fears have any basis in fact. We conclude that they do not. A review of the social science research literature on immigration reveals that assimilation—whether considered intergenerationally or among the most recent waves of immigrants—appears to be progressing roughly as it always has.44 This is particularly the case with respect to linguistic assimilation, which is frequently seen as the most important marker of acculturation. An analysis of the 1990 United States Census, summarized in Table 1, makes that point abundantly clear. The data are for the 19.5 million foreign-born persons counted in the 1990 census who were five years or older, and they show a familiar story: without exception, English proficiency increases over time in the United States, and especially rapidly among younger immigrants. Overall, a majority (53%) of all immigrants in 1990 spoke English only or very well, ranging from 25% of recently arrived immigrants 55 years or older (those who had come in the 1980s) to 90% of those who had immigrated as children three decades before.

Howe Verhovek, Torn Between Nations, Mexican-Americans Can Have Both, N.Y. TIMES, Apr. 14, 1998, at A1. 43 More precisely, the phrase is meant to denote anxiety about a presumed lack of assimilation among recent immigrants. For some, the anxiety may run the other way: that rapid assimilation is erasing cultural practices, languages, and identities. For a critique of the gap between the rhetoric and the realities of assimilation, see Rubén G. Rumbaut, Paradoxes (and Orthodoxies) of Assimilation, 40 SOC. PERSP . 483 (1997). 44 For intergenerational analyses of the assimilation of European-origin immigrants and their descendants, see, e.g., Richard D. Alba, Assimilation’s Quiet Tide, 119 PUB. INTEREST 3 (1995); Lisa J. Nidert & Reynolds Farley, Assimilation in the United States: An Analysis of Ethnic and Generation Differences in Status and Achievement, 50 AM. SOC. REV. 723, 840 (1985); Andrew M. Greeley, The Ethnic Miracle, 45 PUB. INTEREST 20 (1976). For information on the incorporation of contemporary immigrants, see generally ALEJANDRO PORTES & RUBÉN G. RUMBAUT , IMMIGRANT AMERICA: A PORTRAIT (2d ed. 1996).


Table 1.

Percent of Immigrants Who Speak English Only or Very Well,By Age and Decade of Immigration to the United States, 1990

(N=19.5 million foreign-born persons age 5 years or older, 1990 U.S. Census)

Decade of immigration to U.S.Age in 1990

TOTAL5-17 years 18 to 34 years 35-54 years 55 and older

1980-1990

1970-1979

1960-1969

Before 1960

50

79

--

--

39

62

87

90

36

47

64

85

25

31

42

72

39

53

63

75

Source: 1990 U.S. Census, 5% Public Use Microdata Sample (PUMS).

Equally compelling are the broad-brush data on immigrants and their American-born children presented in Table 2, drawn from the merged 1996 and 1997 Current Population Surveys of the United States Census Bureau. The data are representative of the immigrant stock population of the United States, estimated at nearly 55 million people in 1997—including nearly 27 million foreign-born persons (the first generation) and nearly 28 million U.S.-born persons with at least one foreign-born parent (the second generation). Over time in the United States, the proportion of immigrants who become naturalized United States citizens grows significantly, from a tenth of those who had come during the 1990s (most of whom were not yet eligible to apply for citizenship), to over a fourth of those arriving in the 1980s, over half of those coming in the 1970s, over two-thirds of the cohort of the 1960s, and 84% of those who had immigrated before 1960. Yet another kind of stake in the society is indicated by the proportion of immigrants who own their homes, which actually grows faster than citizenship over time in the United States, as the immigrants are incorporated not only culturally (as the language data indicate) but socioeconomically.


Table2.

Assimilation of Immigrants and TheirChildren:Citizenship, Homeownership, Poverty,

and Education by Generation and Decade of Immigration to theUnited States, 1997

Generation(1

st and2

nd)andDecade of immigration

to U.S.

Social and EconomicCharacteristics, 1997

Age(mean

years)

U.S.Citizen(%

)

OwnHome(%

)

PovertyRate(%

)

CollegeGraduat

es(%)

1

FirstGeneration

(foreign-born):

2

1990-19971980-19891970-19791960-1969

Before1960

SecondGeneration

(U.S.-born):

3

27.534.742.152.166.635.1

9

28

53

69

84

100

26

44

62

73

78

67

33

23

15

10

9

16

31

25

26

25

24

34

1 College graduation rate for adults 25 to64 years old.2 Foreign-born persons residing in the U.S. (estimated at over 26million in 1997).3 U.S.-born persons with at least one foreign-born parent (estimated at nearly28 million in 1997).

Source: Current Population Surveys (CPS), merged 1996 and 1997 (March) data In the remainder of this section, we explore the data on assimilation in more detail, focusing on relevant findings regarding linguistic assimilation and ethnic identification from the Children of Immigrants Longitudinal Study (“CILS”), the largest such survey to date in the United States. The study, carried out in Southern California (San Diego) and South Florida (Miami and Fort Lauderdale), followed a large sample of over 5,200 youths from immigrant families representing seventy-seven different nationalities from the junior high school grades through the end of high school in the 1990s.45 In California the

45 An earlier analysis of these issues focused only on the San Diego sample in the study. See id. We present here results from the complete CILS data set.


main nationalities were Mexicans, Filipinos, Vietnamese, Laotians, Cambodians, and smaller groups from Asia and Latin America; in South Florida the main groups included Cubans, Nicaraguans, Haitians, Jamaicans and others from the Caribbean and Latin America. The initial survey was completed in the Spring of 1992 when the students were enrolled in the eighth and ninth grades, in order to avoid the potential bias of differential dropout rates in the later high school years. The sample was evenly divided by sex and nativity (half of the youths were foreign-born, and the other half were U.S.-born with at least one foreign-born parent), with a mean age of 14.2 years. A follow-up survey which succeeded in re-interviewing 82% of the original sample, was completed during 1995-96, and a stratified sample of their parents was also interviewed. In addition, detailed academic information was obtained from the schools in the respective districts.46 A. Linguistic Assimilation A perennial controversy in public debates on immigration concerns the role of bilingual education and perceived threats to the continuing predominance of English as the common national language. In June 1998, California voters by a 61% to 39% margin approved Proposition 227, a popular new initiative dubbed “English for the Children,” which reflects this impulse: it aims to dismantle the state’s bilingual programs and require that all public school instruction be conducted in English. CILS data on language preference and proficiency provide some illuminating information in this regard. Over 90% of these children of immigrants reported speaking a language other than English at home, mostly with their parents. But at the time of the first survey in 1992 (T1) already 73% of the total sample preferred to speak English instead of their parents’ native tongue. This group includes 64% of the foreign-born youth and 81% of the U.S.-born. By the follow-up survey in 1995-96 (T2), the proportion who preferred English had swelled to 88%, including 83% of the foreign-born and 93% of the U.S.-born. The most mother-tongue-retentive group was the Mexican-origin youth living in a Spanish-named city on the Mexican border with a large Spanish-speaking immigrant population and a wide range of Spanish-language radio and television stations. However, even for this group the force of linguistic assimilation was incontrovertible: while at T1 only 32% of the Mexico-born children preferred English, by T2 that preference had doubled to 61%. Additionally, while just over half of the U.S.-born

46 For an overview of the CILS study, see THE NEW SECOND GENERATION (Alejandro Portes ed., 1996). See also Rubén G. Rumbaut, Coming of Age in Immigrant America, RES. PERSP . ON MIGRATION, Jan/Feb 1998, at 1.


Mexican-Americans in San Diego preferred English at T1, that proportion had jumped to four-fifths three years later. Even more decisively, among Cuban-origin youth attending both public and private schools in Miami, 95% of both the foreign-born and the native-born preferred English by T2. A main reason for this rapid language shift in use and preference has to do with increasing fluency in both spoken and written English relative to the level of fluency in the mother tongue. Respondents were asked to evaluate their ability to speak, understand, read and write in both English and the non-English mother tongue. The response format (identical to the item used in the United States census) ranged from “not at all” and “not well” to “well” and “very well.” Over three-fourths of the total sample at both T1 and T2 reported speaking English “very well,” compared to only about a third who reported an equivalent level of spoken fluency in the non-English language. Even among the foreign-born, 69% spoke English very well whereas only 41% spoke their native language just as well. The differences in reading fluency are much sharper still: the percentage of respondents who can read English “very well” is three times the percentage of those who can read a non-English language very well (78% to 24%). The ability to maintain a sound level of literacy in a language is nearly impossible to achieve in the absence of schools that teach it, and of a community that values it and in which it can be regularly practiced.47 As a consequence, the bilingualism of these children of immigrants becomes increasingly uneven and unstable. The CILS data vividly underscore the rapidity with which English triumphs and foreign languages atrophy in the United States—even in a border city like San Diego with the busiest international border crossing in the world, or in Miami, the metropolitan area with the highest percentage of foreign-born in the country. The second generation not only comes to speak, read and write English fluently, but prefers it overwhelmingly over their parents’ native tongue. It bears adding that these results have occurred while the youths were still residing as dependents in their parents’ home, where the non-English mother tongue retains primacy. Once they leave the parental fold to lead independent lives of their own, the degree of English language dominance and non-English language atrophy is almost certain to accelerate. This result is all the more certain among those living outside dense immigrant enclaves. This pattern of rapid linguistic assimilation is constant across nationalities and socioeconomic levels and suggests that, over time, the use of and fluency in

47 This is particularly true for languages with entirely different alphabets and rules of syntax and grammar, such as many of the Asian languages brought by immigrants to California.


foreign languages will inevitably decline. These results both confirm the broader results from the 1990 census depicted above, and directly rebut nativist alarms about the perpetuation of foreign-language enclaves in immigrant communities. The findings suggest that the linguistic outcomes for the third generation—the grandchildren of the present wave of immigrants—will be no different than what has been the age-old pattern in American history: the grandchildren may learn a few foreign words and phrases as a quaint vestige of their ancestry, but they will most likely grow up speaking English only. It is for this reason that the United States has been called a “language graveyard.” Seen in this light, initiatives such as “English for the Children” seem superfluous. They seem aimed less at pursuing the intended goal (teaching English) than at tightening the circle of membership. B. Identity and the “Boomerang Effect” Membership may be as much a state of mind—a definition of the situation—as a set of social facts. That is, people who are formally members may not perceive themselves as full members (for example, citizens who are members of oppressed minority groups); and persons formally excluded from membership may feel and act as if they are full members (for example, long-term permanent resident aliens with a United States citizen spouse and United States citizen children). Moreover, the question of “belonging” in the United States is complicated by this nation’s pluralism. One can identify with a minority group of United States citizens that both perceives itself as distinct from the majority and is so perceived and treated by the majority. Perceptions of belonging, then, are influenced by a host of factors, including social realities, favorable or unfavorable reception by full members, and racial and ethnic identifications. If a sense of belonging is important to assimilation, then one way to seek a measure of that factor is to ask persons about their self-identifications, as was done in the CILS surveys. The way that these young people define themselves is significant, revealing how and where they perceive themselves to “fit” in the society of which they are its newest members. Self-identities and ethnic loyalties can often influence long-term patterns of behavior and outlook, and may also reflect social dynamics weighed with significant long-term political implications.

The CILS surveys included an open-ended question that sought to ascertain the respondent’s ethnic self-identity. Four main types of ethnic identities became apparent: (1) a plain “American” identity; (2) a hyphenated-American identity (e.g., Haitian-American, Vietnamese-American); (3) a


national-origin identity (e.g., Filipino, Cuban, Jamaican); and (4) a pan-ethnic minority group identity (e.g., Hispanic, Latino, Chicano, Asian, Black). In 1992, the largest proportion (41%) chose a hyphenated-American identification; 13% identified as plain “American”; a fourth of the sample identified by national origin; and 16% selected pan-ethnic minority identities. Birth in the United States made a great deal of difference in the type of identity selected. At T1, the foreign-born were four times more likely to identify by national origin than were the U.S.-born. Conversely, the U.S.-born were much more likely to identify as American or hyphenated-American than were the foreign-born. Those and related findings at T1 suggested an assimilative trend from one generation to another.48 That these ethnic self-identifications are patterned along a wide range of social characteristics is made clearer by the profile presented in Table 3. The main types of ethnic identities chosen in the follow-up survey in 1995 (T2) are profiled via an array of social, economic and cultural characteristics measured three years before in 1992 (T1), thus making clear and unambiguous the temporal ordering of the effects of variables hypothesized as potential predictors of identity choices made much later. Virtually without exception, the data in Table 3 show linear progressions, from lower to higher degrees of assimilation, along an identity continuum—from national-origin self-identifications to pan-ethnic, hyphenated and unhyphenated American identities. For example, 93% of all respondents identifying themselves as “American” were born in the United States, compared to 69% of the hyphenated-Americans, 54% of the pan-ethnic identities, and only 25% of those identifying by national origin. Respondents who had one United States-born parent significantly increased their odds of identificational assimilation.49 Citizenship appears to matter over and above nativity: becoming a United States citizen by naturalization adds to the influence of birthplace, perhaps signaling a stake in the society as a full-fledged member, legally as well as subjectively, with an accompanying shift in one’s frame of reference. It is not so much the length of time in the United States, but rather the nature of one’s sociopolitical status that seems to be more determinative of the psychology of identity.

48 For analysis of the determinants of ethnic self-identities based on cross-sectional data from the baseline survey, see Rubén G. Rumbaut, The Crucible Within: Ethnic Identity, Self-Esteem, and Segmented Assimilation Among Children of Immigrants, 28 INT’L MIGRATION REV. 748 (1994). See also Alejandro Portes and Dag MacLeod, What Shall I Call Myself?: Hispanic Identity Formation in the Second Generation, 19 ETHNIC AND RACIAL STUD., 523 (1996). 49 The proportion identifying as "American" who had one U.S.-born parent was ten times that of respondents identifying by national-origin, 42% to 4%.


Similarly, as Table 3 also shows, those identifying as “American” had the highest socioeconomic status, followed by those choosing the hyphenated-American, pan-ethnic, and national-origin identities (who reported the lowest SES). Race played a major role in these self-identifications: 55% of those identifying as “American” self-reported as “white,” compared to only 17% of the hyphenated-Americans, 16% of the pan-ethnics, and 6% of those identifying by national origin. The most acculturated were also the most likely to see themselves as “American,” while those experiencing and expecting discrimination the most were notably less likely to identify as “American.”


INSERT REVISED TABLE 3 HERE


To a considerable extent, the strong impression given by the data in Table 3 is of a generational movement. As one becomes increasingly distant from the original immigration experience and its ethos, one moves toward a greater identificational “Americanization” which is accompanied by upward socioeconomic mobility, increasing acculturation and linguistic assimilation, and decreasing experiences and expectations of discrimination. The “Hyphenated-American” identity, in Table 3, appears to occupy a middle position between the immigrant and the assimilative plain “American” identities. This intergenerational identity bridges two worlds, but may be as difficult to sustain as fluent bilingualism or other additive adaptations have proven to be in American life. In turn, the pan-ethnic identities reflect a different kind of “segmented assimilation”—an identification with American racial minorities in contexts in which class and race play a more determinative role than culture and language. Still, the data overall would lead one to expect greater change over time in a generally assimilative direction: that is, an increase in the proportion identifying as American, with or without the hyphen, and a decrease in the proportion identifying by national origin.

But the results of the 1995 survey (conducted in the months after the passage of Proposition 187 in California) turned conventional expectations on their head, as Table 4 demonstrates. The table presents data collected at both the T1 and T2 surveys, showing changes over time in ethnic self-identifications, experiences, and expectations of discrimination, and perceptions of American society—for the CILS sample as a whole, as well as for the major national-origin groups. At T1, a combined 54% had identified as American or hyphenated-American, while a combined 43% had identified by national origin or with a pan-ethnic label.50 But at T2 the proportion identifying as American plummeted from 13% to 3%, and hyphenates dropped from 41% to 31% of the total—a combined loss of twenty percentage points. Meanwhile, those identifying pan-ethnically increased from 16% to 27%, and national origin identifications increased from 27% to 35%—a combined gain of almost twenty percentage points. This is exactly the opposite of what would have been expected.

The San Diego and Miami stories diverge here. In Southern California, the biggest gainer by far in terms of the self-image of these youths was the foreign nationality identity, chosen by 32% at T1 but by 48% at T2. This shift took place among both the foreign-born and the U.S.-born, most notably among the two largest immigrant groups in the United States: the youth of Mexican and Filipino descent. Pan-ethnic

50 The rest, no more than 4% in either survey, offered other mixed identity responses.


Insert Table 4 Here


identities remained at 16% at T2, but that figure concealed a steep decline among Mexican-origin youth in “Hispanic” and “Chicano” self-identities, and a sharp upswing in the proportion of youths now identifying pan-ethnically as “Asian” or “Asian American,” especially among the smallest groups. The rapid decline of both the plain “American” (cut to less than 2% in San Diego) and hyphenated-American (dropping from 43% to 30%) self-identities points to the growth of a reactive ethnic consciousness. In South Florida, in contrast, the biggest gains overall were in pan-ethnic identities such as “Hispanic” and “Black,” doubling from 17% at T1 to 38% at T2, mainly among of the Latin Americans (except the Nicaraguans) and the Jamaicans, while the percentage identifying by national origin remained unchanged at about one-fifth. Plain American identity dropped sharply from 19% to less than 4% overall, and hyphenated-American identities fell to 30%. The Haitians were an interesting exception in Florida. Resembling the California pattern, the proportion of Haitians selecting a denationalized pan-ethnic identity decreased, while the proportions identifying as “Haitian” and “Haitian-American” increased notably. This response is reflective of the fact that the T2 response was given in the aftermath of the United States’ invasion of Haiti in the Fall of 1994, where, for a change, the interests of the United States government coincided with those of the Haitian émigrés. This instance again suggests the importance of the sociopolitical context in shaping ethnic self-identities. The Cubans, despite being a majority in Miami, by T2 were twice as likely to identify as “Hispanic” (30%) than as “Cuban” (15%), although nearly half (46%) identified as “Cuban-American.” Change over time, in any event, has not been toward assimilative mainstream identities (with or without a hyphen). Some groups (the Mexicans and Filipinos in California, the Haitians and Nicaraguans in Florida) have moved toward a more proudly nationalistic reaffirmation of the immigrant identity. Almost all others have moved toward pan-ethnic minority group identities, as these youths become increasingly aware of the ethnic and racial categories in which they are persistently classified by mainstream society. Surprisingly, this is among a sample of children of immigrants, only 13% of whom self-reported racially as “white.” In both of these variants, the results point to the rise of a “reactive ethnicity”51 and to a growing identification with United States minority groups. That does not mean that these youths exhibit a lack of acculturation—quite to the contrary, as the above data on language especially shows, the process of cultural and linguistic assimilation is probably proceeding more

51 See PORTES & RUMBAUT , supra note 44.


rapidly among these young cohorts than with any others in American immigration history. Nor does the finding of a return to a national-origin self-identity for some groups augur a deepening of transnational connections—which, were they fully developed, might prove mutually beneficial. On the contrary, the evidence shows a decreasing knowledge of, interest in, and attachment to the parental homeland, history and language.52 Instead, the process of growing ethnic awareness among the children of immigrants in the CILS sample appears to be a function of their experiences, expectations, and perceptions of racial and ethnic discrimination in their American present—and their response to societal messages that tell them that they are not, and may never become, full-fledged members. In the last survey, as the middle panel of Table 4 shows, reports of being discriminated against increased to 62% of the sample. Virtually every group reported more experiences of rejection or unfair treatment against themselves as they grew older, with the highest proportions found among the children of Afro-Caribbean and Asian immigrants, followed by Mexicans and other Latin Americans, and the lowest proportions among the Cuban youth in Miami. Among those reporting discrimination, their own race or nationality is overwhelmingly perceived to account for what triggers unfair treatment from others. Such experiences tend to be associated over time with a higher incidence of depressive symptoms, and with the development of a more pessimistic stance about their chances to reduce discriminatory treatment through higher educational achievement. In both surveys the students were asked to agree or disagree with the statement, “No matter how much education I get, people will still discriminate against me.” By T2, the proportion agreeing with the statement was 35%, including substantial majorities of black immigrants from the West Indies and Haiti. Still, it is important to underscore the fact that despite their growing awareness of the realities of American racism and intolerance, most of the youth in the sample (almost two-thirds) continued to affirm a confident belief in the promise of equal opportunity through educational achievement. Even more telling, as shown in the bottom panel of Table 4, 60% of these youths agreed in the T1 survey that “there is no better country to live in than the United States,” and that endorsement grew to 72% three years later—

52 In addition to the evidence already discussed showing a rapid erosion among these youth of the parental language and an increase in both preference for and proficiency in English, other CILS data (not shown) document a very high degree of ignorance of basic historical and geographical knowledge of the parental homeland, although the level of knowledge varies by national origin, with Cubans being most knowledgeable, and Southeast Asians least knowledgeable.


despite a growing anti-immigrant mood in the country during that period. The groups most likely to endorse that view were the children of political exiles from Cuba and Vietnam who found a favorable context of reception in the United States. The groups least likely to agree with that statement were also those who have most felt the weight of racial discrimination: the children of immigrants from Haiti, Jamaica and the West Indies. Milton Gordon’s assimilation sequence, it is well to recall, ultimately required routine social acceptance and an absence of prejudice and discrimination in the “core society.”53 It takes two to tango, after all—and to assimilate.54 But what goes around comes around. To the extent that a climate of intolerance and exclusion persists, these trends may portend potentially significant political alignments and commitments for these children in their adult years. In California, for instance, in boomerang fashion, real or imagined immigrant-bashing may provoke long-term opposition to politicians and political parties so perceived by the children of those immigrants. Similarly, in Florida, the politics of race may become the most salient factor in the adult second generation. Already among first-generation adults, the past few years have seen an unprecedented increase in the number and proportion of immigrants who have applied for naturalization and who have registered to vote—with some notable and unexpected consequences, as in the 1996 election of a Hispanic newcomer, Loretta Sanchez, over former California Representative Robert Dornan in one of the nation’s erstwhile most conservative congressional

53 In MILTON M. GORDON, ASSIMILATION IN AMERICAN LIFE: THE ROLE OF RACE, RELIGION, AND NATIONAL ORIGINS (1964), perhaps the most influential statement on the subject in American social science, Milton Gordon broke down the assimilation sequence into seven stages, of which “identificational assimilation,” such as the formation of a self-image as an unhyphenated American, constituted the end of a process that began with cultural assimilation, proceeded through structural assimilation and intermarriage, and was accompanied by an absence of prejudice and discrimination in the “core society.” Once structural assimilation occurred (i.e., extensive primary-level interaction with members of the “core group”), either in tandem with or subsequent to acculturation, “the remaining types of assimilation have all taken place like a row of ten pins bowled over in rapid succession by a well placed strike.” Id. at 81. 54 Of such inclusive terms of belonging—and what might be called paradoxes of acceptance—Ari Shavit has argued that as American Jews find acceptance and success under the more inclusive model of membership, they become “an endangered species:” “Curiously, it is precisely America’s virtues—its generosity, freedom and tolerance—that are now softly killing the last of the great Diasporas. It is because of its very virtues that America is in danger of becoming the most luxurious burial ground ever of Jewish cultural existence.” Ari Shavit, Vanishing, N.Y. TIMES, June 8, 1997, § 6, at 52. See also MARTIN MARTY, THE ONE AND THE MANY 90 (1997).


districts, Orange County.55 Whether and to what extent such future outcomes may be extrapolated from present trends remain open empirical questions.

VI. PERCEPTION, RECEPTION, AND THE ROLE OF SOCIAL SCIENCE In the preceding sections, we have suggested links between perceptions of assimilation and models of membership. While noting that the relationship might take a number of forms, we have tried to show that concerns about assimilation—based, at least in part, on race and national origin—may contribute to a shift to a less inclusionary membership model, and that this shift might well produce behavior that further complicates the integration of newcomers into American society. We have also tried to show that concerns about assimilation are not strongly borne out by the data. We are left then with a public policy conundrum: namely, that the formulation of public policy, although based on highly questionable assumptions, helps to produce the conditions that affirm the public policies. How do we break away from this misguided self-fulfilling prophecy? Robert Merton, in his seminal formulation of the concept, offered this advice as to “how the tragic, often vicious, circle of self-fulfilling prophecies can be broken”:

The initial definition of the situation which has set the circle in motion must be abandoned. Only when the originating assumption is questioned and a new definition of the situation introduced, does the consequent flow of events give the lie to the assumption. Only then does the belief no longer father the reality . . . These changes, and others of the same kind, do not occur automatically. The self-fulfilling prophecy, whereby fears are translated into reality, operates only in the absence of deliberate institutional controls. . .56

55 See Nancy Cleeland et al., Sanchez Officially the Winner, L.A. TIMES, Nov. 23, 1996, at A1. Orange County, located between San Diego and Los Angeles, is today the fifth largest metropolitan area in the United States in the size of its immigrant-stock population. See RUMBAUT , supra note 18, at Table 1. Mr. Dornan unsuccessfully challenged the results of the 1996 election, charging that fraudulent votes by non-citizens had given Ms. Sanchez the close victory. See H.G. Reza & Peter M. Warren, Dornan to Seek Recount Today, L.A. TIMES (Orange County), Nov. 27, 1996, at B3. He ran against her again in 1998. This time he proclaimed that he was the only “true Latino” in the race because of his anti-abortion stance. See Faye Fiore, California and the West, L.A. TIMES, Oct. 28, 1998, at A3. He again lost to Ms. Sanchez by a decisive 56% to 39% margin. See Esther Schrader, Rout by Dornan’s Ex-Followers, L.A. TIMES (Orange County), Nov. 5, 1998, at A14. 56 MERTON, supra note 2, at 187, 200.


One answer to the question of how a new definition of the situation might be introduced is to encourage policy makers to read more social science, a body of literature that rigorously questions the assumption of growing non-assimilation. This is, not surprisingly, a frequent suggestion from social scientists. Unfortunately, politicians tend to use social science the way they use history—by largely searching for data to support views that they already hold. This is not to say that academic scholarship has no influence over debates on immigration policy. We have already mentioned the work of Schuck and Smith which has sparked congressional reconsideration of birthright citizenship rules. In another notable example, a study conducted in 1987 projected a coming labor shortage57 and was relied upon to formulate immigration legislation that significantly increased the annual number of employment-based visas.58 More typically, though, policy makers employ the prodigious study published in 1997 by the National Research Council of the National Academy of Sciences.59 The conclusions of the panel’s report were careful and nuanced; yet, the conclusions were “spun” by the various participants in the immigration debate to support the participant’s preconceived positions. Advocates for immigration characterized the report as demonstrating that immigration, on balance, aided the United States economy.60 Opponents of a high level of immigration cited the report for the view that immigration imposes significant net fiscal costs on particular localities and harms low-skilled workers.61 Ultimately, arguments about assimilation may turn more on symbols than data. Pro-immigration advocates invoke with pride the Statue of Liberty, its torch shining over New York harbor; nativists invoke with dismay Mexican-flag carrying opponents of Proposition 187 marching in the streets of Los

57 HUDSON INSTITUTE, WORKFORCE 2000: WORK AND WORKERS FOR THE TWENTY-FIRST CENTURY 77-79 (1987). 58 See Immigration and Nationality Act of 1990, Pub. L. No. 101-646, § 121, 104 Stat. 4978, 4987. Lawyers also use social science in litigation. Perhaps the best example is Plyler v. Doe, 457 U.S. 202 (1982), in which the Supreme Court invalidated a Texas statute authorizing the exclusion of undocumented children from public schools. Important to the Court’s decision was data showing that large numbers of undocumented children would remain in the United States demonstrating that the impact of Texas’ policy would be to create a caste of illiterate members of United States society. See id. at 228-29. 59 See generally THE NEW AMERICANS: ECONOMIC, DEMOGRAPHIC, AND FISCAL EFFECTS OF IMMIGRATION (J. Smith & B. Edmonston eds., 1997). 60 See Steve Marshall, Immigration Helps Expand the Economy, Report Says, USA TODAY, May 19, 1997, at 3A; Robert Pear, Academy’s Report Says Immigration Benefits the U.S., N.Y. TIMES, May 18, 1997, at A1. 61 See George Borjas & Richard B. Freeman, Findings We Never Found, N.Y. TIMES, Dec. 10, 1997, at A29.


Angeles. The invocation of symbols draws upon a collective memory and set of narratives that inform concepts of belonging and identity. These narratives serve as filters and classifiers, shaping current perceptions to fit pre-existing schema. This, of course, is not a one-way street. Narratives—definitions of the situation—change as they accommodate new stories, much as immigrants shape America even as they are shaped by America. It is here, perhaps that we can escape the self-fulfilling prophecy. Personal and generational narratives of immigrants can demonstrate the complicated nature of assimilation and loyalty in ways that may resonate with the personal histories of Americans.62 That is, we may discover in ourselves and our histories the senses of belonging that we recognize in immigrants. In so doing, we may understand the tensions—even conflicts—that immigrants experience do not undermine assimilation, but may inevitably accompany it. The use of narratives, drawing on American notions of fairness and commitment, has already achieved some important successes. Stories of elderly immigrants rendered ineligible for federal assistance led to the reversal of parts of the 1996 welfare law, 63 and the narrative of freedom fighters and community roots in the United States produced the Nicaraguan Adjustment and Central American Relief Act of 1997 that granted status to long-term (status-less) residents of the United States.64 In other words, maybe we need more

62 In their canonical formulation of assimilation in American life, R.E. Park and E.W. Burgess wrote:

Not by the suppression of old memories, but by their incorporation in his new life is assimilation achieved. . . . Assimilation cannot be promoted directly, but only indirectly, that is, by supplying the conditions that make for participation. There is no process but life itself that can effectually wipe out the immigrant's memory of his past. The inclusion of the immigrant in our common life may perhaps best be reached, therefore, in cooperation that looks not so much to the past as to the future. The second generation of the immigrant may share fully in our memories, but practically all that we can ask of the foreign-born is participation in our ideals, our wishes, and our common enterprises.

ROBERT E. PARK & ERNEST W. BURGESS, INTRODUCTION TO THE SCIENCE OF SOCIOLOGY 739-40 (2d ed. 1924). 63 See Balanced Budget Act of 1997, Pub. L. No. 105-33, § 5301, 111 Stat. 251, 597; 143 CONG. REC. S8493-01, S8496 (discussing the need to reverse the part of the 1996 Welfare law that denied SSI to elderly immigrants). See also Noncitizen Benefit Clarification and Other Technical Amendments Act of 1998, Pub. L. No. 105-306, 112 Stat. 2926; 144 CONG. REC. H8498-01, H8501 (discussing the need to reverse the part of the 1996 welfare law which denied food stamps to elderly immigrants). 64 See District of Columbia Appropriations Act of 1998, Pub. L. No. 105-100, § 202, 111 Stat. 2160, 2193-94; 143 CONG. REC. S10185-06, S10197 (explaining that advocates of the Act were moved by the plight of the freedom fighters).


anthropology and literature and less sociology and law to influence the narratives that undergird our models of membership. On the political level, the answer seems clearer. Anti-immigrant policies are not integrative policies. They do not further the inclusive project of assimilation, even if they provide quasi-coercive incentives to naturalization.65 A nation concerned about and committed to the successful integration of a large and growing resident immigrant population ought to adopt policies that help orient and acculturate immigrants, provide skills and access, and foster tolerance and non-discrimination. The state should model the behavior, such as attachment, commitment and loyalty, that it seeks from its newest members. Models of membership are not free-standing. Like other self-fulfilling prophecies, those models seem to be influenced by the very phenomenon they purport to be classifying. Preservation of the inclusionary model is important, then, not just on principled grounds, but because the model produces the designed outcome and thereby promotes the national interest. If a moral to this sociolegal story of reception and belonging exists, it echoes an ancient admonition: that states, too, reap what they sow.

65 One of the unintended consequences of the passage of Proposition 187 in California and legislation subsequently passed by Congress to bar medical and social services from legal permanent residents (and bills proposed that would deny citizenship to the U.S.-born children of illegal immigrants) was a rush by non-citizen immigrants to apply for naturalization in California, New York, South Florida, and elsewhere, overwhelming the ability of the INS to process them. See Mirta Ojito, A Record Backlog to Get Citizenship Stymies 2 Million: The Wait for Naturalization, 6 Months in 1996, is Now Three Times That Long, N.Y. TIMES, Apr. 20, 1998, at A1.

Terms of Belonging: Are Models of Membership Self-Fulfilling Prophecies?

Documents