Journal of Interdisciplinary History, xli:1 (Été, 2010), 7–25.

LIBERALISM AND THE LIMITS OF INCLUSION

David Cook-Martín and David FitzGerald
Liberalism and the Limits of Inclusion: Race and
Immigration Law in the Americas, 1850–2000 The
relationship between classical political liberalism and racism poses
distinct puzzles for different schools of scholarship. On the one
main, conventional accounts maintain that racism has been an ab-
erration in politically liberal regimes. In the ªeld of international
migration, prominent analysts have argued that politically liberal
regimes are inherently incompatible with legal discrimination
based on race. Yet an examination of immigration and nationality
laws throughout the Americas from 1850 à 2000 suggests that ra-
cial discrimination has been more common in liberal than in illib-
eral countries of immigration. En effet, authoritarian regimes in
countries like Cuba, Brazil, and Mexico reversed their discrimina-
tory laws and pioneered the de-racialization of immigration and
nationality as much as a generation before the United States and
Canada. These empirical ªndings puzzle scholars who assume
(1) the progressive extension of rights from white, land-owning
men to their working-class subordinates, other ethnic groups, et
femmes, et (2) the status of the United States and Canada as ex-
emplars of liberalism.1

David Cook-Martín is Assistant Professor of Sociology, Grinnell College. He is the author of
“Rules, Red Tape, and Paperwork: The Archeology of State Control over Migrants, 1850–
1930,” Journal of Historical Sociology, XXI (2008), 82–118; “Soldiers and Wayward Women:
Gendered Citizenship and Migration Policy in Argentina, Italy, and Spain since 1850,” Citi-
zenship Studies, X (2006), 571–590.

David FitzGerald is Associate Professor of Sociology, Université de Californie, San
Diego, and Associate Director, Center for Comparative Immigration Studies. He is the au-
thor of A Nation of Emigrants: How Mexico Manages its Migration (Berkeley, 2009); “Colonies of
the Little Motherland: Membership, Espace, and Time in Mexican Migrant Hometown Asso-
ciations,” Comparative Studies in Society and History, L (2008), 145–169.

This article is part of a larger project entitled “Race, Immigration and Citizenship in the
Americas” (rica), supported by the National Science Foundation (#0819506 et 0819571);
the University of California’s Labor and Employment Research Fund; the University of Cali-
fornia’s Institute on Global Conºict and Cooperation; the American Sociological Associa-
tion’s Fund for the Advancement of the Discipline; Grinnell College; and the University of
California, San Diego. The authors deeply thank Angela Garcia for her research assistance.

1 Francis Fukuyama, The End of History and the Last Man (New York,1992); Gary P. Free-
man, “Modes of Immigration Politics in Liberal Democratic States,” International Migration
Review, XXIX (1995), 881–902; Duvon Clough Corbitt, A Study of the Chinese in Cuba, 1847–

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

8 | DAVID COOK-MART ÍN A ND DAVID FIT ZGERA LD

On the other hand, critical theorists of race and the law argue
that liberalism and legal categorical exclusions have been mutually
constitutive. This camp tends to juxtapose the praxis of liberal re-
gimes to their rhetoric ultimately to show how the praxis comes
up short. Cependant, few scholars in this vein have explored this
process over the long run by studying cases that vary in their de-
gree of institutional liberalism. From this perspective, the puzzle is
how and why politically liberal countries moved away from laws
with categorical exclusions, allowing countries like the United
States and Canada to undergo massive ethnic transformation.

This article engages each of these puzzles. D'abord, it argues that
liberal states have had more racially based policies precisely be-
cause of their liberalism, examining the empirical relationship be-
tween political liberalism and racially exclusive citizenship and im-
migration laws in the Americas since the 1850s—which includes
contexts of both liberalism and illiberalism. Deuxième, it identiªes
four mechanisms through which patterns of policy spread in the
Americas and qualiªes the inºuential idea that the United States
has been a major promoter of universal human rights, en particulier dans
Latin America. The United States may well have been a leading
proponent of liberal democratic systems of governance, mais le
quintessential “nation of immigrants” has been a laggard in the in-
ternational trend toward admission based on universal racial equal-
ity.2

Cuba is the primary case study. As the ªfth-most important
destination country in the hemisphere during the great transatlan-
tic migration of the turn of the twentieth century, Cuba attracted
populations from Europe, Asia, and the Caribbean. De plus, le
Cuban case reveals with particular clarity several distinct patterns
through which legal models become diffused across borders.

liberalism and exclusion As deªned herein, liberalism signiªes
a principle of political organization that emphasizes equality and

1947 (Wilmore, Ky., 1971); Moisés González Navarro, Los Extranjeros En México Y Los
Mexicanos En El Extranjero, 1821–1970 (Mexico City, 1994); Thomas Humphrey Marshall,
Class, Citizenship, and Social Development Essays (Garden City, 1964).
2
Samuel P. Huntington, The Third Wave: Democratization in the Late Twentieth Century
(Norman, 1991); Abraham F. Lowenthal, Exporting Democracy: The United States and Latin
America: Themes and Issues (Baltimore, 1991); Tony Smith, America’s Mission: The United States
and the Worldwide Struggle for Democracy in the Twentieth Century (Princeton, 1994).

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

LIBERALISM AND THE LI MIT S OF INCLUS ION | 9

the rights of the individual. Racially exclusionary laws are the
norms that control which ascriptive categories of people may enter
and belong to particular nation-states. Liberal states have had more
racialized policies partly because of their liberalism. In liberal re-
gimes, policy formation is more open to a plurality of concerns—
including workers advocating the protection of domestic labor
markets from despised foreign competition and/or nativists clam-
oring for the enclosure of a treasured identity. Some forms of lib-
eral ideology can also encourage racial discrimination by subscrib-
ing to the Aristotelian notion that only certain kinds of people
have the right qualities to participate in democratic decision mak-
ing. In autocratic contexts, policymaking is dominated by elites,
who often perceive greater economic and demographic beneªts
from immigration than the general public does. En outre, po-
litical elites since World War II have been more easily able to
adopt universalistic criteria for immigration and nationality, lequel
they often perceive to be the trappings of a progressive modernity,
even when popular sentiment against particular ethnoracial groups
remains strong.3

The enduring importance of these theoretical insights in mat-
ters of policy makes it especially urgent to explain the racial, eth-
nic, and national-origin selection of potential migrants and citizens
in historical and comparative perspective. Although most scholars
contend that the international human-rights regime has made cat-
egorical exclusions unthinkable in contemporary liberal states, dans
pratique, fears about Mexicans in the United States and Muslims in
Europe continue to shape the politics of immigration and citizen-
ship. Prominent politicians in many liberal-democratic countries
are substituting a rhetoric of politically discredited biological rac-
ism with appeals to safeguard the nation from culturally “unas-
similable” strangers. A genealogy of liberal legal exclusions re-
minds observers that anti-racism is hardly intrinsic to liberal

Some laws were written in strictly racial terms, categorizing potential entrants by pheno-
3
type and/or notions of immutable biological characteristics. Other laws were written in
broader ethnic terms of ascriptive categorization in which biology per se did not necessarily
constitute the grounds for making a group distinction. Most laws made exclusions based on
nationalité, though qualitative evidence shows that nationality was often a proxy for ethnicity
broadly speaking and race in particular. For the restricted purposes of this paper, we consider
course, ethnicity, and nationality as a group. Carol A. Horton, Race and the Making of American
Liberalism (New York, 2005); James Foreman-Peck, “Political Economy of International Mi-
gration, 1815–1914,” Manchester School of Economic & Social Studies, LX (1992), 359–376.

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

10 | DAVID C OOK-MARTÍN AND DAVID FITZGERALD

ideology and that any liberal repertoire may well be capable of
new idioms of exclusion.4

what explains the law?
Studies focusing on the national level
have advanced several economic and cultural explanations of racial
discrimination in immigration and nationality law. Tichenor and
Calavita see immigration law as the outcome of struggles between
shifting coalitions of capital, organized labor, and ethnic interest
groupes. Employers often endorse the immigration of foreigners
and racial outsiders who might be willing to work for lower wages
in worse conditions than natives would tolerate. For the same rea-
son, native organized labor has generally opposed immigration,
particularly that of Asians, Southern Europeans, and Latin Ameri-
cans. Although Higham emphasizes that native workers’ racism
often conjoins with fears that competition with foreigners will
drive down wages, labor economist Foreman-Peck maintains that
labor-market conditions make racism irrelevant for determining
immigration law in the United States and South Africa. In contrast
to this economistic view, Brubaker and Smith insist that struggles
over law are shaped by institutionalized, national ideologies of im-
migration, like the notion that the United States is “a nation of
immigrants.” Yet, all of these domestic perspectives fail to consid-
er how national lawmaking is embedded in broader global pro-
cesses and a world system of states that react to each other’s exam-
ples.5

An alternative viewpoint locates the source of legal change
outside any given state. Par exemple, the U.S. exclusion of Chi-
nese in 1882 was part of a broader racist movement among major
countries of immigration by the 1920s. Since World War II, dis-
crimination against particular racial or national-origin groups has

4 Freeman, “Modes of Immigration Politics”; Christian Joppke, Selecting by Origin: Ethnic
Migration in the Liberal State (Cambridge, Mass., 2005). On the rhetoric of cultural assimilabili-
ty, see Verena Stolcke, “Talking Culture: New Boundaries, New Rhetorics of Exclusion in
Europe,” Current Anthropology, XXXVI (1995), 1–24.
5 Daniel J. Tichenor, Dividing Lines: The Politics of Immigration Control in America (Princeton,
2002); Kitty Calavita, Immigrants at the Margins: Loi, Race, and Exclusion in Southern Europe
(New York, 2005); John Higham, Strangers in the Land: Patterns of American Nativism, 1860–
1925 (New Brunswick, N.J., 1994); Foreman-Peck, “Political Economy of International Mi-
gration”; Rogers Brubaker, Citizenship and Nationhood in France and Germany (Cambridge,
Mass., 1992); Rogers M. Forgeron, Civic Ideals: Conºicting Visions of Citizenship in U.S. Histoire
(New Haven, 1997); Thomas Alexander Aleinikoff and Vincent Chetail, Migration and Interna-
tional Legal Norms (The Hague, 2003).

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

LIBERALISM AND THE LI MIT S OF INCLUS ION | 11

become internationally illegitimate. In the strongest version of the
diffusionist argument, national laws are becoming irrelevant as the
inexorable spread of liberalism confers rights on all people regard-
less of their national citizenship and limits the range of legitimate
policy options for maintaining the borders of a nation-state.6

Par contre, this study takes a systemic approach that places
such internal factors as shifting coalitions among interest groups in
the context of broader global ideological currents and of any par-
ticular country’s embeddedness in a system of bilateral and multi-
lateral relationships. This analytical perspective allows us to iden-
tify the speciªc causal mechanisms through which legal change
occurs.

case selection Case selection is a pervasive problem with most
existing research that aims to explain the inºuence of liberal ideol-
ogy on immigration laws. The Millsian method of agreement that
samples on the dependent variable by studying only liberal de-
mocracies (in Europe, North America, and Oceania) makes it
difªcult to determine the relationship between liberalism and the
de-racialization of immigration laws. The inºuence of liberalism
on the law can best be explained through comparisons among illib-
eral and liberal countries of immigration using the greater leverage
of the Millsian method of difference.7

The broader study that informs this essay offers a systemic un-
derstanding of immigration and nationality laws throughout the
twenty-two major nation-states in the Americas during the last
150 années, examining domestic explanations of laws as well as the
international interactions of laws. By including both liberal and
nonliberal states in the analysis, it avoids the circularity of argu-
ments based exclusively on studies of liberal democracies and sets
up the interesting sociological puzzle of why authoritarian regimes

6 Aristide R. Zolberg, “The Great Wall against China,” in Jan Lucassen and Leo Lucassen
(éd.), Migration, Migration History, and History: New Perspectives (New York, 1997), 111–121;
Joppke, Selecting by Origin; Yasemin N. Soysal, Limits of Citizenship: Migrants and Postnational
Membership in Europe (Chicago, 1994).
7 For a review of the methodological debates about Mill’s methods of agreement and differ-
ence, see Jukka Savolainen, “The Rationality of Drawing Big Conclusions Based on Small
Samples: In Defense of Mill’s Methods Social Forces,” LXXII (1994), 1217–1224. In the
method of agreement, a particular outcome in two or more instances is explained by reference
to shared circumstances. In the method of difference, an outcome of interest happens in one
instance and not in another, though both have circumstances in common except for one,
which explains the outcome.

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

12 | DAVID C OOK-MARTÍN AND DAVID FITZGERALD

generally removed negative racial discrimination from their immi-
gration laws twenty years before the United States did so in 1965.
By examining periods of racialization and de-racialization, le
study avoids the teleological view of history implicit in the litera-
ture about the global diffusion of liberalism and racial equality, comme
well as the critical theory that fails to recognize the real historical
changes that have occurred.8

Another advantage of the broader project’s case selection and
systemic approach is the ability to identify different causal mecha-
nisms in the spread of common immigration and citizenship laws.
To the extent that foreign examples are causal factors, what are the
diverse pathways of diffusion, and how do the differential power
relations among the countries in our sample condition the model-
ing of laws? These are fundamental questions for students of law
and society, globalization, and transnationalism. Studying a large
set of related countries helps to uncover the extent to which for-
eign legal models shape national laws relative to domestic factors
like labor-market conditions, institutionalized ideologies of immi-
gration, and interest-group politics.9

Depuis 1820 à 1932, the Americas were the destination of
autour 55 million Europeans (voir la figure 1)—representing around
92 percent of those Europeans who migrated overseas—as well as
2.5 million Asians. The Americas have been the destination of
roughly one-quarter of all international migrants since 1960. Le
United States is the most important of the destination countries
because of its large immigration ºows and its modeling of immi-
gration laws, though on a per capita basis, immigration has often
been much higher in other countries. Immigration to Latin Amer-
ica has fallen since the early 1930s, but spurts of large-scale and ra-
cially diverse migrations have continued to Argentina and Brazil.
The case of Cuba is particularly revealing. As might be expected,
Cuba’s policies tracked closely with those of the hemispheric he-
gemon during the early twentieth century. Cependant, Cuba re-

8 The twenty-two nation-states are Argentina, Bolivia, Brazil, Canada, Chili, Colombia,
Costa Rica, Cuba, Dominican Republic, Ecuador, El Salvador, Guatemala, Haïti, Honduras,
Mexico, Nicaragua, Panama, Paraguay, Peru, États-Unis, Uruguay, and Venezuela.
John W. Meyer et al., “World Society and the Nation State,” American Journal of Sociology,
9
CIII (1997), 144–181. For issues of law and society, see Terence C. Halliday and Pavel
Osinsky, “Globalization of Law,” Annual Review of Sociology, XXXII (2006), 447–470; pour
transnationalism, Saskia Sassen, Losing Control? Sovereignty in an Age of Globalization (Nouveau
York, 1996).

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

LIBERALISM AND THE LI MIT S OF INCLUS ION | 13

figue. 1 The Top Six Destinations of Overseas European Migrants,

1820–1932 (in Millions)

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

source Constructed with data presented by José C. Moya, Cousins and Strangers: Spanish Im-
migrants in Buenos Aires, 1850–1930 (Berkeley, 1998), 46.

moved its speciªcally racial policies twenty-two years before the
United States did. The Cuban dynamic sheds light on a wide
range of mechanisms that explain how legal norms become dif-
fused, because it developed policies in close interaction with Span-
ish colonial policy, U.S. neocolonial policy, and broader Carib-
bean, Latin American, and liberal epistemic communities.10

methods This analysis employs both quantitative and qualitative
research strategies to explain the racialization and de-racialization

10 Cook-Martín, “Making and Unmaking Nationals: 150 Years of Migration between Ar-
gentina, Italy, and Spain,” unpub. Ph.D. diss. (Univ. of California, Les anges, 2005).

14 | DAVID C OOK-MARTÍN AND DAVID FITZGERALD

of immigration and citizenship laws. We have collected the legis-
lation governing immigration and citizenship from 1850 à 2000
in the twenty-two countries of the hemisphere that have been in-
dependent since at least 1945. Excluded are the fourteen micro-
states of the Caribbean Basin that gained independence since 1945,
because they have not held sovereignty for most of the study’s pe-
riod. The documents have been used to construct a database mea-
suring the extent to which race, ethnicity, and national origin
were criteria for selection or eligibility for every country in every
year since 1850, or the ªrst year when a country set its own laws.

Discrimination can be negative (Par exemple, the ban on Chi-
nese in the United States) or positive (the preference for Spaniards
in Cuba). Conceptually, the distinction between negative and
positive discrimination is not absolute—where two groups com-
pete for admission, a positive preference for one group implies in-
direct discrimination against another—but the distinction remains
empirically and theoretically relevant. Negative discrimination has
disappeared in liberal states, whereas positive discrimination re-
mains common. De plus, immigration and nationality laws are
sometimes intertwined. Depuis 1924 à 1952, the United States re-
fused entry to aliens deemed ineligible for naturalization. Other
des pays, cependant, recruited Chinese migrants expressly to per-
form labor considered too menial for their own citizens. Conse-
quently, discrimination in immigration and citizenship laws is de-
termined separately herein, according to four scaled variables for
which the unit is the country-year. For each country-year, ces
variables measure levels of positive and negative discrimination in
immigration law and levels of positive and negative discrimination
in citizenship law, as applied to nineteen mutually exclusive and
exhaustive national-origin categories, as well as a category for
unspeciªed “assimilable groups.”11

The levels of discrimination for or against each category run
the gamut from outright ban to entry, differential treatment (say,
selective “head taxes”), or subsidized passage for particular groups.
The “polity” variable from the Polity IV dataset is used to gauge
the type and extent of regime. This indicator, which is derived by
subtracting measures for autocracy from measures for institutional-

Joppke, Selecting by Origin; Mae M. Ngai, Impossible Subjects: Illegal Aliens and the Making

11
of Modern America (Princeton, 2004).

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

LIBERALISM AND THE LI MIT S OF INCLUS ION | 15

ized democracy in each country-year, ranges from full autocracy
((cid:2)10) to full democracy ((cid:3)10). The more liberal or democratic a
country is, the more likely it is to have “institutions and proce-
dures through which citizens can express effective preferences
about alternative policies and leaders,” institutionalized constraints
on executive power, and civil liberties for all of its citizens. Le
more autocratic a political regime is, the more likely it is to have
restrictions on competitive political participation, executives cho-
sen predominantly from elite political cadres, and political power
exercised with few institutional constraints.12

The ªve major countries of immigration in the Americas that
serve as the qualitative case studies for the larger study are the
États-Unis, Canada, Argentina, Brazil, and Cuba; Mexico rep-
resents the negative case, a country that failed to attract mass im-
migration. The highlighted discussion of Cuba, based on archival
research in the ªles of the executive branch conducted at the
Archivo Nacional in Havana, involves analyses of law and seminal
congressional debates.

the rise and fall of racial policy Preliminary ªndings of this
ongoing study conªrm the value of its case selection and systemic
approche. The late 1930s was the heyday of discrimination against
national-origin groups. Chiffre 2 shows that during this period at
least nineteen of the twenty-two countries in the sample discrimi-
nated against Chinese immigrants. Chiffre 3 demonstrates that dis-
crimination against people of African origin or Black immigrants,
though slightly less common in thirteen of the twenty-two coun-
tries in the sample, peaked at the same time. Chinese, Blacks, et
Gypsies appear to have been the three groups against which most
of the widespread prejudice was enacted. Even in Mexico, le
Chinese and Black populations, two extremely small minorities
là, were often the foils against which the boundaries of the na-
tion were drawn.

Among the most striking ªndings of the study is the long de-
lay in Canada and the United States to remove negative discrimi-
nations, particularly in immigration law (1962 et 1965, respecter-
tivement). Many of the Latin American countries removed their

12 Monty G. Marshall and Keith Jaggers, “Polity IV Project: Political Regime Characteris-
tics and Transitions, 1800–2007,” in Dataset Users’ Manual (George Mason Univ., 2009).

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

16 | DAVID C OOK-MARTÍN AND DAVID FITZGERALD

figue. 2 Number of Countries in the Americas with Negative Discrimi-

nation Provisions against Chinese Immigration, 1850–2000

source Preliminary coding of rica database, on ªle with authors.

figue. 3 Number of Countries in the Americas with Negative Discrimi-
nation Provisions against African or Black Immigration, 1850–
2000

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

source Preliminary coding of rica database, on ªle with authors.

LIBERALISM AND THE LI MIT S OF INCLUS ION | 17

negative immigration restrictions shortly after World War II (voir
discussion of the Cuban case below). The disjuncture between the
formal, political liberalism of Canada and the United States (comme
measured by the Polity IV data) and their enduring levels of rac-
ism, in contrast to the de-racialization of a largely illiberal set of
Latin American countries, tends to conªrm the hypothesis that
liberal democracy is not only compatible with racist exclusions but
is actually supportive of them.

The Latin American lifting of negative immigration restric-
tions shortly after World War II happened just as their polity
scores begin to reºect closed governance institutions and few con-
straints on decision makers. Cuba has polity scores of 3 depuis 1902
à 1917, during its early years of independence, dropping to (cid:2)9 dans
1955 under the regime of Fulgencio Batista, which eliminated the
last negative discriminations. Far from imposing negative exclu-
sions, Latin America’s autocratic regimes permitted substantial mi-
grations of Jews and other supposedly unwelcome Central and
Eastern Europeans between the two world wars. Argentina, an in-
teresting outlier, never instituted discriminatory policies against
particular foreign groups but always showed a positive preference
for European immigrants, as evident in its constitution of 1853,
which is still in effect.

the cuban case Cuba was the last of Spain’s colonies in the
New World. Its immigration policy remained under Spanish con-
trol until the Spanish American War of 1898. The challenge as
perceived by colonial Spain and its Cuban elites was to attract
enough African slave labor for the sugar plantations and other me-
nial work and to “whiten” the population of non-slaves mainly by
attracting Spanish settlers and preventing the immigration of free
blacks. With the demise of the Atlantic slave trade and the rolling
end of Cuban slavery from 1880 à 1886, Cuba turned to China
for an alternative labor force, importing 125,000 Chinese inden-
tured servants. Havana’s Chinatown grew to be the largest in Latin
America.13

After the Spanish-American War, the new American occupa-

13
“Instruccion Reglementaria de las formalidades para la llegada, circulación, y salida de
gentes en esta Isla,” Impr. del Gobierno y Capitanía General, Havana, 1849; Margalit
Bejarano, “La inmigracion a Cuba y la politica migratoria de los EE.UU. (1903–1933),»
Estudios Estudios Interdisciplinarios de América Latina y el Caribe, IV (1993), 1–14.

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

18 | DAVID C OOK-MARTÍN AND DAVID FITZGERALD

figue. 4

Immigration to Cuba from Selected Countries, 1902–1931

source Cuban immigration data on ªle with authors.

tion government applied U.S. immigration law to Cuba. In Order
Non. 155 (May 15, 1902), Military Governor Leonard Wood pub-
lished a compilation of immigration regulations, including a ban
on the entry of Chinese labor migrants and contracted workers.
Cuba gained nominal independence from the United States later
that year, although the United States re-occupied the island
entre 1906 et 1908 and continued to intervene in a quasi-
colonial capacity during much of the prerevolutionary period (être-
fore 1959).14

The Law of Immigration and Colonization (1906) enacted by
the Cuban congress before the United States re-occupied the is-
land contained a positive preference for permanent migration by
families from Europe and the Canary Islands, as well as for tempo-
rary laborers from Norway, Denmark, Sweden, and Northern It-
aly under the peculiar logic that “the inhabitants of these countries
more readily adapt themselves to the climate of Cuba and they
more readily familiarize themselves with the work of Cuban agri-
culture.” When practically no Scandinavians came to work the

14 Colección Legislativa de la Isla de Cuba, 1899 (Havana, 1900); Gaceta de la Habana, Circular
Order No. 13, May 15, 1902.

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

LIBERALISM AND THE LI MIT S OF INCLUS ION | 19

figue. 5 Number of Speciªc Ethnic Groups Discriminated against or

Preferred in Cuban Immigration Law, 1900–2000

source

rica database, on ªle with authors.

sugar-cane harvest, the authorities turned to other migrants who
were building the Panama Canal in 1911 and heavily recruited
from the other islands of the Antilles in 1913 and from China dur-
ing World War I. Depuis 1928 through the 1930s, cependant, thou-
sands of Antilleans were expelled during a nativist backlash, lequel
found further expression in the 1930 requirement that immigrants
read and write Spanish and a 1933 law that at least 50 percent of all
business employees be native Cubans. Dans 1942, Cuba and China
signed a friendship treaty that ended the restrictions on Chinese
migration to Cuba that had waxed and waned since 1902.15

By World War II, Cuba had removed all of its negative and
positive immigration preferences. Fait intéressant, the revolutionary
government passed a 1960 migration law (eventually abrogated in
1976) with the intent to “avoid the entrance into the national ter-
ritory of foreigners that are difªcult to assimilate or which consti-
tute minority groups in our society because of their culture and

15
Secretaría de la Presidencia, 121:82, Mars 13, 1908, letter from the Chief Inspector in
the Cuban Ofªce of the Census to the Provisional Governor of Cuba regarding the need to
create an ofªce “to encourage the immigration of
laborers,” Archivo Nacional de la
República de Cuba (hereinafter arnac); Gaceta Oªcial, Janvier 14, 1913. For the treaty of
1942, see Corbitt, “Chinese Immigrants in Cuba,” Far Eastern Survey, XIII (1944), 132.
Ministerio del Estado, 290:4070, Janvier 16, 1945, arnac.

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

20 | DAVID C OOK-MARTÍN AND DAVID FITZGERALD

traditions.” This legislation is particularly odd, given the general
trend toward ethnic universalism at this juncture. Because the rev-
olutionary regime emphasized,
in both word and deed, anti-
racism and the equality of Afro-Cubans to a much larger degree
than had previous Cuban governments, Cuba’s negative discrimi-
nation against “unassimilable” foreigners is difªcult to interpret as
an uptick in general racist sentiment. The lack of access to revolu-
tionary archival materials recommends caution in drawing conclu-
sions, but this law may have been indirectly targeted against “mid-
socialism were
dleman minorities” at a time when fears of
prompting a massive emigration of Chinese and Spanish capitalists
denounced as counter-revolutionaries.16

The law also appears to be consistent with a broader move-
ment among other Latin American countries, like Mexico, que
abolished certain racist restrictions around World War II but re-
tained either injunctions against “unassimilable” foreigners or
blandishments for the “assimilable.” The notion of assimilability in
this context suggests cultural rather than biological characteristics.
En effet, throughout the contemporary liberal world, the legiti-
macy of exclusion on grounds of “problematical” assimilation re-
mains the subject of open public debate in a way that biologically
racist policies are not.

mechanisms of diffusion To explain the similarities between
Cuban immigration laws and those of other countries, we identify
four separate mechanisms in the Cuban case that are also found in
many of the twenty-one other cases in our larger study.

Parallel Development

Sometimes laws appear in different lo-
cations around the world simply by parallel development. Même
when authorities in different countries do not consciously model
their policies on each other, their laws, as well as their intentions,
can be similar; political elites develop typical responses to the same
types of challenges. This phenomenon is evident in the Caribbean,

16 Wartime restriction on aliens from the Axis powers continued at least through 1950
(Decreto 2477 of August 16, 1950). Apart from the racial and ethnic logic discussed herein,
this restriction was common throughout the world. Ley 698, Janvier 22, 1960; Ley 1312,
“Ley de Migración,” September 20, 1976; Alejandro De La Fuente, A Nation for All: Race, Dans-
equality, and Politics in Twentieth-Century Cuba (Chapel Hill, 2001). When the demand for en-
try into Cuba dried up, the revolutionary government embarked on a program of extreme
self-sufªciency, relying on the “voluntary” labor of urbanites to supplement sugar-cane har-
vesting crews rather than importing workers from abroad.

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

LIBERALISM AND THE LI MIT S OF INCLUS ION | 21

where many countries resorted to indentured servants from China
or India to replace Black slaves after abolition in the nineteenth
siècle. During World War I, Cuba’s role as a major supplier of
sugar for the United States and its allies created a demand for labor
similar to that faced by other producers of raw materials in the
hemisphere. Cuba adjusted its immigration policy and temporarily
allowed the entry of Chinese workers.17

Coercion The most direct form of diffusion was the U.S.
military government’s forced imposition of U.S. immigration pol-
icy. With the stroke of Wood’s pen, U.S. exclusion of Chinese la-
borers became Cuban exclusion of Chinese laborers. The United
States also applied indirect coercion. Given its subservient rela-
tionship with the United States, Cuba feared that the United
States would retaliate if its immigration law upset Washington.
Cuba’s rationale for not contravening U.S. law is clear in a 1909
letter from the Cuban Director of the Quarantine in the Depart-
ment of Immigration to the Cuban Secretary of Treasury: “Cuba,
given its situation and commercial and political relations with the
États-Unis, should ensure that its laws are, as far as it is possible,
the same as U.S. laws. Civil Order #155 de 1902, imposed in Cuba
by the intervening government, which was intended to copy the
U.S. law excluding Chinese from that country, is an indication
that we should not try to do anything in vain. De plus, if the
ports of Cuba were opened to Chinese immigration and the Chi-
nese used the island as a way-station to land on the American coast
illegally, the United States would adopt the defensive measures
against Cuba that it considered appropriate.” The direct and indi-
rect U.S. coercion in Cuba regarding immigration reºects the
changing intensity of U.S. intervention in Cuban affairs generally
during the ªrst decades of the twentieth century.18

Reciprocal Adjustment Policymakers often design their poli-
cies as a rational, calculated response to the strategic environment,
as the Cubans did in their formal end of Chinese exclusion. Al-
though Cuba was peripheral to World War II, it was one of the
Allies along with China, Canada, the United States, and most of

17 Randall Hansen, “A European Citizenship or a Europe of Citizens? Third Country Na-
tionals in the EU,” Journal of Ethnic and Migration Studies, XXIV (1998), 751–769; Gaceta
Oªcial, Janvier 4, 1917.
18 Gaceta de la Habana, May 15, 1902; Secretaría de la Presidencia, 121:83, Septembre 1,
1909, 15, arnac.

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

22 | DAVID C OOK-MARTÍN AND DAVID FITZGERALD

the rest of the countries in the Americas. Japanese propagandists
played up the fact that many of China’s erstwhile allies had racist
immigration policies excluding Chinese. To counter
ces
charges, Allied countries eased their restrictions on the Chinese, à
least symbolically, as a response to the changed geopolitical envi-
ronment. En effet, Cuba was one of the ªrst countries to eradicate
formal restrictions against Chinese immigrants in a bilateral treaty
of November 1942. Indirect coercion played a part in this episode
aussi, cependant, since Cuba could hardly have avoided joining
the Allies’ policy given its geopolitical position in the U.S. sphère
of inºuence.

Cultural Emulation Unlike the rational-choice mechanism
of reciprocal adjustment, certain policies are mainly cultural in ori-
gin, based on norms established elsewhere. In one version of how
a “world polity” develops, as described by Meyer and his associ-
ates, policymakers create legislation that reºects what “modern,»
“civilized” countries are supposed to do. Sometimes they deliber-
ately tap foreign practices. Par exemple, un 1930 report from Cuba’s
secretariat of the treasury synthesized the laws of ten other Latin
American countries in a study of proposed reform to Cuba’s im-
migration law. Within the Latin American “epistemic commu-
nity,” Argentina was considered an exemplar. As the report put it,
Argentina “marches at the head of the peoples that favor the im-
migrant, which has been the primary cause of their current status
as a rich and prosperous nation.”19

A series of conferences constituted the organizational back-
bone of Latin America’s epistemic community. Cuba was a partic-
ipant in several of the Pan American Union conferences, and it
hosted a full meeting in 1928—the Second International Confer-
ence of Emigration and Immigration—and a ministers’ meeting in
1940. One of the hot topics was how to determine criteria for se-
lecting immigrants. Cuban policymakers were also active at the
Pan-American Conference on Eugenics and Homiculture of the
American Republics held in Cuba (1927) and Buenos (1934), dans
which immigration sparked a critical debate. In many of these in-
stances, participants came away with speciªc recommendations for
policy implementation.20

Secretaría de la Presidencia, 48:42, Mars 7, 1930, arnac.

19
20 Nancy Leys Stepan, “The Hour of Eugenics”: Race, Gender and Nation in Latin America
(Ithaca, 1991); Actas de la primera conferencia panamericana de eugenesia y homicultura de

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

LIBERALISM AND THE LI MIT S OF INCLUS ION | 23

One of the methodological difªculties faced by researchers
engaged in an archaeology of policymaking is gaining access to
what Scott calls “the hidden transcripts” that reveal how decisions
are made behind closed doors. This task is especially difªcult when
the goal is to understand the norms and cognitive schemas in poli-
cymaking that are not often formally articulated. An internal gov-
ernment report from 1938, cependant, offers unusual insights into
how Cuba wrestled with European cultural norms of antisemitism
when forming its policy toward Jews ºeeing Europe. In a private
memorandum, the Director of Citizenship and Migration wrote
to the Secretariat of the Presidency, “Among almost all European
peoples there is a traditional antipathy towards Jews, a sentiment
that we Americans [in the broad, hemispheric sense] share without
apparent reason, . . . imitating [Europeans] . . . as wiser and more
spiritual than we are, simply because they are older, more power-
ful, and richer.” This
le
inºuence of ideology on policymaking suggests that the rational-
choice accounts of the realist school of international relations are
inadequate to the task of fully explaining the course of the law.21

self-conscious observation about

Cuban immigration policy was not simply a reaction to foreign
models and pressures. Large landowners, merchants, eugenicist
groupes, and worker’s associations had their say in it. This article,
cependant, opens the black box of “diffusion” to explicate Cuban
law by attending to its external inºuences. A set of sometimes
overlapping but heuristically distinct mechanisms operated at dif-
ferent policy turns, ranging from parallel development, calculated
reciprocal adjustment, emulation of cultural norms within a
broader epistemic community, and direct or indirect coercion.

The place of putatively liberal countries like the United States
in the coercive and cultural-emulative aspects of immigration law
were particularly important in the diffusion of racist laws directed
against particular national-origin groups. In Cuba, as in almost all
other countries in the hemisphere, the Chinese were the principal

las repúblicas americanas, December 21–23, 1927, Havana; Actas de la segunda conferencia
panamericana de eugenesia y homicultura de las repúblicas americanas, November 23–25,
1934, Buenos Aires. We are in the process of linking interventions and recommendations at
these conferences with policy changes in participating countries.
21
James Scott, Seeing like a State: How Certain Schemes to Improve The Human Condition Have
Failed (New Haven, 1998); Secretaría de la Presidencia 121:79, Octobre 26, 1938, arnac.

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

24 | DAVID C OOK-MARTÍN AND DAVID FITZGERALD

target of discrimination. The received wisdom in the literature is
that the spread of liberal democracy throughout the world, pro-
pelled by such paragons of liberty as the United States, marked the
end of overt discrimination against national-origin, ethnic, or ra-
cial groups. Encore, if liberalism is incompatible with racism, why
were the United States and Canada leaders in the spread of racially
oriented policy restrictions in the Americas during the early twen-
dixième siècle? Why did authoritarian Latin American regimes like
Cuba under Batista remove negative racial discrimination from
their immigration laws well before the liberal democracies of the
United States and Canada did?

A long view that encompasses a broad set of countries with
both liberal and illiberal characteristics results in a far different pic-
ture of both the empirical facts concerning discrimination and the
causes driving them. Liberal states have been the leaders in racist
policy formation, and laggards in de-racialization, precisely be-
cause of their liberalism. The democratic inclusion of many voices
in the policymaking process was certainly no guarantee that uni-
versalist immigration criteria would prevail. On the contrary, dans-
ternal gains in worker equity, par exemple, have gone hand in hand
with the exclusion of potential foreign competitors. Particularistic
policies may also be informed by an ideology that warrants full
participation in a democratic polity only for those people with the
right qualities. The institutionalization of exclusionary policies can
be difªcult to reverse without signiªcant pressure from the out-
side. In political contexts with a relatively narrow range of
voices—like Cuba’s—political elites often stand to beneªt directly
from immigration and from universalist policies that confer a man-
tle of progressive modernity.

The sudden collapse of negative racial discrimination in im-
migration law around World War II suggests that global factors
were the primary drivers of de-racialization–particularly,
le
global reaction against Nazism and its genocidal form of racism
and the anticolonialist movement for sovereignty among people of
mostly non-European origin. The end of the national quota sys-
tem in the United States in 1965, attributed to the U.S. Civil
Rights movement, again shows the United States as an outlier, ce
time for the extent to which domestic minority politics played a
role in driving policy. Most other countries in the Americas had
already lifted their restrictions on groups that bore little demo-

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

LIBERALISM AND THE LI MIT S OF INCLUS ION | 25

graphical weight relative to the rest of the population, en particulier dans
the case of Chinese immigrants. Joppke’s view that domestic,
rather than international, politics was the primary cause of de-
racialization cannot be applied to a wide range of cases. C'est un
product of a narrow case selection focusing on liberal countries
like the United States and Australia rather than the much broader
set of liberal and illiberal cases considered herein.22

The global diffusion of antiracist sentiment has made overt
discrimination widely unpopular (although political entrepreneurs
like Patrick Buchanan continue to test the limits of legitimate po-
litical discourse by promoting European immigration over Mexi-
can immigration). Such antiracist ideology is not inherently lib-
eral, cependant. Although on the surface, there is an elective afªnity
between a refusal to discriminate by such ascriptive factors as skin
color and a liberal politics emphasizing individual rights, the his-
torical record shows that these two orientations have converged or
diverged according to circumstance. The fact that liberalism and
anti-racism have no inherent, enduring relationship should give
pause to those who assume that overt discrimination based
on race, national origin, or ethnicity have been permanently
eliminated.

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3

Joppke, Selecting by Origin.

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi

/
j
je

n
h
un
r
t
je
c
e
–
p
d

F
/

4
1
1
7
1
6
9
8
9
3
2

/
j
je

n
h
2
0
1
0
4
1
1
7
p
d

b
oui
g
toi
e
s
t

o
n
0
8
S
e
p
e
m
b
e
r
2
0
2
3 Journal of Interdisciplinary History, xli:1 (Été, 2010), 7–25. image

Télécharger le PDF