970 AJS Volume 113 Number 4 (January 2008): 9701022

2008 by The University of Chicago. All rights reserved.

0002-9602/2008/11304-0002$10.00
The Making and Unmaking of Ethnic
Boundaries: A Multilevel Process Theory
1
Andreas Wimmer
University of California, Los Angeles
Primordialist and constructivist authors have debated the nature of
ethnicity as such and therefore failed to explain why its charac-
teristics vary so dramatically across cases, displaying different de-
grees of social closure, political salience, cultural distinctiveness, and
historical stability. The author introduces a multilevel process theory
to understand how these characteristics are generated and trans-
formed over time. The theory assumes that ethnic boundaries are
the outcome of the classicatory struggles and negotiations between
actors situated in a social eld. Three characteristics of a eldthe
institutional order, distribution of power, and political networks
determine which actors will adopt which strategy of ethnic boundary
making. The author then discusses the conditions under which these
negotiations will lead to a shared understanding of the location and
meaning of boundaries. The nature of this consensus explains the
particular characteristics of an ethnic boundary. A nal section iden-
ties endogenous and exogenous mechanisms of change.
TOWARD A COMPARATIVE SOCIOLOGY OF ETHNIC BOUNDARIES
Beyond Constructivism
The comparative study of ethnicity rests rmly on the ground established
by Fredrik Barth (1969b) in his well-known introduction to a collection
1
Various versions of this article were presented at UCLAs Department of Sociology,
the Institute for Migration Research and Intercultural Studies of the University of
Osnabru ck, Harvards Center for European Studies, the Center for Comparative Re-
search of Yale University, the Association for the Study of Ethnicity at the London
School of Economics, the Center for Ethnicity and Citizenship of the University of
Bristol, the Department of Political Science and International Relations of University
College Dublin, and the Department of Sociology of the University of Go ttingen. For
helpful comments and challenging critiques, I should like to thank Klaus Bade, Fredrik
Barth, Michael Bommes, John Breuilly, Rogers Brubaker, Marian Cadogan, Hartmut
Esser, Jon Fox, Matteo Fumigalli, Nazgol Ghandnoosh, Philip Gorski, Wesley Hiers,
John Hutchinson, Eric Kaufmann, Matthias Ko nig, Sinisa Malesevic, Tariq Modood,
Ethnic Boundaries
971
of ethnographic case studies. Barth broke away from the Herderian canon
in anthropology, according to which each ethnic group represented a his-
torically grown, uniquely shaped ower in the garden of human cultures.
2
Instead of studying each of these cultures in a separate ethnography, Barth
and his collaborators observed how the boundaries between two ethnic
groups are maintained, even though their cultures might be indistinguish-
able and even though individuals and groups might switch from one side
of the boundary to the other. Barths approach to ethnicity thus no longer
resembled an exercise in Linnean taxonomy but in social ecology.
Barth pioneered what later became known as constructivism: the
claim that ethnicity is the product of a social process rather than a cultural
given, made and remade rather than taken for granted, chosen depending
on circumstances rather than ascribed through birth. In the following two
decades, prolonged battles emerged between devotees of this constructivist
perspective and adherents to older views that were more in line with
Herderian notions of the binding power of ethnicity and culture. This
debate has often been framed in dichotomous terms: primordialism,
which underlined that ethnic membership was acquired through birth
and thus represented a given characteristic of the social world, was
pitted against instrumentalism, which maintained that individuals
choose between various identities according to self-interest. Essentialism
was opposed to situationalism, the former privileging the transcontex-
tual stability provided by ethnic cultures while the latter showed how
individuals identify with different ethnic categories depending on the logic
of the situation. Modernists attributed the salience of ethnicity to the
rise of the modern nation-state, while perennialists insisted that ethnicity
represented one of the most stable principles of social organization in
human history. Scholars who insisted on the subjectively felt reality and
deeply rooted character of ethnic identity argued against those for whom
ethnic distinctions were primarily driven by the changing interests of
individual or collective actors.
3
Orlando Patterson, Abigail Saguy, Peter Stamatov, Paul Statham, Art Stinchcombe,
Ivan Szelenyi, Yasuko Takezawa, Eddie Telles, Jennifer Todd, Sarah Zingg, and Lynne
Zucker. Special thanks go to Miche` le Lamont, whose invitation to a conference pro-
vided the initial stimulus for writing this article and who continued to support the
project through its various phases. All errors of fact and thought unfortunately remain
my sole responsibility. Direct correspondence to Andreas Wimmer, Department of
Sociology, 264 Haines Hall, University of California, Los Angeles, California, 90095.
E-mail: awimmer@soc.ucla.edu
2
See Herder ([1784] 1968). On Herders inuence on the contemporary study of eth-
nicity see Wimmer (in press).
3
These binary oppositions appeared in various constellations and combinations. In
the eyes of some, they aligned along a grand battle line separating constructivist-
instrumentalist-circumstantialist-interest approaches from the essentialist-primordial-
American Journal of Sociology
972
This article attempts to transcend these debates.
4
I argue that the em-
pirical and analytical questions that they raise cannot be solved by def-
initional ontologyby trying to nd out what ethnicity really is. The
past decades have produced an impressive variety of case studies in which
we nd examples that tand contradictany of the positions sum-
marized above, as will be shown in the following section. The denitional
debates may have diverted our efforts away from understanding why
ethnicity appears in such variable forms. While there is a substantial body
of work illustrating the contrasting properties of ethnic, national, or racial
boundaries across usually two or three examples, little has been done to
explain the entire range of empirically documented variation through
comparative theory building and research. This article is certainly not a
successful execution of this task; rather, it is intended as a substantial
beginning and as an invitation to other scholars to further advance this
agenda.
The article makes a twofold contribution to this project. First, it offers
a systematic description of the wide variety of ethnic constellations that
empirical research has brought to light and shows that none of the existing
comparative hypotheses sufces to make sense of these differences. Four
principal dimensions of variation are identied: different degrees of po-
litical salience of ethnic boundaries, of social closure and exclusion along
ethnic lines, of cultural differentiation between groups, and of stability
over time.
Second, I outline an analytically more sophisticated and empirically
more promising theory designed to explain why the process of ethnic group
formation produces such different outcomes. The model leads from the
macrostructural level to the agency of individuals and aggregates their
actions back to the macrostructural level. It thus represents a dynamic
ist-perennialist-identity position. However, some debates crisscrossed this divide. For
example, constructivists who emphasized individual choice and economic interests
argued with other constructivists who conceived identity formation as a collective
process.
4
During the 1980s, various attempts were made at reconciling these positions and
arriving at a theoretical synthesis (McKay 1982; Bentley 1987; Keyes 1981; G. Scott
1990; Nagata 1981). The mainstream debate, however, continued to oscillate between
the various pairs of oppositions. By the end of the 1990s, constructivism had gained
the upper hand over essentialism, instrumentalism over primordialism, and circum-
stantialism over perennialism. Contrary positions are still expressed today and with
much more sophistication than in decades before (see Roosens 1994; Hirschfeld 1996;
Gil-White 1999, 2001) but seem to be no longer in control of mainstream discourse.
Routine references to the constructed, changing, and power-driven character of
ethnicity that one nds in todays literature illustrate the contemporary hegemony of
constructivism. Primordialism, essentialism, and perennialism have, however, survived
in unacknowledged form in some ethnic studies departments and in migration studies
(Wimmer, 2007) as well as in conict research (Brubaker 2004).
Ethnic Boundaries
973
process theory focused on how social forms are generated and transformed
over time. In a nutshell, the model explains the varying features of ethnic
boundaries as the result of the negotiations between actors whose strat-
egies are shaped by the characteristics of the social eld. It proceeds
through four steps, each corresponding to a separate section.
In a preliminary step, I provide an inventory of possible strategies of
ethnic boundary making that individual and collective actors might pur-
sue. In a second step, I discuss three characteristics of social elds that
explain which actors will pursue which strategies (the macrostructural
level): (1) the institutional framework determines which types of bound-
ariesethnic, social class, gender, villages, or otherscan be drawn in a
meaningful and acceptable way in a particular social eld; (2) the position
in a hierarchy of power denes the interests according to which actors
choose between different possible levels of ethnic differentiation; (3) who
exactly will be included in the actors own ethnic category depends on
the structure of her political alliances. In the third step, I explain how
the ensuing classicatory and political struggles between actors advocat-
ing different ethnic categories may lead to a more or less encompassing
consensus over the topography, character, and rightful consequences of
boundaries (the agency level). Finally, it is shown that the nature of this
consensus explains the characteristics of ethnic boundaries: their varying
degrees of political salience, social closure, cultural differentiation, and
historical stability (leading back to the structural level).
This multilevel process model of ethnic boundary making represents,
to the best of my knowledge, the rst attempt at systematically explaining
the varying character and consequences of ethnic boundaries. It thus goes
beyond the dominant approaches in comparative ethnicity that either try
to get at the nature of the ethnic phenomenon as such, develop static
typologies of different ethnic congurations, or outline in broad strokes
the world historical forces that have given ethnic, racial, or national di-
visions their current signicance.
Dening the Field
Following the tradition established by Max Weber ([1922] 1985, p. 237),
I dene ethnicity as a subjectively felt sense of belonging based on the
belief in shared culture and common ancestry. This believe refers to cul-
tural practices perceived as typical for the community, to myths of a
common historical origin, or to phenotypical similarities (see Weber 1978,
pp. 38598; Schermerhorn 1970; Erikson 1993; Jenkins 1997; Cornell and
Hartman 1998). In this broad understanding of ethnicity, race is treated
American Journal of Sociology
974
as a subtype of ethnicity,
5
as is nationhood: if phenotypical features are
used as indicators of group membership, we speak of ethnosomatic groups;
if members of an ethnic community have developed nationalist aspirations
and demand (or control) a state of their own, we describe such categories
and groups as nations (Jenkins 1997, chap. 6; Weber 1978, pp. 92126;
Smith 1986). Further subtypes of ethnicity can be distinguished depending
on the type of markers that are used to substantiate the belief in shared
culture and ancestry, most importantly ethnoreligious, ethnoregional, and
ethnolinguistic categories and groups.
Subsuming race under ethnicity runs against the folk use of these
terms in the United States. Race is associated with African-Americans,
while ethnicity commonly refers to the less consequential distinctions
among the dominant white group based on different European countries
of origin. From W. Lloyd Warners Yankee City studies onward (Sollors
1986, pp. 2123), mainstream American sociology treated race and eth-
nicity as phenomena of a different order (see van den Berghe 1991; Feagin
and Feagin 1993; Omi and Winant 1994; Bonilla-Silva 1999; Cornell and
Hartman 1998), reecting the dramatically different fate that the descen-
dents of African slaves and European immigrants experienced over the
past two centuries. While using a terminology that contradicts domestic
common sense is inconvenient, adopting this common sense for compar-
ative purposes would be even more problematic (see Loveman 1997; Kiv-
isto 2003).
First, treating race as fundamentally different from ethnicity overlooks
the fact that one and the same group of individuals might be treated as
a race at one point in history and as another type of ethnic category at
another: in the 16th and 17th centuries, African slaves in the United States
were primarily dened as pagans, and their English masters as Christians.
Only after about 1680 was this ethnoreligious distinction gradually re-
placed by the ethnosomatic differentiation between white and Negro
(Jordan 1968). Second, phenotypical differences are often evoked as one
among other markers of ethnic distinction, as the racialization of ethnicity
in Rwanda and Burundi and many other contexts with a history of ethnic
violence shows. Third, distinguishing between race as xed, imposed, and
exclusionary, on the one hand, and ethnicity as uid, self-ascribed, and
voluntary, on the other hand, would not do justice to constellations (such
as among Serbs in Kosovo, Albanians in Serbia) where ethnic groups
experience degrees of forced segregation, exclusion, and domination usu-
ally associated with race. Thus, there is no clear-cut line between eth-
5
The list of authors who dene race as a special case of ethnicity includes Gordon
(1964); Wallman (1986, p. 229); Sollors (1991, chap. 1); Anthias (1992); Loveman (1997);
Patterson (1997, p. 173); Nagel (2003, chap. 2); and Banton (2003).
Ethnic Boundaries
975
nosomatic and other types of ethnicity that would justify establishing
entirely separate objects of analysis to be addressed with different ana-
lytical tools.
Perhaps it is useful to briey address the political worries that seem to
motivate opponents of an encompassing denition in the United States.
They argue that subsuming race as a particular form of ethnicity is part
of a sinister neoconservative agenda (Omi and Winant 1994, chap. 1)
meant to negate the role that racist ideologies have played in the colo-
nization of the world and to deny that racial exclusion might be relevant
in contemporary U.S. society and beyond (Bonilla-Silva 1999, p. 899;
Winant 2000, p. 179). However, an encompassing denition does not imply
that race no longer matters in the United States. Quite to the contrary,
it allows one to see how much it matters by situating the U.S. case in a
comparative horizon. Within that horizon, we will nd societies with
phenotypical variation among the population but without racialized
groups (Sanjek 1996, p. 56; Horowitz 1971), societies without phenotyp-
ical variation but racially dened groups in stark opposition to each other,
6
and nonracialized systems of ethnic differentiation that are as exclusionary
as race is in the United States. An encompassing denition not only allows
us to situate the U.S. experience better but also prevents us from mis-
interpreting the specic ethnosomatic order of this particular society as
a universal form of social organization and then projecting this form onto
other societies across the globe (see the philippica of Bourdieu and Wac-
quant [1999]; Bonnett 2006).
Having defended my denition of ethnicity, I will elaborate briey on
the notion of boundary used in this article. A boundary displays both a
categorical and a social or behavioral dimension. The former refers to
acts of social classication and collective representation; the latter to ev-
eryday networks of relationships that result from individual acts of con-
necting and distancing. On the individual level, the categorical and the
behavioral aspects appear as two cognitive schemes. One divides the social
world into social groupsinto us and themand the other offers
scripts of actionhow to relate to individuals classied as us and them
under given circumstances. Only when the two schemes coincide, when
ways of seeing the world correspond to ways of acting in the world, shall
I speak of a social boundary.
7
6
See the distinction between red humans and white humans among the Rendille
described by Schlee (2006, p. 82).
7
The best discussion of the relationship between the two dimensions of ethnicity is
still Mitchell (1974); with regard to the boundary concept in general, see Lamont (1992,
chap. 1). An example of a categorical distinction with few behavioral consequences is
the sharp moral boundary most contemporary Americans drawagainst atheists (Edgell,
Gerteis, and Hartman 2006).
American Journal of Sociology
976
A further clarication might be necessary in order to avoid a potential
misunderstanding. The concept of boundary does not necessarily imply
that the world is composed of sharply bounded groups. As I will show
below, ethnic distinctions may be fuzzy and boundaries soft, with unclear
demarcations and few social consequences, allowing individuals to main-
tain membership in several categories or switch identities situationally.
The concept of boundary does not imply closure and clarity, which vary
in degree from one society, social situation, or institutional context to
another. It represents one of the foremost tasks of the comparative study
of ethnicity to account for such varying degrees of boundedness.
FOUR DIMENSIONS OF VARIATION
The past decades of research have produced hundreds of ethnographic
studies, contrasting case comparisons, and historiographies of ethnic
groups and boundaries. Seen together, they offer a breathtaking panorama
on a variety of ethnic forms. Here, I review and organize this literature
by outlining four dimensions of variation along which an individual case
could be situated.
8
Each will bring to light different empirical and ana-
lytical challenges that the comparative study of ethnicity has so far failed
to address in a systematic way.
The Political Salience of Boundaries
The rst challenge is to understand why some ethnic boundaries are
politically salient while others are not. When boundaries are salient, po-
litical alliances are more likely to be formed between coethnics than be-
tween individuals on opposite sides of a boundary. In Switzerland, for
example, not a single political party, trade union, or major civil society
organization is organized on the basis of language (Wimmer 2002, chap.
8). In Northern Ireland, by contrast, politics is conceived as a matter of
ethnoreligious power relations, and political loyalties rarely cross the eth-
noreligious divide. How are we to explain comparatively such varying
degrees of political salience?
This question is relevant not only from a comparative perspective but
also from a case study point of view because many systems of ethnic
classication are of a multilevel character: they comprise several nested
segments of differentiationin contrast, for example, to gender classi-
8
For other attempts at laying out the dimensions of variability in ethnic forms, see
Horowitz (1971); Cohen (1981); Shibutani and Kwan (1965, pp. 4851). Arthur Stinch-
combe (2006) recently described general forms of variation in the features of social
boundaries.
Ethnic Boundaries
977
cations or ranked social estatesall of which might become the main
focus of political loyalty. The following example illustrates this widespread
feature of ethnicity. A Southern Californian may identify as Blue Hmong
as opposed to White Hmong, as Hmong in opposition to other persons
of Vietnamese origin, as Vietnamese in contrast to other Asian nation-
alities, as Asian-American in opposition to African-Americans and Euro-
Americans, or as American from a global perspective.
9
Which of these
potential lines of cleavage will be politically relevant, and which ones will
not?
Several attempts have been made to address the salience question. The
situationalist approach, developed by anthropologists working in com-
plex, plural societies (Okamura 1981, but see also Galaty 1982),
10
offers
a straightforward answer: the salience of the various levels of differen-
tiation depends on the logic of the situation and the characteristics of the
persons interacting. Thus, in the example above, a political activist will
emphasize his Blue Hmong identity when struggling with White Hmongs
over which groups cultural heritage will be recognized by the Californian
government. When traveling in Europe, he will be treated as and identify
with being American and will have to defend the foreign policy of his
government.
However, there are social forces beyond those emerging from specic
social contexts that make certain levels of categorical distinction more
important than others for a persons overall life chances. Whatever the
situational relevance of a Blue-White Hmong boundary, a persons as-
signment to the racialized category of Asian will be more important for
college ofcers when they decide whom to admit to their programs or for
political entrepreneurs who design electoral strategieseven if his per-
sonal identity may situationally be dened in other terms (Kibria 2002,
chap. 3). Following Despres (1975) and others in the pluralist school, we
may thus want to identify those categorical cleavages that are the most
consequential and salient for the overall structuring of political relations
in a society. The framework outlined in later sections will identify these
social forcesinstitutions, power, and networksthat are most likely to
produce such effects of structuration, to borrow Anthony Giddens term.
A second approach derives the salience of ethnic categories from the
dynamics of economic competition. Ethnic boundaries that correspond to
groups in competition on the labor market will be politically more relevant
9
For a discussion of this aspect of ethnicity, see Moerman (1965); Keyes (1976); Oka-
mura (1981); Galaty (1982); Jenkins (1997); Brubaker (2004, chap. 2); Waters (1990,
pp. 5258).
10
Compare also the contextualist arguments in Cornell and Hartman (1998, chap.
6); Jenkins (1997, pp. 6370).
American Journal of Sociology
978
than those that cut across lines of economic interest. This solution to the
salience problem is at the core of the only genuinely comparative tradition
in the eld of ethnic studies, stretching from Abner Cohens work in the
1970s to Amy Chuas recent best-seller (see Cohen 1974; Patterson 1975;
Banton 1983; Bonacich 1974; OSullivan 1986; Olzak and Nagel 1986;
Chua 2004; Chai 1996, 2005 ). Competition theory indeed helps to un-
derstand the situation of trading minorities for which ethnic networks
represent a considerable advantage in the provision of cheap credit and
labor (see Landa 1981; Ward and Jenkins 1984; Boissevain et al. 1990;
Wintrobe 1995). The broader claims, however, proved to be problematic.
The economic structures of labor markets are poor predictors of where
the most salient fault lines in the ethnic landscape come to lie, as the
following two examples illustrate.
Olzak (1993) studied U.S. cities during the high tide of immigration
before World War I, to conrm the competition argument. However, in-
creasing job segregation and reduced competition between African-Amer-
ican immigrants from the South and the established labor force did not
decrease the salience of the black-white boundary. Quite to the contrary,
most of the violence was directed against black migrants rather than those
from Europe (Lieberson 1980), even though it was the latter who increas-
ingly competed for the same jobs as local Euro-Americans.
11
A recent
study by Dina Okamoto (2003) also nds results that directly contradict
the predictions of competition theory: higher degrees of occupational seg-
regation between Asian-Americans and others increases the likelihood of
pan-Asian mobilization, while more competition decreases such
mobilization.
It seems that economic competition theory does not help to understand
who is seen as a legitimate competitor and who is not. The dynamic of
ethnic boundary formation follows a political logic that cannot be derived
in any straightforward way from economic incentive structures.
12
More
often than not, the distinction between legitimate and illegitimate com-
petitors in the economic eld maps onto that between national majority
and minoritya thesis to which I will return.
A third answer to the problem of salience is provided by scholars who
believe that the visibility of ethnic markers determines which cleavage
will be the most relevant for social interactions and political life. Various
authors (Hale 2004; van den Berghe 1997) have maintained that differ-
11
For more extensive empirical critiques of the competition argument, see Horowitz
(1985, pp. 10535) regarding the trading minority model and Be langer and Pinard
(1991) and Wimmer (2000) regarding labor market competition theory.
12
This point has been made by Belanger and Pinard (1991) and by Espiritu (1992,
chap. 1).
Ethnic Boundaries
979
ences in physical appearance are more likely to be used to drawboundaries
because they are easy to recognize and thus cognitively economical. Ac-
cording to another group of authors, racialized boundaries originated in
colonial conquest, slavery, and postemancipation segregation and thus
will be more politically salient than the less exclusionary boundaries be-
tween ethnic groups (Isaac 1967; Omi and Winant 1994; Bonilla-Silva
1996; Cornell and Hartman 1998). This is certainly a reasonable assess-
ment of the contemporary situation in the United Statesbut it proves
to be difcult to generalize once we enlarge the horizon both historically
and cross-nationally.
13
In the history of American ethnic and race relations, groups such as
Jews (Saks 1994), Irish (Ignatiev 1995), and Italians (Guglielmo 2003) that
were once considered to be phenotypically ambivalent and probably even
belonging to other races are now considered white ethnics. The per-
ception of racial difference and associated practices of racial discrimi-
nation seem to shift over time and do not depend on objective phe-
notypical appearance alone. In other plantation societies of the NewWorld
that do not know the American one drop rule, the location of boundaries
on the somatic continuum varies even more. In Puerto Rico, the denition
of white expanded considerably over time to include individuals of
mixed background previously considered colored (Loveman and
Muniz 2006). In Brazil, the classication of similar-looking individuals
into ethnosomatic types varies according to a number of contextual factors
(Sansone 2003, chap. 1). In Colombia, people with the same somatic fea-
tures might be black in one region of the country (Wade 1995) but not
in another (Streicker 1995). The difculties of deriving the salience of
boundaries from racial differences appear even more clearly if we com-
pare across societies. As Hoetink noticed some time ago, one and the
same person may be considered white in the Dominican Republic or
Puerto Rico, . . . coloured in Jamaica, Martinique or Curacao . . . [and]
may be called a Negro in Georgia (Hoetink 1967, p. xii).
Social Closure and Groupness
A second challenge is to understand which ethnic boundaries are relevant
for the structures of social networks and the access to resources that they
enable. Some ethnic groups have rmly closed themselves off against
outsiders. In other cases, relationships oweasily across ethnic boundaries.
Sometimes, ethnic boundaries are associated with high levels of discrim-
ination and exclusion; sometimes they do not matter for hiring and ring,
marrying and divorcing, befriending and feuding. What is the best way
13
The best discussion of this remains Horowitz (1971, pp. 24044).
American Journal of Sociology
980
to organize and describe such various social consequences of ethnicity?
Richard Jenkins (1994) proposes to distinguish between an ethnic category,
which may be entirely imposed by powerful outsiders and is associated
with high degrees of discrimination and exclusion, and an ethnic group
based on self-identication and a shared sense of belonging.
14
However, the distinction between group and category is not one of
principle, as Jenkins notices, because imposed categories may over time
be accepted as a category of self-identication and thus transformed into
a group. The black-white divide in U.S. society, to give an example, has
been reproduced despite immigration from countries where other modes
of classication prevail. Second-generation immigrants from Cape Verde
(Ito Adler), Haiti (Woldemikael 1989), and working-class children of West
Indian immigrants living in innercity neighborhoods (Waters 1999) come
to identify with the imposed category of blackwhile their parents still
vehemently emphasized their national identity in order to counter the
stigma of blackness.
15
If ethnicity can be both a categoryimposed by outsidersand a
groupembraced by its membersa dichotomous distinction obviously
looses its value. We might want to replace it with a continous variable.
A good starting point is Max Webers discussion of ethnic group formation
as a process of social closure (cf. Loveman 1997). High degrees of closure
imply that a boundary cannot be easily crossed and that it is consequential
for everyday life because it denies access to the resources that have been
monopolized by the dominant group.
16
Social closure does not occur ex-
clusively in such hierarchical relationships, however, but may be of a
more symmetric nature, as when Indian peasant villages in Mexico each
control their own piece of communal land and deny access to outsiders
(Wolf 1957). To be sure, social closure is not a universal feature of ethnic
14
On the distinction between group and category (i.e., individuals sharing an ethnic
trait), see also McKay and Lewis (1978). A Nepalese example nicely illustrates what
Jenkins means by ethnic category: The majority of Rajopadhyaya Brahmans of the
Katmandu valley, Gellner writes, do not today see themselves as Newars, do not
call themselves Newars, do not speak Newar to their children, and to not support
Newar ethnic activism. Yet they are seen as Newars by many others, an identication
. . . which they themselves reject (Gellner 2001, p. 6).
15
The mechanisms that lead to the internalization of imposed boundaries are well
known from social psychology. Several studies have shown that low-status group
members are more likely to identify with their own category when the boundaries are
perceived as impermeable (Mummendey et al. 1999); another line of work demonstrates
that high prejudice leads to more identication with ones group as a rst step of
establishing a positive self-concept (Branscombe, Schmitt, and Hervey 1999).
16
In such contexts, the theory of identity choice, as developed by Patterson (1975),
Lustick (2000), or Laitin (1995a), is of little help, because the choices by individuals
placed in subordinate categories are much less consequential for their own lives than
the ones made by more powerful actors.
Ethnic Boundaries
981
group boundaries. The literature offers a range of ethnographic examples
where no such closure has occurred and where permeable boundaries are
of little consequence for access to resources.
17
We are thus well advised
to distinguish between different degrees of closure and to try to understand
under which conditions these emerge.
Another dimension of variation follows from this. Depending on the
degree of closure, ethnic boundaries may or may not separate groups
in the sociological sense of the term, implying widely shared agreement
on who belongs to which category, as well as some minimal degree of
social cohesion and capacity for collective action. Ignoring this variability,
many authors have fallen back into a groupist default language, to use
Rogers Brubakers term (Brubaker 2004). These authors assume, rather
than demonstrate, that an ethnic category represents an actor with a single
purpose and shared outlook.
18
Such ontological collectivism overlooks,
however, that ethnic categories may shift contextually and that there
might be substantial disagreement among individuals over which ones
are the most appropriate and relevant ethnic labels.
19
The list of well-
documented examples is quite long.
20
In such contexts, we may well speak
of ethnicity without groups (Brubaker 2004) or ethnicity without bound-
aries. Moermans description of the uid, fuzzy, and overlapping modes
17
On the identity choice among white Americans, see Waters (1990); for the back and
forth switching between Tatar and Bashktiar categories in Tatarstan, see Gorenburg
(1999); for the change of self-identication in Latin America, see Lancester (1991) and
Wade (1995).
18
See also the critique by Chai (1996).
19
For a recent example of groupist analysis, see Ross (2001).
20
Gorenburg (2000) reports that the identication with Tartar nationalismvaries across
occupational groups; Sanjek (1981) describes how individuals group tribal-ethnic cat-
egories in different ways in urban Ghana; according to Starr (1978), who did research
in prewar Beirut, the classication of an individual depends on the context of inter-
action and the ethnic characteristics of the classifying person; Levine (1987) reports
how different systems of ethnic and caste classications in Nepal may be used in
different contexts; Berreman (1972) arrived at similar ndings regarding ethnic and
caste classication in North India; Labelle (1987) shows that the use of ethnoracial
labels in Haiti varied, among other things, by social class; in Nicaragua, it depends
on how formal the situation of interaction is (Lancester 1991); Harriss (1980, chap.
5) research in Brazil found widespread disagreement in the use of ethnoracial categories
for the same persons and even different classications for siblings; research by Landale
and Oropesa (2002) highlights the varied strategies of self-identication of Puerto
Ricans in the United States. To make things even more complex, some ethnographic
studies have shown that even the self-classication by individuals may be context
dependent and variable (e.g., Jime nez [2004] on contemporary Californians of Mexican
and white parentage; Campbell, Lee, and Elliott [2002] on northeastern China under
the Qings; Nagata [1974] on urban Malaysia; Mayer [1962] on rural migrants in urban
South Africa; Waters [1990, pp. 3638] on suburban white ethnics in the United States;
Russell [1997] on the Yahka of East Nepal).
American Journal of Sociology
982
of ethnic classication in northern Thailand represents the locus classicus
for this assertion (Moerman 1965).
I would like to note, again, that these examples represent one end of
a continuum only. An equally diverse sample could be cited as support
for the opposite proposition: that ethnic boundaries are drawn unambig-
uously, are relevant for many different domains of everyday life, are agreed
upon by a vast majority of individuals, and form the basis for collective
action and resource mobilization. In Gil-Whites example from Mongolia,
there is little disagreement among his interviewees that a Mongol is a
Mongol even if born from a Kazah mother and brought up among Kazahs
(Gil-White 1999).
21
Northern Ireland could be cited as another society
where variation in the use of ethnoreligious categories is rather limited,
the consequence of a long history of segregation, endogamy, and conict
(Ruane and Todd 1996).
22
Various scholars have observed that classi-
catory variability and ambiguity are greatly reduced through violence and
war (most explicitly, Smith [1981] and Appadurai [1998]). Who are the
Albanians? to paraphrase the title of Moermans article, is maybe too
easy a question to deserve an answer in present day Kosovo. Given this
wide spectrum of variation, it is certainly useful to distinguish between
various degrees of groupness, as Jenkins (1997, p. 50) put it, and to
attempt to explain these comparatively.
23
Cultural Differentiation
Contrary to Barths famed dictum that it is the boundary that matters
in ethnic relations and not the cultural stuff they enclose (Barth 1969b,
p. 15), a number of authors, including Barth (1994) himself some 30 years
later, have noted that this stuff may indeed make a difference. In the
landscape of cultural variation, to use a metaphor coined by Tim Ingold
21
A similar argument centering around the notion of participants primordialism is
offered by Roosens (1994). His example are rst-generation Spanish immigrants in the
Netherlands.
22
This does not preclude, obviously, a great deal of dissent over the meaning and
political implications of those boundaries, as the Northern Ireland example illustrates.
Ethnographic research shows that there is space for local negotiations over the im-
plications of the religious divide in daily interactions (Harris 1972; Burton 1978).
Individuals may blur one categorical dimension of the boundary (e.g., by associating
with Catholics in a sports club), as long as they are straight on other dimensions (e.g.,
not dealing with anybody with open sympathies for the IRA).
23
Allowing for the existence of ethnic groups does not imply ontological collectivism:
they might be thought of as aggregate consequences of individual-level processes and
mechanisms (see the discussion in Wimmer [2007]; for a useful distinction between
ontological and methodological collectivism/individualism, see Hedstro m [2005, pp.
7074]).
Ethnic Boundaries
983
(1993), we may observe discontinuities and ruptures: a graben between
tectonic plates, or an abrupt change in soil composition and vegetation,
to push the geological metaphor. Ceteris paribus, we expect that ethnic
boundaries will follow some of these more dramatic cultural ruptures,
such as those brought about by long-distance migration or conquest.
24
We
would indeed be surprised if rst-generation Chinese merchants in Ja-
maica would not see themselves and be perceived by Afro-Caribbeans as
ethnically differentat least among the rst generation of immigrants.
25
If cultural difference and ethnic boundaries do coincide in this way,
they can reinforce each other in a two-way process. Cultural differenti-
ation may make a boundary appear quasi natural and self-evident, while
social closure along ethnic lines may reinforce such differences through
the invention of new cultural diacritics,
26
such as when Chinese traders
in Jamaica converted to Catholicism to set themselves apart from the rest
of the population and stabilize the boundary (Patterson 1975).
However, this again only represents one end of a continuum. In other
constellations, ethnic boundaries do not divide a population along obvious
cultural lines but unite individuals who follow quite heterogeneous cul-
tural practices. Examples include multilingual, multireligious national
communities such as the Swiss who managed, to the bewilderment of
observers such as Tocqueville, John Stuart Mill, Ernest Renan, Max We-
ber, and Karl Deutsch, to develop a strong sense of belonging and to draw
sharp boundaries toward immigrants from neighboring countries (Wim-
mer 2002, chap. 8). Another example is the Maconde, who are perceived
and perceive themselves as a distinct ethnic group despite vast cultural
differences between migrants from Mozambique and town dwellers in
Tanzania and despite the fact that they are divided into endogamous
24
Max Weber saw migration and conquest as prime forces of ethnic group formation
(Weber 1978, pp. 38598; see also Keyes 1981). Schermerhorn (1970) adds the emergence
of pariah groups and indigenous isolates in settler societies to the list of ethnicity
generating dynamics.
25
See the novel on a Chinese trader in Jamaica by Powell (1998); on the Chinese in
Mississippi, see Loewen (1971).
26
This argument has been made by different authors and in different analytical lan-
guage. Bentley has used Bourdieus habitus theory to explain why cultural differences
easilyyet not automaticallytranslate into perceptions of ethnic difference (Bentley
1987; cf. also Wimmer 1994). Cornell argues that if an ethnic groups identity is
primarily built around shared values, as opposed to shared interests, this culture may
act as a lter for the perception of interests and thus inuence the strategies of
boundary maintenance (Cornell [1996]; cf. Barth [1994]; the lter argument can also
be found in Keyes [1981]). Hale takes a cognitive perspective and argues, in a neo-
Deutschean mode, that communication barriers such as those represented by language
differences will make it more likely that individuals nd the boundary meaningful
and will use the corresponding linguistic markers as clues to make cognitive sense of
the social world and reduce uncertainty (Hale 2004).
American Journal of Sociology
984
castes (Saetersdal 1999).
27
Finally, where ethnic boundaries originally did
coincide with cultural difference, the boundary may nevertheless be
blurred subsequently and eventually break down completelysuch as
among the Chinese in Guyana (Patterson 1975) or Cuba (Corbitt 1971)
and countless other cases of assimilation.
Stability
A nal challenge for the comparative understanding of ethnicity is that
some groups and boundaries are tenacious and change only slowly, over
the course of many generations, while in other contexts, substantial shifts
in the ethnic landscape may occur during the lifespan of an individual.
28
It seems that ethnic boundaries cannot always be redened or changed
ad libitum, as radically constructivist interpretations of Barths writings
suggested. Following Katherine Verdery, we would be well advised to
situate the situationalims of radical constructivism (Verdery 1994).
It seems that the degree of stability is linked to various modes of trans-
mitting ethnic membership. The most stable boundaries are found among
peoples who identify individuals through multigenerational, unilineal de-
scent lines, such as among Mongols, Pathans, Jews (Gil-White 1999), and
Germans. More unstable boundaries, one could argue, are those dened
by behavioral, rather than genealogical, membership criteria. Among the
Vezo of Madagascar, for example, one is considered being Vezo if one
behaves like a typical Vezo and lives the lifestyle of a Vezo, indepen-
dent of the ethnic background of ones parents (Astuti 1995).
29
Whatever the correlates are of more or less stable boundaries, the con-
27
Other examples would include the Tat in Dagestan, which include Christian, Jewish,
and Muslim sections; the Karen of Thailand and Burma, which comprise adherents
of Protestantism, Catholicism, animist religions, Buddhism, and several syncretist re-
ligions (Keyes 1979); Kachin groups in northern Burma that speak Jinghpaw or Lisu
(cf. Leach 1954), or the Hadiyya in Ethiopia which comprise Muslim, Protestant, and
Catholic sections (Brauka mper 2005).
28
Examples are the melting away of Yugoslavians from the 1980s onward, the swell-
ing of the ranks of self-identied Indians in the wake of the red power movement in
the United States (Nagel 1995), the identity shifts between Han and Manchu in the
eastern provinces of China under the Qing (Campbell et al. 2002), similar oscillations
between Tatar and Baskir categories in central Russia during Soviet rule (Gorenburg
1999), and the spectacular spread of the Chetri caste in Nepal through intermarriage
(Ramble 1997).
29
This echoes the discussion of open vs. closed citizenship regimes, which allow for
more or less easy naturalization of immigrants and thus more or less stable boundaries
between nationals and foreigners. Access to citizenship is easier, it has been argued,
when the nation is dened in terms of political behavior; national boundaries are more
stable and impermeable, on the other hand, where membership is dened by ancestry
(Brubaker 1992; Alba 2005).
Ethnic Boundaries
985
trast between ethnic categories that perdured over thousands of years
the Jewish community being the most prominent exampleand those
that have been invented, adopted, and forgotten within a generation, such
as the Ciskeian nation of the apartheid era (Anonymous 1989), is striking
enough to ask for a comparative explanation.
Intellectual Genealogies
So far, I have shown that explaining different degrees of political salience,
social closure, cultural differentiation, and historical stability represents
a major challenge for the comparative sociology of ethnic group formation.
I have also demonstrated that the existing literature offers little help in
addressing this task. In what follows, I outline a theoretical framework
that might represent a rst step toward an analytically more sophisticated
theory that allows one to explain the wide range of ethnic forms that the
comparative literature has brought to light. The model draws inspiration
from three research traditions.
The rst goes back to Max Weber, who conceived ethnicity as a mode
of drawing boundaries between individuals and thus creating social
groups. This focus on group making stands in opposition to studies of
collective identity in social psychology (cf. Le Vine and Campbell 1972,
pt. 3; Scheff 1994) and of group relations in both sociology (Pettigrew
1980; Banton 1983) and social psychology (Dovidio, Glick, and Rudman
2005), which all take for granted thatrather than explain whya society
is divided along ethnic lines. The genealogy of this boundary and group-
making approach leads from Weber to Fredrik Barth (1969a), Miche`le
Lamont (2000), Mara Loveman (1997), Richard Alba (2005), Charles Tilly
(2006), and other contemporary writers.
The second tradition is the study of ethnicity as the outcome of a
political and symbolic struggle over the categorical divisions of society.
This line of research was initiated by political anthropologists of the Man-
chester school such as Boissevain, Turner, and others, and later canonized
and popularized by Pierre Bourdieu (1991). Today, it includes Bentley
(1987), Loc Wacquant (1997), Rogers Brubaker (2004), Wimmer (1994,
2004), and many others who rely on Bourdieus framework. Some an-
thropologists and historians work along similar lines but have adopted
Antonio Gramscis concept of hegemony (Mallon 1995; Comaroff and
Comaroff 1992; Omi and Winant 1994), to be discussed later on in this
essay. Both the Bourdieusian and the Gramscian strands developedlargely
as a response to, and in opposition to, a range of other theories that
continued the Herderian line of thinking in conceiving ethnicity as an
identity based on the shared culture and values of a group.
The third intellectual source for this project is the emerging institu-
American Journal of Sociology
986
tionalist tradition in the study of ethnic politics. Institutions provide in-
centives for actors to draw certain types of boundariesethnic rather
than class or gender, for exampleand to emphasize certain levels of
ethnic differentiation rather than others. While some have emphasized
macropolitical institutional transformations, such as the shift fromindirect
to direct rule (Hechter 2004) or the spread of the nation-state form (Bru-
baker 1996; Meyer et al. 1997; Wimmer and Min 2006), others have looked
at mesolevel and microlevel institutional mechanisms that lead actors to
emphasize certain ethnic boundaries over others (Posner 2005; Koopmans
et al. 2005). This institutionalist approach contrasts with various micro-
sociological traditions that see ethnic boundaries as emerging from the
minutiae of cognition, action, or interaction, variously conceived as con-
versational encounters (as in the ethnomethodologist tradition pursued by
Day [1998]), performative enactments (Sharp and Boonzaier 1994), ratio-
nal choices (e.g., Kuran 1998) or the cognitive processing of information
(Fryer and Jackson 2003).
ELEMENTARY STRATEGIES OF ETHNIC BOUNDARY MAKING
This article draws these three traditions of research together and integrates
them into a unied theoretical framework. It derives the topography and
character of ethnic boundaries from the institutional structures, the net-
work of alliances, the distribution of power, and the dynamics of repre-
sentational politics that they shape. The model is presented in several
steps. The rst one, to be undertaken in this section, consists of taking
stock of the various possible strategies of ethnic boundary making that
may be pursued by different actors in different social contexts. Summa-
rizing a diverse empirical literature, I distinguish between ve types of
such strategies:
30
those that seek to establish a newboundary by expanding
the range of people included; those that aim at reducing the range of the
included by contracting boundaries; those that seek to change the meaning
of an existing boundary by challenging the hierarchical ordering of ethnic
categories; those that attempt crossing a boundary by changing ones own
categorical membership; those that aim to overcome ethnic boundaries
by emphasizing other, crosscutting social cleavages through what I call
30
The typology has been inspired by Lamont and Bails work on destigmatization
strategies (Lamont and Bail 2005); Zolberg and Woons (1999) distinction between
boundary crossing, blurring, and shifting; and Donald Horowitzs (1975) discussion of
amalgation, incorporation, division, and proliferation as strategies of categorical fusion
and ssion. In the sociopsychological literature on attitude change, a similar distinction
between de-categorization, common ingroup identity, and intergroup differentia-
tion is being made (Gonzalez and Brown 2003).
Ethnic Boundaries
987
strategies of boundary blurring. In the following paragraphs, I present
the bare bones of this typology, referring the reader to another article on
the subject for more details (Wimmer 2007).
Shifting Boundaries through Expansion
Actors may create a more encompassing boundary by grouping existing
categories into a new, expanded category. Examples from across the world
at different levels of aggregation and from various historical periods can
be found in the literature. Many modernizing empires have created, from
the 18th century onward, larger ethnic minorities out of smaller groups
in order to tighten and centralize the system of indirect rule over their
subjects. Similarly, colonial authorities grouped various previously in-
dependent tribes and other local communities into larger ethnic entities,
often by appointing chiefs or other representatives, as numerous works
in colonial historiography have shown. Not all such strategies of minority
making, it should be noted, have been successful in creating durable and
salient boundaries.
Other examples of ethnogenesis come from newly nationalizing states
after independence. Several of the ethnic categories that imperial admin-
istrations had created were further grouped into ethnoregional blocks by
politicians who attempted, with varying success, to establish a larger
political base in order to compete more successfully in the new national
political arena, as a whole tradition of research in political anthropology
has shown. Similarly, nationalizing states in the West have grouped var-
ious minorities into larger entities for the purpose of administration and
some would sayhegemonic control.
Perhaps the most consequential form of boundary expansion in the
modern world is nation building: making French out of peasants, Prov-
encales, and Normands; Brazilians out of whites, blacks, and browns;
Jamaican Creoles out of Afro-Caribbeans, Europeans, and Chinese; and
so forth. Not all such strategies, it should again be noted, have been
successful.
Shifting Boundaries through Contraction
The opposite strategy is to promote narrower boundaries than those al-
ready established in the social landscape. Ethnic localism may be an
especially attractive strategy for individuals and groups that do not have
access to the centers of communication and whose radius of action remains
conned to narrower geographic spaces. The indigenous groups of Mexico
provide a good example. Their social world was once dened by imperial
polities that had established wide areas of cultural commonality. After
American Journal of Sociology
988
conquest and the dismantlement of these empires and kingdoms, indig-
enous peasants started to draw an ethnic boundary separating their mu-
nicipality, the new center of their political, social, and spiritual universe,
from the rest of the worlda formidable symbolic weapon against the
claims to exclusivity and cultural superiority that the Spanish-speaking
elites made when distinguishing themselves as gentes de razo n (people
of reason) from the indigenous majority as gentes naturales (Wimmer
1995).
Another example of boundary contraction is the insistence, among mid-
dle-class, second-generation Chinese and Koreans in Los Angeles, that
they should be referred to and treated as Chinese-Americans and Korean-
Americans rather than lumped together under the term Asian (Kibria
2002)similar to immigrants from the West Indies who ght for recog-
nition as Jamaicans and Trinidadians in order to avoid being cate-
gorized as black (Waters 1999).
Inversion
In contrast to expansion and contraction, the strategy of normative in-
version does not target the location of the boundary but the hierarchical
ordering of ethnic groups. The category of the excluded and despised
comes to designate a chosen people who are morally, physically, and cul-
turally superior to the dominant groupthe Umwertung der Werte
(transvaluation) that Nietzsche so profoundly detested. Examples of nor-
mative inversion abound. The most widely known in the Western world
is probably the cultural nationalism among African-Americans in the
United States and the African nationalism in South Africa. To be sure,
not all attempts at inversion were successful, and not all despised and
dominated groups have developed such powerful political movements.
Repositioning
Repositioning describes a strategy in which the principles of hierarchy
are not contested (as they are in normative inversion) nor are boundaries
expanded or contracted. Rather, an actor seeks to change her own position
within an existing hierarchical boundary system. Status change may be
pursued individually or, much less often, by repositioning ones entire
ethnic category within a multitiered hierarchy. Assimilation and passing
are the main strategies for individuals to shift sides and escape a minority
stigma. Both can be found in an enormous variety of social contexts,
including among contemporary immigrant minorities, Jewish converts in
19th-century Europe, Dalit groups embracing Islamin prewar India, Mex-
ican Indians after the revolution, and Polish workers in prewar Germany.
Ethnic Boundaries
989
The best example of collective repositioning is perhaps what anthro-
pologists have called caste climbing. By adopting the lifestyle of the
upper castes and strategically demanding certain jajmani services from
members of other castes (a central feature of local caste systems), a group
may slowly acquire a higher standing in the ritual hierarchy (Bailey 1969).
Other examples are the Chinese of Mississippi who managed, although
originally classied as nonwhite, to cross the caste boundary onto the
other side (Loewen 1971).
Blurring Boundaries
Boundary blurring aims to overcome ethnicity as a principle of catego-
rization and social organization altogether. Other, nonethnic principles are
promoted in order to undermine the legitimacy of ethnic, national, or
racial boundaries. Examples include such explicitly antinationalist or-
ganizations as the Communist International, radical Islamic movements
that dream of the restoration of a supranational caliphate, some forms of
transnational and transethnic feminism, as well as less organized, less
politically salient forms of boundary blurring. Transethnic localism rep-
resents a good example for the latter, such as in Sophiatown in the 1950s,
a township outside Johannesburg (Hannerz 1994). Africans, Jews, and
immigrants had formed what they perceived as a cosmopolitan culture
inspired by American Jazz, British fashion, and continental literary styles.
They saw this urban lifestyle, at least in part, as a counterculture that
would stand against the racial classication and segregation imposed by
an emerging apartheid regime.
Emphasizing civilizational commonalities is another way to blur ethnic
boundaries. Perhaps the most politically salient example is to underline
membership in one of the world religions, especially Islam, Buddhism, or
Christianity. Out of many possible cases, I may cite a recent study on
British Pakistanis whose Muslim identity is more salient in daily life than
the Pakistani category ofcially assigned to them by the state (Jacobson
1997). Similarly, Maghrebinian immigrants in France emphasize their
membership in the umma, rather than their national origin or immigrant
status (Lamont, Morning, and Mooney 2002).
Even more encompassing boundaries are drawn when individuals pur-
sue what Lamont has called universalizing strategies. Universal moral
qualities and membership in the human family are often evoked, so it
seems, by the most excluded and stigmatized groups, such as working-
class African-Americans (Lamont 2000) or African immigrants in France
(Lamont, Morning, and Mooney2002), refugees from Kosovo in Central
Europe (Karrer 2002, chap. 12), or Muslim caste groups in Hyderabad
(Ali 2002).
American Journal of Sociology
990
INSTITUTIONS, POWER, AND NETWORKS
Actors are obviously not free to choose whatever strategy they like best
whether to invert the normative hierarchy or simply cross the boundary
into the dominant group. In this section, I discuss three types of constraints
that all derive from the structures of the social eld within which actors
are situated. They are constrained, rst, by the institutional environment
that makes it appear more plausible and attractive to draw certain types
of boundariesethnic, class, regional, gender, tribal, or others. Second,
the distribution of power denes an individuals interests and thus which
level of ethnic differentiation will be considered most meaningful. Third,
the network of political alliances will inuence whom the boundaries will
include and who will not be counted as one of us.
Institutions
Institutional frameworks specify the historical context within which the
dynamics of ethnic boundary making unfolds. Much has been written
about world historical trends that have shaped these contexts: the colo-
nization of the non-Western world, the racialization of its populations
(Balibar 1988), and at the same time its division into ethnic domains
(Mamdani 1996); the role of forced labor and slavery in the making of
the Americas and the various ethnosomatic constellations that it produced
(Patterson 2005); and the spread of the nation-state in the postcolonial
era and the ways in which this has transformed the dynamics of ethnic
politics (Wimmer and Min 2006). The model to be outlined here treats
these world historical developments as exogenous. It focuses on how par-
ticular types of political institutionswhatever the macrohistorical pro-
cesses that led to their emergence and global spreadshape the strategies
of ethnic boundary making that actors pursue.
More specically, I focus on the peculiarities of the institution of the
nation-state, which dominates contemporary politics across the globe. An
analysis of the incentives that it provides for ethnic politics offers a crucial
starting point to understand why much of contemporary politics is about
drawing, maintaining, and shifting the boundaries of ethnicity, race, or
nationhood. This argument draws upon a growing tradition of research
that looks at the interplay between nation building and the making of
ethnic minorities (Young 1976; Williams 1989; Verdery 1994; Wimmer
2002; Mann 2005).
While it would be exaggerating to maintain that empires or premodern
territorial states were not at all interested in shaping and policing ethnic
boundaries, the change from empire to nation-state provided new incen-
tives for state elites to pursue strategies of ethnicas opposed to other
Ethnic Boundaries
991
types ofboundary making.
31
On the one hand, the principle of ethnon-
ational representativity of governmentthat like should rule over likes
became de rigueur for any legitimate state. It provided the main insti-
tutional incentives for state elites to systematically homogenize their sub-
jects in cultural and ethnic terms, usually by expanding the boundaries
of their own group and declaring their own ethnic background, culture,
and language as forming the national pot into which everyone else should
aspire to melt. On the other hand, the nation-state also needs to dene
its territorial boundaries in ethnic terms. The transethnic, universal prin-
ciples of imperial rulein the name of Allah, the spread of civilization,
revolutionary progressmeant that the boundaries of a polity were never
dened in ethnonational terms. In modern nation-states, however, only
territories populated by the nation should be integrated into the polity.
Dening the ethnic boundaries of the nation therefore is of central political
importance, and state elites are encouraged to pursue the strategies of
nation building and minority making outlined above.
The nation-state also provides institutional incentives for nonelites, es-
pecially political entrepreneurs among ethnic minorities, to emphasize
ethnic rather than other social divisions. The principle of ethnonational
representativity can be turned on its head by applying it to the minorities
themselves. Minorities can thus be transformed, through a strategy of
normative inversion, into nations (Wimmer 1993). Evoking the logic of
ethnonational representativity, they can demand an independent state for
their own group or at least fair representation within an existing state
to have the minority culture respected and honored in national museums,
to have its language recognized as an ofcial idiom to be taught in schools
and universities, and so forth.
For the population at large, the nation-states also provides incentives
to pursue ethnic boundary-making strategies: majority members might
discriminate against minorities in the day-to-day interactions on the job,
marriage, and housing markets and feel justied, if not encouraged, to
do so because they have become dignied as representing the people of
a particular state and thus entitled to a privileged seat in the social theater.
They might enforce the boundary toward minorities or encourage bound-
ary expansion by assimilating minority members into the national family.
Minorities are encouraged to cross the boundary into the national majority
and pursue strategies of passing and assimilation that will overcome the
consequences of the new structure of exclusion and discrimination or, to
the contrary, to divert the stigma associated with their minority status
31
I explore the relation between the nation-state and politicization of ethnicity ingreater
detail in Nationalist Exclusion and Ethnic Conicts (Wimmer 2002).
American Journal of Sociology
992
through boundary blurring, emphasizing the village, the continent, or
humanity as the main focus of identity and source of human dignity.
The ethnic logic of the nation-state thus shapes the boundary-making
strategies of many actors and comes to permeate many different social
elds. The precise way in which the boundary between the nation and
its various others are drawn varies substantially from society to society,
as a large body of comparative research has shown (most recently Bail
[in press]). The nature of this boundary then determines the kind of the
claims that ethnic minorities make in the public domain. In Britain, the
racialized boundaries of the nation are reected in the ethnosomatic modes
of self-identication by migrant organizations, while none of the migrant
organizations in France portray their constituency as a racial minority
but instead emphasize their status as politically and legally excluded. In
the Netherlands, Germany, and Switzerland, national identities are more
prevalent while race as an identifying marker is almost absent from the
discursive repertoire of minority politicsconforming to the way the na-
tional majority denes its boundaries toward immigrant others (Koop-
mans et al. 2005, chap. 4).
32
The task remains to explain how these varying
denitions of the national boundary came into beinga topic to be ad-
dressed in the remainder of this section.
Before we proceed, however, two qualications are in order. The above
does not imply that the boundaries of inclusion and exclusion are drawn
along ethnic or national lines in all institutional elds and in all situations
(cf. Bommes 2004; Brubaker et al. 2007). In the emergency rooms of
hospitals in the contemporary United States, to give an example, dis-
tinctions based on ethnicity, nationality, or race are considered inappro-
priate, while distinguishing between bodies with life-threatening and non-
life-threatening injuries is part of the institutional routine. Outside
emergency rooms, however, when it comes to the treatment of diseases
that pose no threat to immediate survival, some hospitals may inquire
about the legal status of Spanish-speaking immigrants (see, e.g., Preston
2006) or may give black patients less care than Anglo-American patients
with similar health problems (Thomson 1997). It is a matter of empirical
analysis to determine how far the ethnonational master scheme of modern
society has penetrated these institutional domains in a particular case.
Second, other institutions also inuence the dynamics of ethnic bound-
ary making once modern nation-states have been established, producing
further variation across cases. Democratization politicizes and deepens
the boundary between national majority and ethnic minorities, as it pro-
vides additional incentives for politicians to appeal to the shared interest
32
For other research that shows how ethnic claims making depends on institutionalized
opportunity stuctures, see Ireland (1994) and Okamoto (2006).
Ethnic Boundaries
993
of the people and unravel the machinations of its ethnic enemies (for
evidence, see Manseld and Snyder [2005]). The shift from one-party
regimes to democratic multiparty governments may also entail incentives
to emphasize other levels of ethnic differentiation hitherto of little political
signicance (Posner 2005). Similar effects can be observed when the in-
stitution of federalism is introduced (see the Ethiopian case study by
Brauka mper [2005]).
33
Power
Thus, the institution of the nation-state provides strong incentives for
elites and nonelites alike to emphasize ethnic rather than other types of
boundaries, using the various strategies outlined above. But there are
many different ways of drawing ethnic boundaries, because systems of
ethnic classications often entail various, segmentally nested levels of
differentiation, as noted above. Which level of ethnic differentiation an
individual will emphasize depends on her position in the hierarchies of
power that the institutional order establishes. The effects of power are
twofold.
34
First, an actor will prefer that level of ethnic differentiation that is
perceived to further her interests, given her endowment with economic,
political, and symbolic resources. The best model to understand this pro-
cess is the theory of frame selection offered by Hartmut Esser (Esser 2002;
Kroneberg 2005). It describes how actors rst choose a cognitive scheme
appropriate to the institutional environment and conducive to their per-
ceived interest and then the script of action most suitable to attain the
goals dened by the scheme. Depending on information costs and the
logic of the situation, both choices are made either in a fully conscious,
reexive mode of reasoning or in a semiautomated, spontaneous way. It
should be underlined that in this model the perception of interests is not
independent of the institutional environment and the cognitive frames
that have already been routinized. I will discuss such path dependency
effects later on in this article, focusing on the types of boundaries that
are more likely to produce them.
33
Supranational institutions provide other and sometimes contradicting sets of incen-
tives. On the effects of European Union conditionality on minority politics in Eastern
European candidate countries, see the literature cited in Kymlicka (2007, p. 41 n. 26).
On the the political opportunities offered by the supranational indigenous rights regime,
see Passy (1999).
34
I dene power by referring to the three Weberian/Bourdieusian dimensions of social
stratication: economic assets and income, the possibility to inuence other actors
choices even against their will (political power), and the Kapital der Ehre, the honor
and prestige associated with ones social standing.
American Journal of Sociology
994
But even where a particular ethnic boundary has already been estab-
lished and routinized in everyday cognition and action, individuals have
a choice between different interpretations and instantiations of the ethnic
scheme. They will choose that particular version that allows themto claim
an advantagous position vis-a` -vis other individuals of the same ethnic
category, as the following example illustrates. Miche`le Lamont and her
collaborators have accomplished a series of ethnographic studies on how
African-Americans draw social boundaries in order to counter stigmati-
zation and exclusion. Marketing specialists pursue a strategy of inversion
by emphasizing the power of consumption and the hipness of black
culture. They thus draw a line between insiders and outsiders that places
themselvesas experts in the production and consumption of things fancy
and as members of the black communityat the very top of the symbolic
hierarchy (Lamont and Molna r 2001). The highly educated and successful
upper-middle class, by contrast, stresses professional competence, intel-
ligence, and achievement as criteria to identify the morally and socially
superiorthus relying on the classic scheme of meritocracy to establish
equality between black and white (Lamont and Fleming 2005). Finally,
working-class African-Americans draw on religious universalism and un-
derline the value of caring personalities to emphasize that they belong to
the right side of the moral divide. Each of these groups thus interpretes
the black-white categories in such a way as to give legitimacy to their
own claims to moral worth and social standing and to place themselves
at the top of the prestige pyramid.
35
Second, the endowment with power resources not only determines
which strategy of ethnic boundary making an individual will pursue but
also how consequential this will be for others. Obviously enough, only
those in control of the state apparatus can use the census and the law to
enforce a certain boundary. Only those in control of the means of violence
will be able to force their ethnic scheme of interpretation onto reality by
killing Catholics, Shiites, or Furs, or resettling Tatars and Ger-
mans a` la Stalin, thus making Catholics, Shiites, Furs, Tatars, and Ger-
mans. Discrimination by those who control decisions over whom to hire,
where to build roads, and to whom to give credit is much more conse-
quential than the discriminatory practices of subordinate individuals and
groups.
35
Other examples could be cited to underline the point. Contrast the game of ethnic
identity choice that white, middle-class suburbanites in the United States are playing
(Waters 1990) with the rather anxious insistence on the relevance of the black racial
divide among their working-class peers (Lamont 2000). Many studies have shown that
educational background (or class status) explains most of the variance that we nd in
how sharply majority members draw a boundary toward minorities/immigrants (e.g.,
Betz 1994; Mugny et al. 1991; Semyonov et al. 2006).
Ethnic Boundaries
995
However, we should not overstate the hegemonic power of dominant
modes of ethnic boundary making. While powerful actors can make their
vision of the social world publicly known and consequential for the lives
of all, subordinates may develop counterdiscourses and other modes of
dividing the social world into groups than those propagated by the dom-
inant actors (cf. the notion of hidden transcripts by J. Scott [1990]).
Sometimes an imposed category is countered by a strategy of boundary
contraction: insisting on being Jamaican rather than black (Waters
1999), a Zinacanteco rather than Indio (Wasserstrom 1983). Sometimes
boundary expansion is the answer: being Muslim rather than a Pakistani
(Jacobson 1997) or a child of God rather than a black person (Lamont
2000). In still other contexts, boundary blurring is the counterhegemonic
strategy of choice: checking the other race box on the U.S. census (Al-
maguer and Jung 1999).
To recognize the possibility and existence of such counterdiscourses
or of resistance in more romantic termsis of crucial importance for
the model that is being proposed here. It helps to avoid equating strategies
of classication by powerful actors with the formation of groups in ev-
eryday life and thus allows a crucial question to be asked: under which
conditions do subordinate actors pursue counterstrategies, and under
which do they embrace the categorical distinction imposed upon them,
thus transforming the category into a group and the classicatory dis-
tinction into a social boundary? I return to this question below.
Networks
Institutional frameworks and power differentials explain if and what
strategies of ethnic boundary making actors will choose. They will adopt
ethnic classicationsrather than distinguighing between classes, gen-
ders, religions, villages, tribesif there are strong institutional incentives
to do so, and they will choose that level of ethnic differentiation and that
interpretation of an existing boundary that ensures that the individual is
a full member of the category of worthy, righteous, and dignied. But
where exactly will the boundaries between us and others be drawn?
Which individuals will be classied to which ethnic groups? Here, net-
works of political alliances come into play, the third characteristic of social
elds in the framework that I propose here.
36
I hypothesize that the reach of political networks will determine where
36
Such networks are in turn structured by the institutional framework (which denes
who actors are and what kind of resources they may use to pursue which types of
strategies) as well as the distribution of power (which inuences the possibilities for
forming stable alliances between persons with different resource endowments).
American Journal of Sociology
996
the boundaries between ethnic us and them will be drawn.
37
This can
be illustrated with examples of nation building. The political alliances of
state elites in the early periods of nation-state formation are most con-
sequential for the location of the boundary between nation and minority,
as the comparative literature shows. Anthony Marx (1999) explains how
different constellations of conict and alliance led to the inclusion of large
sections of the population of African descent into Brazils nation-building
project and to their exclusion in the United States and South Africa.
Modifying Marxs point slightly, we may argue as follows: when slavery
was abolished and restricted forms of democracy introduced, Brazils elite
relied on an extensive network of clientelist ties stretching far into the
intermediate class of mixed racial origin that had emerged in previous
centuries. In the United States, however, this intermediate class was com-
posed of Anglo-American peasants and tradesmen (Harris 1980, chap. 5),
and no transracial political ties had previously developed. Accordingly,
Brazils new political elites aimed at integrating and mixing peoples of
different racial origin,
38
while in the United States the nation was imagined
as white and mixing conceived and treated as a horribilum to be avoided
at all costs (Ringer 1983; Hollinger 2003).
39
The lack of well-established
transracial political networks helps explain why nation building in Amer-
ica was set off against the black population as its inner other, rather
than against the nation of competing neighboring states as in much of
Europe.
Similar lessons can be drawn from a least similar case comparison
involving Switzerland, Iraq, and Mexico (Wimmer 2002). It shows that
the reach of elite political networks in the early days of nation-state for-
mation determine which groups will be considered part of a national
project. In Switzerland, the new political elite relied on already established
civil society networks that stretched across French, German, and Italian-
37
A related hypothesis plays an important role in social movement research. It has
been shown that movements are mobilized along existing networks and that the rel-
evant boundaries become salient also on the level of identity and categorization (Bear-
man 1993; Zelizer and Tilly, in press). That the boundaries of networks and ethnic
categories coincide is one of the most important mechanisms explaining ethnic soli-
darity, as research in experimental economics has shown (Habyarimana et al. 2006).
38
Similarily, such transracial political ties were formed during the wars of indepen-
dence in Cuba (Helg 1995) and explain why the nation was imagined in a comparatively
inclusive way.
39
The Populist Party or the Readjuster coalition in Virginia that attempted to build
a transracial political network from scratch failed to break the white transclass
alliance established during the war and institutionalized within the Democratic Party.
On the rise and fall of the Readjuster movement, see Dailey (2000); on the defeat of
the Populist Party and the control of Democrats over the black vote, see Goodwyn
(1978, pp. 187200); Hicks ([1931] 1961, pp. 25154).
Ethnic Boundaries
997
speaking cantons when they mobilized a following to compete in the arena
of electoral politics. This explains Switzerlands exceptional history of
multiethnic nation building. Those networks were limited to a Creole-
mestizo elite in newly independent Mexico, and the vast majority of the
indigenous populations remained excluded from the nation-building pro-
ject up until the Mexican Revolution. The segregation of political net-
works along ethnoreligious lines in preindependent Iraq prevented the
rise of a popular Iraqi nationalism once the country was released from
the colonial leash. No independent civil-society organizations were al-
lowed under the Baaths ethnocratic dictatorship, and transethnic alli-
ances like those that had formed from the 1940s onward within the Com-
munist Party were destroyed. Once the American invasion led the Iraqi
state to collapse, political alliances rarely crossed the ethnoreligious di-
vides, and politics quickly became a matter of the balance of power be-
tween ethnoreligious blocks (Wimmer 2003).
STRUGGLING OVER BOUNDARIES: CONFLICT AND CONSENSUS
It follows from the previous analysis that different actors will pursue
different strategies of boundary making, depending on their position in
the hierarchies of power and the structure of their political networks. If
they want their preferred ethnic classication to be accepted by others
and the associated boundaries of inclusion and exclusion generally en-
forced and socially respected, they have to convince others of their view
of society. They thus have to enter a negotiation process with other actors
that may prefer other types of boundaries. We are now ready to consider
this interactional dynamics and analyze under which conditions they may
lead to a shared understanding of the location and meaning of ethnic
boundaries. But how is such consensus possible between actors who pur-
sue different strategies and are motivated by diverging interests?
The perhaps most prominent answer to this question is the one provided
by scholars working in the (neo-)Gramscian tradition.
40
They assume that
subordinates consent to the cultural models developed by elites, including
categories of ethnic or national belonging, thus stabilizing the underlying
system of political and economic domination. The precise ways in which
this consent is conceptualized diverge widely, however, not least because
of the many ambiguities in Gramscis own writings (Anderson 1976). Some
scholars emphasize the overwhelming denitional power of dominant ac-
tors. Subordinates passively receive and internalize hegemonic discourses,
40
The Marxist tradition of conceiving such agreements as a sign of false consciousness
(cf. Kasr 1979) has now been abandoned. For attempts at transposing Bourdieus
habitus theory into the domain of ethnicity, see Bentley (1987) and Wimmer (1994).
American Journal of Sociology
998
thus leaving no room for autonomous agency. This interpreation of he-
gemony makes it impossible to understand why subordinates sometimes
pursue counterhegemonic strategies such as boundary blurring, inversion,
or crossing.
41
More promising are other followers of Gramsci, notably
Roseberry (1994), Grandin (2000), and Mallon (1995), who underline the
informed, partial, and strategic nature of consent by subordinates and
show that elites are bound by the hegemonic accord as well, even if this
may at times go against their immediate self-interest. In this interpretation,
hegemony denotes a partial consensus between groups and individuals,
which reects a particular constellation of power and alliance.
This variant of neo-Gramscianism comes close to the theory of cultural
compromise that I have developed elsewhere (Wimmer 2002, chap. 2;
2005). According to this theory, a consensus between individuals and
groups endowed with different resources is more likely to emerge if their
interests at least partially overlap and strategies of classication can there-
fore concur on a shared view.
42
It is then possible to agree that a particular
ethnic boundary indeed represents the most important division of the
social world. Interest overlap does not necessarily imply that interests are
identical, however. Quite to the contrary, a consensus may result from
the exchange of different economic, political, and symbolic resources
between individuals occupying different social positions. Apartial overlap
of interests therefore reects a particular structure of inequality and po-
litical alliances in a social eld.
43
Let me illustrate the usefulness of this concept of cultural compromise
with some examples. Perhaps the most interesting is the spread of the
idea of the national community. What compromise does underlie this
consensus? The elite of a newly established nation-state promotes the
41
For such a Foucauldian interpretation of Gramsci, see Comaroff and Comaroff
(1991); Omi and Winant (1994, p. 66). For critiques, see Donham (2001); Merry (2003).
42
For experimental support for this assumption, see the sociopsychological research
tradition established by Thibaut (1968). For a similar approach in political philosophy,
see John Rawlss (1987) notion of an overlapping consensus.
43
I prefer this theory of cultural consensus over the neo-Gramscian framework for
three closely related reasons. First, it does not imply a dichotomous view according
to which a society is necessarily composed of two classes with opposite intereststhe
Marxian legacy in the Gramscian framework. Second, the language of cultural con-
sensus leaves no doubt that subordinate actors are capable of developing their own
classicatory practices. It thus avoids the implication that individuals act and think
against their true interests that is part of the conceptual baggage of hegemony, at
least in the dominant interpretation of Gramscian writings (cf. Gramsci 2001, p. 145).
Finally, the concept of hegemony was coined as an argument to support certain political
strategies among Russian revolutionaries, within the Comintern and later the Italian
Communist Party (Anderson 1976) and the New Left. The concept bears the marks
of this political history and does not travel particularly well to other constellations
outside of the orbit of these ideological preoccupations in the West.
Ethnic Boundaries
999
expansion of the boundary of the nation in order to give legitimacy to
increased state centralization and administrative control that the shift
from indirect to direct rule has brought about (Hechter 2000). On the
other hand, individuals of varying ethnic backgrounds may accept the
offer of assimilation and cross the boundary into the nation because this
allows them to claim equal treatment before the law, while access to justice
previously depended on ones social status and wealth. Assimilation into
the nation also increases the chance that their voice will be heard now
that the government claims to rule in the name of the people, while
beforehand political participation was limited by birth to certain clans,
families, or ethnosocial strata (Wimmer 2002). Thus, the nation-building
strategy pursued by state elites may be mirrored by subordinate strategies
of boundary crossing through individual assimilation or collective repos-
itioning. The exact nature of the nation-building process therefore depends
on the constellation of power and political alliances that sustain it, as the
previous discussion of the United States, Brazil, Mexico, Switzerland, and
Iraq has already suggested (cf. also Mallon [1995]).
Cultural consensus is also negotiated at lower levels of social organi-
zation, however, including in environments characterized by face-to-face
interactions and dense social networks.
44
In a previous work on indigenous
communities of Mexico and Guatemala, I have shown how the ongoing
negotiation between local elites and peasant farmers may result in agree-
ments on different types of ethnic boundaries, depending on the cong-
uration of power between actors and the exchange equilibrium it induces.
One example is the exchange of the political loyalty of peasant farmers
for collective goods provided by the local elite, most importantly the
defense of the communitys land holdings against the encroachment of
agricultural entrepreneurs or other peasant communities. For both sets
of actors, the idea of the local ethnic community as the prime locus of
political solidarity and as the spiritual center of the universe makes sense
and subsequently becomes institutionalized and routinized in many elds
of social life, including religion (Wimmer 1995). Similarily, Mallon (1995)
and Grandin (2000) have described the local and regional hegemonies
that bind together members of ethnic communities in the Sierra Norte de
Puebla and in Quetzaltenango, despite sharp differences in economic and
political power.
Such local consensus is not limited to village communities but may also
44
For a theory of ethnic identity that emphasizes this interactional, situational level
of the negotiation process, see Eder et al. (2002). A good empirical example of the
negotiation dynamics at the individual level is provided by Baileys analysis of how
an adolescent of Dominican origin situationally emphasizes his black, Hispanic, or
American identities (Bailey 2000).
American Journal of Sociology
1000
emerge in modern urban environments, as research on the boundary strug-
gles in immigrant neighborhoods in Switzerland shows. Here, the con-
sensus is much thinner than in the previous examples but still has powerful
consequences for the dynamics of ethnic boundary making. Despite dis-
agreement on who legitimately belongs to the morally, socially, and cul-
turally acceptable circle of persons, and what the appropriate standards
of judgment may be, men and women, old and young, established im-
migrants and autochthons agree that recently arrived refugees from the
former Yugoslavia bring trouble, indecency, and violence (Wimmer 2004).
This consensus on the categorical boundaries of belonging is reected in
the structures of social closure. Network data show that there are almost
no personal relationships between immigrants from former Yugoslavia
and established residents. Those excluded from the realms of the morally
decent and socially acceptable, however, do not share this view of the
social world. They pursue a strategy of blurring by emphasizing universal
moral qualities that make the division of the world into ethnonational
groups appear wrong and unjustiable (Karrer 2002, chap. 12).
As this last example illustrates, a consensus over boundaries may not
include the entire population. In the Swiss case, the boundary is one-
sided; that is, only the long-established neighborhood residents agree on
its relevance and legitimacy. We may refer to this as an asymmetrical
consensus. In other cases the consensus is partial. Most people would
agree on the topography of boundarieswho belongs on which side
but individuals on either side disagree strongly on the nature and the
political meaning of the ethnic divide. In Northern Ireland, there is little
dissent as to who is a Catholic and who is a Protestant, even if on the
local level there is room for negotiation and occasional boundary blurring
(Harris 1972; Burton 1978). Yet views on the signicance and political
implications of the religious divide diverge sharply. In the United States,
the one drop rule draws a sharp line between black and white and
is largely accepted by individuals on both sides, with only a small minority
advocating its blurring by adding a mixed race category. But disagree-
ment about the meaning and political implications of the boundary, as
over the legitimacy of afrmative action, are perhaps as pronounced today
as ever (Hochschild 2003). In such cases, Sandra Wallman wrote, the
boundary is not a conceptual fence over which neighbors may gossip or
quarrel. It becomes instead a Siegfried line across which any but the
crudest communications is impossible (Wallman 1978, p. 212).
Such struggle and contestation are characteristic of all cultural com-
promises, even when no open disagreements appear. According to the
theory of cultural consensus, every group and every individual constantly
tries to interpret the cultural compromise in ways that seem to justify
their own demands, to validate their own actions, and to represent their
Ethnic Boundaries
1001
own private vices as public benets. The notion of cultural compromise
therefore does not lead back to a functionalist view of society where
conicts and change vanish from sight. A cultural compromise merely
limits the horizon of possibilities within which individuals can argue in
their search for power and recognition. A cultural compromise may thus
be more or less encompassing. It may be limited to elites and counterelites,
or it may be shared by larger segments of the population. It may be more
or less stable, more or less reversible, more or less detailed and elaborated.
All these variations are, according to the analytical framework offered
here, dependent on the constellation of interests produced by institutional
patterns, hierarchies of power, and structures of political alliances.
BOUNDARY FEATURES
So far, I have offered a series of hypotheses to explain under which con-
ditions a widely shared consensus over ethnic boundaries will arise. We
are left with the task of explaining the varying nature of these boundaries,
or, more specically, their political salience, cultural signicance, social
closure, and historical stability. I will argue that these characteristics vary
according to the degree of power inequality as well as the reach of the
consensuswhether it is partial or encompassing, asymmetric or sym-
metric. Institutions and networksthe other main variables in the
modelinuence whether ethnic boundaries matter at all, and if they do,
whom they encompass and whom they exclude. They are less important
for understanding the properties of the boundary. In the following, I sug-
gest some preliminary hypotheses of how the degree of inequality and the
reach of consensus shape boundary features.
Closure, Salience, Differentiation
The more encompassing a compromisethat is, the more symmetric and
complete it isthe less politically salient a boundary will be. When the
location, meaning, and implications of a boundary are widely accepted,
it will be taken for granted on an everyday basis and impossible to chal-
lenge in the political arena. An encompassing consensus also allows cul-
tural differentiation to proceed smoothly since adding new cultural dia-
critics appears as a natural process when everyone agrees that the social
world is composed of ethnic groups with different cultures. On the other
hand, where there is no agreement on either the social location of ethnic
boundaries, let alone their consequences for the allocation of resources,
we expect ethnicity to be politically more salient. Thus, the question of
where boundaries lie and what the legitimate consequences of being an
American Journal of Sociology
1002
X rather than a Y should be may move to center stage in the political
drama.
Let us now consider how degrees of inequality affect the nature of
ethnic boundaries. Where power differentials between individuals of dif-
ferent ethnic backgrounds are high, degrees of social closure are also high,
as Cornell and Hartman have postulated (Cornell and Hartman 1998,
chap. 6). Those who have successfully set themselves apart from the rest
of the population as ethnic others and managed to monopolize economic,
political, or symbolic resources will try to police the ethnic boundary and
make assimilation and other strategies of boundary crossing difcult. The
more the maintenance of privilege depends on collective group member-
ship, such as in the Herrenvolk democracy of the postCivil War Amer-
ican South, the more ercely strategies of closure will be pursued. Con-
versely, where market forcessuch as a meritocratic system of elite
recruitment through expensive private schools and universitiesensure
status reproduction, tendencies of closure may weaken. Social closure and
high degrees of groupness in turn will lead, as we have learned from
Max Weber (1978, pp. 34148) and Pierre Bourdieu (1982), to cultural
differentiation because those who set themselves apart reinforce the
boundary by adding new cultural diacritics in order to show how cul-
turally different and inferior the subordinated groups are. This reinforces
the taken for grantedness of the boundary, which leads to further and
ongoing cultural differentiation, and so forth.
At the other end of the continuum, low degrees of inequality may make
strategies of boundary enforcement and policing less likely and in any
case less successful because the power to contest boundaries through in-
version, shifting, or blurring is more equally distributed across a popu-
lation. The results are low degrees of social closure and less cultural
differentiation. In many cases, the boundary will be contested, fuzzy,
varied, and soft enough to let observers agree, even those most inclined
to groupist thinking, that there is no clearly identiable ethnic group
of which a traditional anthropologist could write an ethnography.
Stability and Path Dependency
The relative stability of a boundarythe last of the four dimensions of
variation to be addressedderives from the three other characteristics
discussed above. Where boundaries are not politically salient, where de-
grees of closure and hierarchization are low, when cultural differentiation
has not produced an empirical landscape with clearly demarcated terri-
tories of cultural similarity, classicatory ambiguity and complexity will
be high and allow for more individual choice. Accordingly, boundaries
will change more easily. On the other end of the continuum, powerful
Ethnic Boundaries
1003
effects of path dependency develop (cf. Mahoney 2000). If ethnic bound-
aries correspond to cultural difference, they represent a plausible empirical
landscape against which any new classicatory discourse will have to
argue; if high degrees of social closure characterize an ethnic hierarchy,
a crosscutting, newly dened ethnic boundary needs to be advocated by
actors possessing considerable political power and legitimacy; if political
networks are aligned along an ethnic boundary, it will be difcult to
establish crosscutting alliances.
Such effects of path dependency are reinforced through the sociopsy-
chological process of identication. When members of an ethnic category
self-identify and are identied by others as belonging to a group with
little ambiguity, when they share easy-to-identify cultural repertoires of
thinking and acting, and when they are tied together by strong alliances
in day-to-day politics, we expect strong emotional attachment to such
ethnic categories to emerge (Brubaker 2004, pp. 4647.). Ethnic identity
will be thicker than in other contexts, and group members will be pre-
pared to incur high costs to defend the culture and honor of their com-
munity and the authenticity of its culture, thus stabilizing a boundary
even in situations of profound social change.
To put it differently, thick identities reduce the range of strategic
options that actors dispose ofthey will thus be more likely to choose
the scheme of interpretation and the script of action that corresponds to
the ethnic category in question, they will be more likely to dene their
interests in terms of those of the entire ethnic community, and they will
be more likely to respond to group pressure from their ethnic peers (cf.
Cornell 1996). Under these circumstances, identity may indeed assume
primacy over interests, as some authors in the identity school have
observed and unfortunately assumed to be a universal characteristic of
ethnicity per se.
Figure 1 summarizes the hypothesis regarding boundary features into
a three dimensional graph. It shows how the three boundary features
(with values plotted on the z-axis) are expected to vary depending on the
degree of power inequality (on the y-axis) as well as the reach of a con-
sensus (the x-axis). These hypotheses could be tested using various re-
search strategies, including comparative historical methods, multisite
eldwork, or cross-national studies. The data problems to be overcome
for a statistical test are quite formidable, however. So far ethnic boundaries
have rarely been treated as an outcome to be explained (but see Chai
2005) but rather as an independent variable that inuences explananda
such as economic growth or the propensity of civil wars. While there are
several indices that measure ethnic diversity either in demographic (Fea-
ron 2003) or political terms (Cederman, Girardin, and Wimmer 2006) or
that indicate the level of political mobilization of ethnic groups (Gurr
American Journal of Sociology
1004
Fig. 1.Boundary features as a function of the nature of consensus
1993), no data set exists that describes the nature of ethnic boundaries,
their degrees of closure, salience, or stability.
DYNAMICS OF CHANGE
In the preceding section, I outlined the major mechanisms that stabilize
a boundary by reducing the range of strategic options from which actors
choose. Certain ethnic boundaries therefore will be more resistant to stra-
tegic reinterpretation or blurring than others. Path dependency, however,
is not a deterministic concept. Under certain historical circumstances, a
path may be abandoned, and change becomes possible.
45
Following the
central tenets of the model outlined so far, three mechanisms of change
can now be discussed: rst, the eld characteristics (institutional frame-
works, power distributions, or political alliances) may change because
45
See the mechanisms of unlocking described by Castaldi and Dosi (2006) and Kathy
Thelens work on slow, cumulative change over longer periods of time (Thelen 2004).
Ethnic Boundaries
1005
new institutions, resources, or actors are introduced (exogenous shift).
Second, these eld characteristics may change endogenously as the in-
tended and unintended consequences of the strategies pursued by various
actors (endogenous shift). Third, new strategies diffuse into a social eld
and are adopted by certain actors (exogenous drift). These three sources
of change will be discussed subsequently.
1. Exogenous shift.Major political events such as imperial conquest
or nation-state formation transform the institutional structure, which in
turn provides incentives to pursue new strategies of boundary making
while letting go of old ones. Similar patterns of transformation can be
triggered by comparatively less dramatic institutional shifts. Dan Posner
shows how the democratization of Zambia resulted in a process of bound-
ary expansion (Posner 2005). In the postcivil rights era in the United
States, the shift to an ethnically based system for distributing state re-
sources has provided incentives for political actors and individuals to
organize social movements on the basis of ethnic claims.
46
The structure of power relations and political alliances can change
exogenously through various processes. New actors may enter a eld,
such as when international organizations become actively involved in the
ethnic politics of a country. The interventions of the European Union in
the candidate countries of Eastern Europe (Kymlicka 2007, chap. 6) or
the engagement of the UN and other international organizations for the
protection of indigenous rights in various Latin American countries are
examples here (see Conklin and Graham 1995; Warren 1998). Interna-
tional migration may also change the constellation of actors quite dra-
matically. These new actors also offer new opportunities for forming al-
liances and thus provide an impetus to redraw ethnic boundaries.
Exogenous processes may also shift the power base of actors, as the
following example illustrates. The resources that Latin American state
elites controlled dwindled when they were forced by nancial markets
and the International Monetary Fund to shift toward a policy of lean
government. Clientelist, corporatist forms of political incorporation broke
down and reduced the attractiveness of the nationalist, encompassing
classication. Political networks no longer extended from the centers of
power to the indigenous hinterland. Both factors together led to the rise
of ethnonationalist movements (Yashar 2005).
2. Endogenous shift.Boundaries may also change endogenously due
to the cumulative consequences of the strategies pursued by actors. If all
members of a particular ethnic category pursue a strategy of boundary
crossing into another group, and if members of this second group pursue
a strategy of boundary expansion and allow such assimilation, the rst
46
See Glazer and Moynihan 1975; a case study is provided by Padilla (1986).
American Journal of Sociology
1006
ethnic group will slowly disappear over timeas has happened to Mishars
and Teptiars in Russia (Gorenburg 1999) or as seems to be the case among
the Mayas of Belize (Gregory 1976) or the French speakers of Alberta
(Bouchard 1994).
A second endogenous mechanism is that small changes in the mix of
strategies pursued by individuals may cascade into dramatic shifts in the
structure of ethnic boundaries, as Kuran (1998) has shown, because they
may tip the dynamics of interaction and negotiation between actors
toward a new consensus. Such cascades may in turn empower (or di-
sempower) political movements who claim to represent the interests of
an ethnic group and who intentionally aim at redrawing the landscape
of ethnic divisions.
47
Third, if such movements are successful, they may not only manage to
shift the consensus over the location and meaning of boundaries in their
direction but also destabilize and denaturalize existing hierarchies of
power, institutional structures, and political alliances. These shifts in the
distribution of power, institutional order, and networks of alliance in turn
lead actors to pursue new strategies of boundary making and transform
their bargaining power in the process of negotiation and contestation,
leading to a further transformation of the system of ethnic boundaries
until a new equilibrium is reached.
The Mexican Revolution provides an apt illustration for this feedback
mechanism of endogenous change. The revolutionary wars mobilized
large sections of the indigenous population and provided the basis for
their integration into a new, pervasive network of clientelist relationships
managed and controlled by the emerging one-party regime. These political
networks supported, as I have shown elsewhere (Wimmer 1995, chap. 3;
cf. also Mallon 1995), a new concept of the Mexican nation. While Mex-
icans were imagined in the prerevolutionary period as consisting of criollo
elites who felt called to keep the racially inferior Indios in check, the
revolutionaries now conceived of the Mexican people as an amalgam of
Indian and Spanish cultures and peoples. As the model described in this
article predicts, the expansion of political networks was mirrored in an
expanded concept of the nation, resulting in a massive process of boundary
crossing by those indigenous villages most closely involved in the revo-
lutionary struggles and thus most integrated into the emerging clientelist
power apparatus. Accordingly, they quickly ceased to think of themselves
as anything other than Mexican (cf. the case study of Friedrich [1970]).
3. Exogenous drift.The system of ethnic boundaries may also change
because actors adopt new strategies that were not part of existing rep-
47
For other tipping models, see Laitin (1995b); for a descriptive approach, see Nagel
(1995).
Ethnic Boundaries
1007
ertoires. Innovative actors, who recombine separate schemes of thinking
and acting, may invent these new strategies or they may, more often than
not, be adopted from the outside. Examples are the global diffusion of
the strategy of normative inversion pursued by the U.S. American civil
rights movement, which has inspired not only red power (Nagel 1995)
and other ethnic minority movements (Takezawa 1995) in the United
States itself but also the political mobilization of Quebecois in Canada,
Catholics in Northern Ireland, postcolonial immigrants in the United
Kingdom, blacks in Brazil (Telles 2004), and so forth. Another strategy
of inversion is the discourse of indigenousness that has been adopted
by many ethnic minorities in Latin America and beyond (cf. Niezen 2003):
by Crimean Tatars, Roma, Afro-Latin Americans, Kurds, Palestinians,
Abkhas, Chechens, Tibetans, and Dalits (Kymlicka 2007, p. 285). Even
more important in world historical terms has been the global spread of
nationalismthe principle that ethnic and political boundaries should
coincideand corresponding strategies of ethnic boundary making from
the middle of the 19th century onward. This diffusion process has pro-
foundly changed the political outlook of the globe and transformed it from
a world of empires to one of nation-states (Wimmer and Min 2006)thus
globalizing an institution that provides strong incentives for the further
ethnicization of social and political life.
SYNOPSIS AND OUTLOOK
I have now discussed all the different elements of a multilevel process
theory of ethnic boundary making that promise to address the empirical
and analytical challenges faced by the eld of comparative ethnicity today.
The rst part of the model consists of three basic features of a social eld
that together determine which actors will pursue which strategy of ethnic
boundary making (see g. 2). First, the institutional order provides in-
centives to draw boundaries of a certain type. More specically, I have
discussed how the modern nation-state entices elites and subordinates
alike to distinguish, both in the political arena and in their private lives,
between ethnic us and them, rather than between man and women,
rich or poor, carpenters and college professors, and the like.
However, such institutional frameworks do not determine which level
of ethnic differentiation will be emphasizedwhether the Blue Hmong,
Hmong, Vietnamese, Asian, or American identities, to come back to an
example introduced ealier. The choice depends on the position in the
hierarchy of power. Actors will choose that level of ethnic distinction that
will best support their claims to prestige, moral worth, and political power.
Networks of political alliances, nally, will determine the precise location
American Journal of Sociology
1008
of the boundary, that is, who will be included in the group of the culturally
authentic, morally dignied, and politically entitled. These three eld
characteristics thus determine, in a probabilistic way to be sure, which
actors will pursue which strategies of ethnic boundary making.
In the next step, I looked at how these actors pursing different strategies
of boundary making interact with each other. Consensus over the social
topography and meaning of ethnic boundaries may or may not evolve
from these ongoing negotiations. I have maintained that consensus will
emerge where institutional structures, power differences, and networks
of alliance create a zone of mutually benecial exchange between actors,
a sphere of overlapping interests around which strategies of boundary
making can converge. My primary example for such consensus was nation
building, where the boundary expansion strategies of state elites and the
assimilation strategies of minority individuals converge. Other, more local
level examples referred to indigenous peasant communities in Mexico and
immigrant neighborhoods in Switzerland.
In the last step, I discussed how the nature of consensus shapes the
characteristics of boundaries: whether they remain largely categorical or
actually have consequences for the everyday web of social relationships
(degree of closure and groupness), how signicant the cultural differences
between individuals on opposite sides of the boundary will be (cultural
differentiation), and how far a boundary will be relevant for the forging
of political alliances (political salience). The model predicts, in a nutshell,
that the higher the degree of ethnic inequality and the more encompassing
the consensus between actors, the more closure and cultural differentiation
we expect to observe. The more inequality and the less consensus, on the
other hand, the more politically salient boundaries will be.
Finally, I identied four mechanisms that either stabilize or change a
system of ethnic boundaries. Highly salient, socially closed and culturally
marked ethnic groups will produce high degrees of identication among
its members and thus stabilize a boundary through path dependency
effects. Shifts in the structure of a social eldand thus in the strategies
pursued by individuals and the nature of the consensus they might reach
can be brought about by new institutions (such as through conquest,
revolution, or democratization), new actors (as through migration or the
emergence of new transnational actors), or new power resources. These
are treated as exogenous to the model, as is the invention and diffusion
of new strategies of ethnic boundary making, such as the global spread
of nationalism. The intended and unintended consequences of action rep-
resent an endogenous mechanism of change: successful ethnopolitical
movements intentionally transform eld structures through concerted po-
litical action while unintended consequences may cascade into shifts in
the location and meaning of ethnic boundaries.
Ethnic Boundaries
1009
Fig. 2.A processual model of the making and unmaking of ethnic boundaries
The theoretical framework introduced here departs from other ap-
proaches in several important ways. First, it does not follow the static
logic of standard typologies in comparative ethnicity. These distinguish
societies in which ethnicity coincides with social class from those where
it crosscuts class divisions (Horowitz 1971), or societies with high from
those with low degrees of ethnic institutional pluralism (van den Berghe
1967; Smith 1969), or societies where ethnic groups are segregated from
more integrated ones (Hunt and Walker 1979), or postnationalist Western
societies from the primordially ethnic rest (Heisler 1991), and so forth.
While these typologies conne themselves to outlining different forms and
functions of ethnicity, the model presented here explains these as the
outcome of a cycle of reproduction and transformation composed of var-
ious stabilizing and transformative feedbacks.
Second, a multilevel process theory does not offer a simple formula
relating dependent to independent variables as in mainstream social
science, for example, by predicting the degree of political salience of eth-
American Journal of Sociology
1010
nicity from levels of gross domestic product, democratization, or ethno-
linguistic heterogeneity (see the attempt by Chai [2005]). Rather, it is a
generative model where variables are dependent or independent de-
pending upon which phase in the cycle of reproduction and transformation
we focus. The model thus concurs with a series of recent approaches in
sociology (Abbott 1998; Emirbayer 1997), political science (Greif and Lai-
tin 2004; Thelen 2003), and economics (Acemoglu, Johnson, and Robinson
2004), which emphasize that in order to understand the logic of social life
we should focus on the processes that generate and transform its varying
forms (Cederman 2005).
Like other such models and similar to evolutionary models in biology
(cf. Lieberson and Lynn 2002), it is empirically void. That is, it needs
to be tailored to the relevant social and historical context in order to arrive
at a concrete prediction of which ethnic boundary we expect to result
from the dynamics of negotiation and contestation. The model thus does
not represent a lawlike universal operator but an analytical framework
for generating context-specic, local predictions. More specically, one
needs to rst ll in the historically grown character of existing bound-
aries (their salience, closure, cultural differentiation, etc.) before specifying
the institutional constraints, the distribution of power, and the structure
of alliances that prevail in a social eld at a particular point in history
to then understand the dynamics of negotiation and contestation that will
make a specic path of transformation more likely than others.
Finally, the model is more complex than others because it integrates
existing insights from both the macro and micro sociological traditions,
rather than pursuing only one avenue of research, such as rational choice
theories or, on the other end of the spectrum, the various world-system
approaches. It therefore covers several levels of analysisfrom the coun-
try level down to the micro processes of boundary contestation in everyday
life. It species the mechanisms that link these levels by showing how
macro social phenomena, such as institutional structures, the distribution
of power, and political alliances, inuence micro behaviorsuch as the
choice of particular strategies of boundary making. It also analyses how
the interplay of various strategies (the dynamics of consensus and conict)
in turn reects back on macro structures, that is, the nature of ethnic
boundaries that characterize a social eld. The model therefore offers a
full circle explanation, as specied by Coleman (1990), Bunge (1997),
and Hedstro m (2005), leading from macro to micro and back to the macro
level again.
That the model is of a processual nature, empirically unspecied and
of a multilevel nature, does not mean that it cannot be tested empirically.
It contains a range of comparative hypotheses that are meant to explain
in which societies and contexts ethnicity will be relevant, which actor will
Ethnic Boundaries
1011
pursue which type of ethnic boundary-making strategy, under which con-
ditions a more or less encompassing consensus over the location and
meaning of such boundaries will emerge, why such boundaries are more
or less politically salient, whether they imply more or less social closure,
how correlated they are with cultural differences, and so forth. Some of
these hypotheses have already been robustly tested. The idea that the
institution of the nation-state gives ethnic boundaries a new meaning, for
example, has been solidly conrmed by quantitative cross-national, com-
parative historical, and ethnographic research. Other hypotheses, most
importantly those concerning the boundary properties themselves, have
to wait for serious empirical testing in the future.
The aim of my model, then, is to situate these more specic empirical
propositions within an encompassing theoretical framework. Obviously
enough, this framework itself cannot be subjected to an empirical test
and thus cannot be falsied.
48
Its ambition is situated on a different
level: rst, to foster the conversation between the disjointed and segre-
gated elds of macro sociological, comparative historical approaches to
ethnicity, race, and nationalism, on the one hand, and the micro socio-
logical and ethnographic traditions, on the other hand. The goal is not
integration on a mere rhetorical level, but to identify as precisely as pos-
sible the mechanisms that link the various levels and domains on which
these school of research have traditionally focused. Second, the paradig-
matic framework offered here is meant to move the debate forward by
showing that the most prominent theories of ethnicityfrom primordi-
alism to constructivism, from instrumentalism to identity theoryare best
seen as descriptions of particular ethnic constellations, rather than as
general theories of ethnicity. The major challenge ahead that this paper
has identied and tried to address is to comparatively explain the emer-
gence, stabilization, and transformation of these various forms of ethnicity.
REFERENCES
Abbott, Andrew. 1998. Transcending General Linear Reality. Sociological Theory 6:
16986.
Acemoglu, Daron, Simon Johnson, and James Robinson. 2004. Institutions as the
Fundamental Cause of Long-Run Growth. Working Paper no. 10481. National
Bureau of Economic Research, Cambridge, Mass.
Alba, Richard D. 2005. Bright versus Blurred Boundaries: Second Generation
Assimilation and Exclusion in France, Germany, and the United States. Ethnic and
Racial Studies 28:2049.
Ali, Syed. 2002. Collective and Elective Ethnicity: Caste among Urban Muslims in
India. Sociological Forum 17:593620.
48
This understanding of the relation between empirical test and theoretical propositions
is based on Stegmu ller (1985).
American Journal of Sociology
1012
Almaguer, Toma s, and Moon-Kie Jung. 1999. The Enduring Ambiguitites of Race in
the United States. Pp. 21339 in Sociology for the Twenty-rst Century, edited by
Lila Abu-Lughod. Chicago: University of Chicago Press.
Anderson, Perry. 1976. The antinomies of Antonio Gramsci. New Left Review I/100:
565.
Anonymous. 1989. Ethnicity and Pseudo-Ethnicity in the Ciskei. In The Creation of
Tribalism in Southern Africa, edited by Leroy Vail. London: James Currey.
Anthias, Floya. 1992. Connecting Race and Ethnic Phenomena. Sociology 26:
42138.
Appadurai, Arjun. 1998. Dead Certainty: Ethnic Violence in an Era of Globalization.
Public Culture 10:22547.
Astuti, Rita. 1995. The Vezo are not a kind of people: Identity, Difference and
Ethnicity among a Fishing People of Western Madagascar. American Ethnologist
22:46482.
Bail, Christopher. In press. The Conguration of Symbolic Boundaries against
Immigrants in Europe. American Sociological Review.
Bailey, Benjamin. 2000. Language and Negotiation of Ethnic/Racial Identity among
Dominican Americans. Language in Society 29:55582.
Bailey, Frederick G. 1969. Strategems and Spoils: A Social Anthropology of Politics.
London: Basil Blackwell.
Balibar, Etienne. 1988. Racisme et Nationalisme. In Race, nation, classe: Les Identites
amigues, edited by Etienne Balibar and Immanuel Wallerstein. Paris: Editions la
De couverte.
Banton, Michael P. 1983. Racial and Ethnic Competition. Cambridge: Cambridge
University Press.
. 2003. Teaching Ethnic and Racial Studies. Ethnic and Racial Studies 26:
488502.
Barth, Fredrik, ed. 1969a. Ethnic Groups and Boundaries: The Social Organization
of Culture Difference. London: Allen & Unwin.
. 1969b. Introduction. In Ethnic Groups and Boundaries: The Social
Organization of Culture Difference, edited by Fredrik Barth. London: Allen&Unwin.
. 1994. Enduring and Emerging Issues in the Analysis of Ethnicity. In The
Anthropology of Ethnicity: Beyond Ethnic Groups and Boundaries, edited by Hans
Vermeulen and Cora Govers. Amsterdam: Het Spinhuis.
Bearman, Peter. 1993. Relations into Rhetorics: Local Elite Social Structure in Norfolk,
England, 15401640. New Brunswick, N.J.: Rutgers University Press.
Belanger, Sarah, and Maurice Pinard. 1991. Ethnic Movements and the Competition
Model: Some Missing Links. American Sociological Review 56:44657.
Bentley, Carter. 1987. Ethnicity and Practice. Comparative Studies in Society and
History 29:2455.
Berreman, Gerald D. 1972. Social Categories and Social Interaction in Urban India.
American Anthropologist 74:56786.
Betz, Hans-Georg. 1994. Radical Right-Wing Populism in Western Europe. New York:
St. Martins Press.
Boissevain, Jeremy, et al. 1990. Ethnic Entrepreneurs and Ethnic Strategies. In
Ethnic Entrepreneurs: Immigrant Business in Industrial Societies, edited by Roger
Waldinger, Howard Aldrich, Robin Ward, et al. Newbury Park, Calif.: Sage.
Bommes, Michael. 2004. U

ber die Aussichtslosigkeit ethnischer Konikte in

Deutschland. Pp. 15584 in Friedens- und Koniktforschung in Deutschland: Eine
Bestandesaufnahme, edited by Ulrich Eckern, Leonie Herwartz-Emden, and Rainer-
Olaf Schultze. Wiesbaden: VS-Verlag.
Bonacich, Edna. 1974. A Theory of Ethnic Antagonism: The Split Labor Market.
American Sociological Review 37:54759.
Ethnic Boundaries
1013
Bonilla-Silva, Eduardo. 1996. Rethinking Racism: Toward a Structural
Interpretation. American Sociological Review 62:46580.
. 1999. The Essential Social Fact of Race. American Sociological Review 64:
899906.
Bonnett, Alastair. 2006. The Americanization of Anti-racism? Global Power and
Hegemony in Ethnic Equity. Journal of Ethnic and Migration Studies 32:
10831103.
Bouchard, Michel. 1994. Ethnic Strategies: Integration, Accommodation, and
Militantism: The Case of the Francophones in Peace River. Canadian Ethnic
Studies 26:12440.
Bourdieu, Pierre. 1982. La Distinction: Critique social du jugement. Paris: Les E

ditions
de Minuit.
. 1991. Identity and Representation: Elements for a Critical Reection on the
Idea of Region. Pp. 22028 in Language and Symbolic Power, edited by Pierre
Bourdieu. Cambridge, Mass.: Harvard University Press.
Bourdieu, Pierre, and Loic Wacquant. 1999. On the Cunning of Imperialist Reasons.
Theory and Society 16:4158.
Branscombe, N. R., M. T. Schmitt, and R. D. Hervey. 1999. Perceiving Pervasive
Discrimination among African Americans: Implications for Group Identication and
Well-Being. Journal of Personality and Social Psychology 77:13549.
Brauka mper, Ulrich. 2005. Controversy over Local Tradition and National Ethiopian
Context: Case Study of the Hadiyya. Pp. 36376 in Afrikas Horn: Akten der Ersten
Internationalen Littmann-Konferenz 2. bis 5. Marz in Mu nchen, edited by Walter
Raunig and Steffen Wenig. Wiesbaden: Harrassowitz.
Brubaker, Rogers. 1992. Citizenship and Nationhood in France and Germany.
Cambridge, Mass.: Harvard University Press.
. 1996. Nationalism Reframed: Nationhood and the National Question in the
New Europe. Cambridge: Cambridge University Press.
. 2004. Ethnicity without Groups. Cambridge, Mass.: Harvard University Press.
Brubaker, Rogers, Margit Feinschmidt, Jon Fox, and Liana Grancea. 2007. Nationalist
Politics and Everyday Ethnicity in a Transylvanian Town. Princeton, N.J.: Princeton
University Press.
Bunge, Mario. 1997. Mechanism and Explanation. Philosophy of Social Sciences 27:
41065.
Burton, Frank. 1978. The Politics of Legitimacy: Struggles in a Belfast Community.
London: Routledge & Kegan Paul.
Campbell, Cameron, James Z. Lee, and Mark Elliott. 2002. Identity Construction and
Reconstruction: Naming and Manchu Ethnicity in Northeast China, 17491909.
Historical Methods 35:10115.
Castaldi, Carolina, and Giovanni Dosi. 2006. The Grip of History and the Scope for
Novelty: Some Results and Open Questions for Path Dependence in Economic
Processes. In Understanding Change: Models, Methodologies, and Metaphors, edited
by Andreas Wimmer and Reinhart Ko ssler. Houndsmill: Palgrave.
Cederman, Lars-Erik. 2005. Computational Models of Social Forms: Advancing
Generative Process Theory. American Journal of Sociology 110:86493.
Cederman, Lars-Erik, Luc Girardin, and Andreas Wimmer. 2006. Getting Ethnicity
Right: An Expert Survey on Power Distributions among Ethnic Groups. Paper
presented at the annual convention of the American Political Science Association,
Philadelphia, August.
Chai, Sun-Ki. 1996. A Theory of Ethnic Group Boundaries. Nations and Nationalism
2:281307.
. 2005. Predicting Ethnic Boundaries. European Sociological Review 21:
37591.
American Journal of Sociology
1014
Chua, Amy. 2004. World on Fire: How Exporting Free Market Democracy Breeds
Ethnic Hatred and Global Instability. New York: Anchor Books.
Cohen, Abner. 1974. The Lesson of Ethnicity. In Urban Ethnicity, edited by Abner
Cohen. London: Tavistock.
. 1981. Variables in Ethnicity. In Ethnic Change, edited by Charles Keyes.
Seattle: University of Washington Press.
Coleman, James. 1990. Foundations of Social Theory. Cambridge, Mass.: Harvard
University Press.
Comaroff, Jean, and John Comaroff. 1991. Of Revelation and Revolution, Vol. I:
Christianity, Colonialism, and Consciousness in South Africa. Chicago: University
of Chicago Press.
Comaroff, John, and Jean Comaroff. 1992. Ethnography and the Historical
Imagination. Boulder, Colo.: Westview Press.
Conklin, Beth, and Laura Graham. 1995. The Shifting Middle Ground: Amazonian
Indians and Eco-Politics. American Anthropologist 97:695710.
Corbitt, Duvon Clough. 1971. A Study of the Chinese in Cuba, 18471947. Wilmore,
Ky.: Asbury College.
Cornell, Stephen. 1996. The Variable Ties That Bind: Content and Circumstance in
Ethnic Processes. Ethnic and Racial Studies 19:26589.
Cornell, Stephen, and Douglas Hartman. 1998. Ethnicity and Race: Making Identities
in a Changing World. Thousand Oaks, Calif.: Pine Forge Press.
Dailey, Jane. 2000. Before Jim Crow: The Politics of Race in PostemancipationVirginia.
Chapel Hill: University of North Carolina Press.
Day, Dennis. 1998. Being Ascribed, and Resisting, Membership of an Ethnic Group.
Pp. 15170 in Identities in Talk, edited by Charles Antaki and Sue Widdicombe.
London: Sage.
Despres, Leo. 1975. Ethnicity and Resource Competition in Plural Societies. The
Hague: Mouton.
Donham, Donald. 2001. Thinking Temporally or Modernizing Anthropology.
American Anthropologist 103:13449.
Dovidio, John F., Peter Samuel Glick, and Laurie A. Rudman, eds. 2005. On the Nature
of Prejudice: Fifty Years after Allport. Oxford: Blackwell.
Eder, Klaus, Bernd Giesen, Oliver Schmidke, and Damian Tambini. 2002. Collective
Identities in Action: A Sociological Approach to Ethnicity. Aldershot: Ashgate.
Edgell, Penny, Joseph Gerteis, and Douglas Hartman. 2006. Atheists as Other: Moral
Boundaries and Cultural Membership in American Society. American Sociological
Review 71:21134.
Emirbayer, Mustafa. 1997. Manifesto for a Relational Sociology. American Journal
of Sociology 103:281317.
Erikson, Thomas H. 1993. Ethnicity and Nationalism: Anthropological Perspectives.
London: Pluto.
Espiritu, Yen Le. 1992. Asian American Panethnicity: Bridging Institutions and
Identities. Philadelphia: Temple University Press.
Esser, Hartmut. 2002. Soziologie: Spezielle Grundlagen, Band 6; Sinn und Kultur.
Frankfurt: Campus.
Feagin, Joe R., and Clairence Booher Feagin. 1993. Racial and Ethnic Relations.
Englewood Cliffs, N.J.: Prentice-Hall.
Fearon, James D. 2003. Ethnic and Cultural Diversity by Country. Journal of
Economic Growth 8:195222.
Friedrich, Paul. 1970. Agrarian Revolt in a Mexican Village. Englewood Cliffs, N.J.:
Prentice-Hall.
Fryer, Roland Gerhard, and M. Jackson. 2003. Categorical Cognition: APsychological
Model of Categories and Identication in Decision Making. Working Paper no.
9579. National Bureau of Economic Research, Cambridge, Mass.
Ethnic Boundaries
1015
Galaty, John G. 1982. Being Maasai; Being People of Cattle: Ethnic Shifters in
East Africa. American Ethnologist 9:120.
Gellner, David. 2001. How Should One Study Ethnicity and Nationalism?
Contributions to Nepalese Studies 28:110.
Gil-White, Francisco. 1999. How Thick Is Blood? The Plot Thickens: If Ethnic Actors
Are Primordialists, What Remains of the Circumstantialist/Primordialist
Controversy? Ethnic and Racial Studies 22:789820.
. 2001. Are Ethnic Groups Biological Species to the Human Brain? Current
Anthropology 42:51553.
Glazer, Nathan, and Daniel Patrick Moynihan. 1975. Introduction. Pp. 111 in
Ethnicity: Theory and Experience, edited by Nathan Glazer and Daniel Patrick
Moynihan. Cambridge, Mass.: Harvard University Press.
Gonzalez, Roberto, and Rupert Brown. 2003. Generalization of Positive Attitudes as
a Function of Subgroup and Superordinate Group Identications in Intergroup
Contact. European Journal of Social Psychology 33:195214.
Goodwyn, Lawrence. 1978. The Populist Moment: A Short History of the Agrarian
Revolt in America. Oxford: Oxford University Press.
Gordon, Milton M. 1964. Assimilation in American Life: The Role of Race, Religion
and National Origin. Oxford: Oxford University Press.
Gorenburg, Dmitry. 1999. Identity Change in Bashkorostan: Tatars into Bashkirs and
Back. Ethnic and Racial Studies 22:55480.
. 2000. Not with One Voice: An Explanation of Intragroup Variation in
Nationalist Sentiment. World Politics 53:11542.
Gramsci, Antonio. 2001. Selections from the Prison Notebooks. London: Electric Book
Company.
Grandin, Greg. 2000. The Blood of Guatemala: A History of Race and Nation. Durham,
N.C.: Duke University Press.
Gregory, James R. 1976. The Modication of an Interethnic Boundary in Belize.
American Ethnologist 3:683708.
Greif, Avner, and David Laitin. 2004. A Theory of Endogenous Institutional Change.
Working paper. Center on Democracy, Development, and the Rule of Law, Stanford,
Calif.
Guglielmo, Thomas. 2003. No Color Barriers: Italians, Race, and Power in the
United States. In Are Italians White? edited by Jennifer Guglielmo and Salvatore
Salerno. New York: Routledge.
Gurr, Ted R. 1993. Minorities at Risk: A Global View of Ethnopolitical Conict.
Washington, D.C.: U.S. Institute of Peace Press.
Habyarimana, James, Macartan Humphreys, Daniel Posner, and Jeremy Weinstein.
2006. Why Does Ethnic Diversity Undermine Public Goods Provision? Paper
presented at the annual meeting of the American Political Science Association,
Philadelphia, August.
Hale, Henry E. 2004. Explaining Ethnicity. Comparative Political Studies 37:45885.
Hannerz, Ulf. 1994. Sophiatown: The View from Afar. Journal of Southern African
Studies 20:18193.
Harris, Marvin. 1980. Patterns of Race in the Americas. Westport, Conn.: Greenwood
Press.
Harris, Rosemary. 1972. Prejudice and Tolerance in Ulster: A Study of Neighbours
and Strangers in a Border Community. Manchester: Manchester University Press.
Hechter, Michael. 2000. Containing Nationalism. Oxford: Oxford University Press.
. 2004. From Class to Culture. American Journal of Sociology 110:400445.
Hedstro m, Peter. 2005. Dissecting the Social: On the Principles of Analytical Sociology.
Cambridge: Cambridge University Press.
Heisler, Martin Ol. 1991. Ethnicity and Ethnic Relations in the Modern West. Pp.
American Journal of Sociology
1016
2152 in Conict and Peacemaking in Multiethnic Societies, edited by Joseph V.
Montville. New York: Lexington.
Helg, Aline. 1995. Our Rightful Share: The Afro-Cuban Struggle for Equality,
18861912. Chapel Hill: University of North Carolina Press.
Herder, Johann Gottfried. (1784) 1968. Ideen zur Philosophie der Geschichte der
Menschheit. Hildesheim: Olms.
Hicks, John D. (1931) 1961. The Populist Revolt: A History of the Farmers Alliance
and the Peoples Party. Lincoln: University of Nebraska Press.
Hirschfeld, Lawrence A. 1996. Race in the Making: Cognition, Culture and the Childs
Construction of Human Kinds. Cambridge, Mass.: MIT Press.
Hochschild, Jennifer. 2003. Who cares who killed Roger Ackroyd?: Narrowing the
Enduring Divisions of Race. Pp. 15569 in The Fractious Nation? Unity and
Division in Contemporary American Life, edited by Jonathan Rieder. Berkeley and
Los Angeles: University of California Press.
Hoetink, H. 1967. The Two Variants in Caribbean Race Relations: A Contribution to
the Sociology of Segmented Societies. Oxford: Oxford University Press.
Hollinger, David A. 2003. Amalgamation and Hypodescent: The Question of
Ethnoracial Mixture in the History of the United States. American Historical
Review 108 (5): 136390.
Horowitz, Donald. 1971. Three Dimensions of Ethnic Politics. World Politics 23:
23244.
. 1975. Ethnic Identity. Pp. 11140 in Ethnicity: Theory and Experience, edited
by Nathan Glazer and Daniel Patrick Moynihan. Cambridge, Mass.: Harvard
University Press.
. 1985. Ethnic Groups in Conict. Berkeley and Los Angeles: University of
California Press.
Hunt, Chester, and Lewis Walker. 1979. Ethnic Dynamics: Patterns of Intergroup
Relations in Various Societies. Holmes Beach, Fla.: Learning Publications.
Ignatiev, Noel. 1995. How the Irish became White. New York: Routledge.
Ingold, Tim. 1993. The Art of Translation in a Continuous World. In Beyond
Boundaries: Understanding, Translation and Anthropological Discourse, edited by
Gisli Pa lsson. London: Berg.
Ireland, Patrick. 1994. The Policy Challenge of Ethnic Diversity: Immigrant Politics
in France and Switzerland. Cambridge, Mass.: Harvard University Press.
Isaac, Harold R. 1967. Group Identity and Political Change: The Role of Color and
Physical Characteristics. Daedalus 96:35375.
Jacobson, Jessica. 1997. Religion and Ethnicity: Dual and Alternative Sources of
Identity among Young British Pakistanis. Ethnic and Racial Studies 20:23856.
Jenkins, Richard. 1994. Rethinking Ethnicity: Identity, Categorization and Power.
Ethnic and Racial Studies 17:197223.
. 1997. Rethinking Ethnicity: Arguments and Explorations. London: Sage.
Jime nez, Toma` s R. 2004. Negotiating Ethnic Boundaries: Multiethnic Americans and
Ethnic Identity in the United States. Ethnicities 4:7597.
Jordan, Winthrop D. 1968. White over Black: American Attitudes toward the Negro,
15501812. Chapel Hill: University of North Carolina Press.
Karrer, Dieter. 2002. Der Kampf um Integration: Zur Logik ethnischer Beziehungen
in einem sozial benachteiligten Stadtteil. Opladen: Westdeutscher.
Kasr, Nelson. 1979. Explaining Ethnic Political Participation. World Politics 31:
36588.
Keyes, Charles. 1976. Towards a New Formulation of the Concept of Ethnic Group.
Ethnicity 3:20213.
. 1979. Introduction. In Ethnic Adaptation and Identity: The Karen on the
Thai Frontier with Burma, edited by Charles Keyes. Philadelphia: Institute for the
Study of Human Issues.
Ethnic Boundaries
1017
. 1981. The Dialectics of Ethnic Change. Pp. 452 in Ethnic Change, edited
by Charles Keyes. Seattle: University of Washington Press.
Kibria, Nazli. 2002. Becoming Asian American: Second-Generation Chinese and
Korean American Identities. Baltimore: Johns Hopkins University Press.
Kivisto, Peter. 2003. The View from America: Comments on Banton. Ethnic and
Racial Studies 26:52836.
Koopmans, Ruud, Paul Statham, Marco Giugni, and Florence Passy. 2005. Contested
Citizenship: Immigration and Cultural Diversity in Europe. Minneapolis: University
of Minnesota Press.
Kroneberg, Clemens. 2005. Die Denition der Situation und die variable Rationalita t
der Akteure: Ein allgemeines Modell des Handelns. Zeitschrift fu r Soziologie 34:
34463.
Kuran, Timur. 1998. Ethnic Norms and Their Transformation through Reputational
Cascades. Journal of Legal Studies 27:62359.
Kymlicka, Will. 2007. Multicultural Odysseys: Navigating the New International
Politics of Diversity. Oxford: Oxford University Press.
Labelle, Micheline. 1987. Ideologie du couleur et classes sociales en Haiti. Montreal:
University of Montreal Press.
Laitin, David. 1995a. Marginality: A Microperspective. Rationality and Society 7:
3157.
. 1995b. National Revivals and Violence. Archives Europeennes de Sociologie
36:343.
Lamont, Miche` le. 1992. Money, Morals, Manners: The Culture of the French and
American Upper Class. Chicago: University of Chicago Press.
. 2000. The Dignity of Working Men: Morality and the Boundaries of Race,
Class, and Immigration. Cambridge, Mass.: Harvard University Press.
Lamont, Miche` le, and Christopher A. Bail. 2005. Sur les frontie` res de la
reconnaissance: Les Cate gories internes et externes de lidentite collective. Revue
Europeenne des Migrations Internationales 21:6190.
Lamont, Miche` le, and Crystal Fleming. 2005. Everyday Antiracism: Competence and
Religion in the Cultural Repertoire of the African American Elite. Du Bois Review
2:2943.
Lamont, Miche` le, and Vira g Molna r. 2001. How Blacks Use Consumption to Shape
Their Collective Identity: Evidence from African-America Marketing Specialists.
Journal of Consumer Culture 1:3145.
Lamont, Miche` le, Ann Morning, and Margarita Mooney. 2002. North African
Immigrants Respond to French Racism: Demonstrating Equivalence through
Universalism. Ethnic and Racial Studies 25:390414.
Lancester, Roger N. 1991. Skin Color, Race, and Racism in Nicaragua. Ethnology
34:33952.
Landa, Janet T. 1981. A Theory of the Ethnically Homogenous Middleman Group:
An Institutional Alternative to Contract Law. Journal of Legal Studies 10:34962.
Landale, Nancy F., and R. S. Oropesa. 2002. White, Black, or Puerto Rican? Racial
Self-Identication among Mainland and Island Puerto Ricans. Social Forces 81:
23154.
Leach, Edmund R. 1954. Political Systems of Highland Burma: A Study of Kachin
Social Structure. London: Athlone.
Levine, Nancy E. 1987. Caste, State, and Ethnic Boundaries in Nepal. Journal of
Asian Studies 46:7188.
Le Vine, Robert, and Donald Campbell. 1972. Ethnocentrism: Theories of Conict,
Ethnic Attitudes and Group Behaviour. New York: Wiley.
Lieberson, Stanley. 1980. A Piece of the Pie: Black and White Immigrants since 1880.
Berkeley and Los Angeles: University of California Press.
Lieberson, Stanley, and Freda B. Lynn. 2002. Barking up the Wrong Branch: Scientifc
American Journal of Sociology
1018
Alternatives to the Current Model of Sociological Science. Annual Review of
Anthropology 28:119.
Loewen, James W. 1971. The Mississippi Chinese: Between Black and White.
Cambridge, Mass.: Harvard University Press.
Loveman, Mara. 1997. Is Race Essential? American Sociological Review64:89198.
Loveman, Mara, and Jeronimo Muniz. In press. How Puerto Rico Became White:
Boundary Dynamics and Inter-Census Racial Classication. American Sociological
Review 72 (6).
Lustick, Ian. 2000. Agent-Based Modelling of Collective Identity: Testing
Constructivist Theory. Journal of Articial Societies and Social Simulation 3, no.
1 (January). http://www.soc.surrey.ac.uk/JASSS/3/1/1.html.
Mahoney, James. 2000. Path Dependency in Historical Sociology. Theory and Society
29:50748.
Mallon, Florencia. 1995. Peasant and Nation: The Making of Postcolonial Mexico and
Peru. Berkeley and Los Angeles: University of California Press.
Mamdani, Mahmood. 1996. Citizen and Subject: Contemporary Africa and the Legacy
of Late Colonialism. London: James Currey.
Mann, Michael. 2005. The Dark Side of Democracy: Explaining Ethnic Cleansing.
Cambridge: Cambridge University Press.
Manseld, Edward D., and Jack Snyder. 2005. Electing to Fight: Why Emerging
Democracies Go to War. Cambridge, Mass.: MIT Press.
Marx, Anthony W., ed. 1999. Making Race and Nation: A Comparision of the United
States, South Africa, and Brazil. Cambridge: Cambridge University Press.
Mayer, Philip. 1962. Migrancs and the Study of African in Towns. American
Anthropologist 64:57692.
McKay, James, and Frank Lewis. 1978. Ethnicity and Ethnic Group: A Conceptual
Analysis and Reformulation. Ethnic and Racial Studies 1:41227.
McKay, Jay. 1982. An Explanatory Synthesis of Primordial and Mobilizationist
Approaches to Ethnic Phenomena. Ethnic and Racial Studies 5:395420.
Merry, Sally Engle. 2003. Hegemony and Culture in Historical Anthropology: A
Review Essay on Jean and John Comaroffs Of Revelation and Revolution.
American Historical Rieview 108:46070.
Meyer, John, John Boli, George M. Thomas, and Francisco O. Ramirez. 1997. World
Society and the Nation-State. American Journal of Sociology 103:14481.
Mitchell, J. Clyde. 1974. Perceptions of Ethnicity and Ethnic Behaviour: An Empirical
Exploration. Pp. 135 in Urban Ethnicity, edited by Abner Cohen. London:
Tavistock.
Moerman, Michael. 1965. Ethnic Identication in a Complex Civilization: Who Are
the Lue? American Anthropologist 67:121530.
Mugny, Gabriel et al. 1991. Independence and Interdependence of Group Judgments:
Xenophobia and Minority Inuence. European Journal of Social Psychology 21:
21323.
Mummendey, A., T. Kessler, A. Klink, and R. Mielke. 1999. Strategies to Cope with
Negative Social Identity: Predictions by Social Identity Theory and Relative
Deprivation Theory. Journal of Personality and Social Psychology 76:22945.
Nagata, Judith. 1974. What Is a Malay? Situational Selection of Ethnic Identity in
a Plural Society. American Ethnologist 1:33150.
. 1981. In Defense of Ethnic Boundaries: The Changing Myths and Charters
of Malay Identity. In Ethnic Change, edited by Charles Keyes. Seattle: University
of Washington Press.
Nagel, Joane. 1995. American Indian Ethnic Revival: Politics and the Resurgence of
Identity. American Sociological Review 60:94765.
. 2003. Race, Ethnicity, and Sexuality: Intimate Intersections, Forbidden
Frontiers. Oxford: Oxford University Press.
Ethnic Boundaries
1019
Niezen, Ronald. 2003. The Origins of Indigenism: Human Rights and the Politics of
Identity. Berkeley and Los Angeles: University of California Press.
Okamoto, Dina G. 2003. Toward a Theory of Panethnicity: Explaining Asian
American Collective Action. American Sociological Review 68:81142.
. 2006. Institutional Panethnicity: Boundary Formation in Asian-American
Organizing. Social Forces 85:125.
Okamura, Jonathan. 1981. Situational Ethnicity. Ethnic and Racial Studies 4:45265.
Olzak, Susan. 1993. The Dynamics of Ethnic Competition and Conict. Stanford, Calif.:
Stanford University Press.
Olzak, Susan, and Joane Nagel, eds. 1986. Competitive Ethnic Relations. New York:
Academic Press.
Omi, Michael, and Howard Winant. 1994. Racial Formation in the United States:
From the 1960s to the 1990s. New York: Routledge & Kegan Paul.
OSullivan, Katherine. 1986. First World Nationalisms: Class and Ethnic Politics in
Northern Ireland and Quebec. Chicago: University of Chicago Press.
Padilla, Felix. 1986. Ladino Ethnicity in the City of Chicago. In Competitive Ethnic
Relations, edited by Susan Olzak and Joane Nagel. New York: Academic Press.
Passy, Florence. 1999. Supranational Political Opportunities as a Channel of
Globalization of Political Conicts: The Case of the Rights of Indigenous Peoples.
Pp. 14869 in Social Movements in a Globalizing World, edited by Donatella della
Porta, Hanspeter Kriesi, and Dieter Rucht. London: Macmillan.
Patterson, Orlando. 1975. Context and Choice in Ethnic Allegiance: A Theoretical
Framework and Caribbean Case Study. Pp. 30549 in Ethnicity: Theory and
Experience, edited by Nathan Glazer and Daniel Patrick Moynihan. Cambridge,
Mass.: Harvard University Press.
. 1997. The Ordeal of Integration: Progress and Resentment in Americas
Racial Crisis. Washington, D.C.: Civitas/Counterpoint.
. 2005. Four Modes of Ethno-Somatic Stratication: The Experience of Blacks
in Europe and the Americas. In Ethnicity, Social Mobility, and Public Policy:
Comparing the USA and UK, edited by Glenn C. Loury, Tariq Modood, and Steven
M. Teles. Cambridge: Cambridge University Press.
Pettigrew, Thomas F., ed. 1980. The Sociology of Race Relations: Reection and
Reform. New York: Free Press.
Posner, Daniel. 2005. Institutions and Ethnic Politics in Africa. New York: Cambridge
University Press.
Powell, Patricia. 1998. The Pagoda. San Diego: Harcourt Brace.
Preston, Julian. 2006. Texas Hospitals Reect Debate on Immigration. New York
Times, July 18.
Ramble, Charles. 1997. Tibetan Pride of Place: Or, Why Nepals Bhotiyas Are Not
an Ethnic Group. Pp. 32550 in Nationalism and Ethnicity in a Hindu Kingdom:
The Politics of Culture in Contemporary Nepal, edited by David Gellner, Joanna
Pfaff-Czarnecka, and John Whelpton. Amsterdam: Harwood Academic Publishers.
Rawls, John. 1987. The Idea of an Overlapping Consensus. Oxford Journal of Legal
Studies 7:125.
Ringer, Benjamin B. 1983. We the People and Others: Duality and Americas
Treatment of Its Racial Minorities. New York: Tavistock.
Roosens, Eugeen E. 1994. The Primordial Nature of Origins in Migrant Ethnicity.
In The Anthropology of Ethnicity: Beyond Ethnic Groups and Boundaries, edited
by Hans Vermeulen and Cora Govers. Amsterdam: Het Spinhuis.
Roseberry, William. 1994. Hegemony and the Language of Contention. Pp. 35566
in Everyday Forms of State Formation: Revolution and the Negotiation of Rule in
Modern Mexico, edited by Gilbert M. Joseph and Daniel Nugent. Durham, N.C.:
Duke University Press.
American Journal of Sociology
1020
Ross, Marc Howard. 2001. Psychocultural Interpretations and Dramas: Identity
Dynamics in Ethnic Conict. Political Psychology 22:15778.
Ruane, Joseph, and Jennifer Todd. 1996. The Dynamics of Conict in Northern Ireland:
Power, Conict and Emancipation. Cambridge: Cambridge University Press.
Russell, Andrew. 1997. Identity Management and Cultural Change: The Yakha of
East Nepal. In Nationalism and Ethnicity in a Hindu Kingdom: The Politics of
Culture in Contemporary Nepal, edited by David Gellner. Amsterdam: Harwood
Academic Publisher.
Saetersdal, Tore. 1999. Symbols of Cultural Identity: A Case Study from Tanzania.
African Archaeological Review 16:12135.
Saks, Karen Brodkin. 1994. How Did Jews become White Folks? Pp. 78102 in
Race, edited by Steven Gregory and Roger Sanjek. New Brunswick, N.J.: Rutgers
University Press.
Sanjek, Roger. 1981. Cognitive Maps of the Ethnic Domain in Urban Ghana:
Reections on Variability and Change. Pp. 30528 in Language, Culture, and
Cognition, edited by Ronald W. Casson. New York: Macmillan.
. 1996. The Enduring Inequalities of Race. Pp. 113 in Race, edited by Roger
Sanjek and James R. Gregory. New Brunswick, N.J.: Rutgers University Press.
Sansone, Livio. 2003. Blackness without Ethnicity: Constructing Race in Brazil.
Houndsmills, U.K.: Palgrave.
Scheff, Thomas. 1994. Emotions and Identity: A Theory of Ethnic Nationalism. In
Social Theory and the Politics of Identity, edited by Craig Calhoun. Oxford:
Blackwell.
Schermerhorn, Richard A. 1970. Comparative Ethnic Relations: A Framework for
Theory and Research. New York: Random House.
Schlee, Gu nther. 2006. Wie Feindbilder entstehen: Eine Theorie religioser und
ethnischer Konikte. Mu nchen: Beck.
Scott, George M. 1990. A Resynthesis of the Primordial and Circumstantial
Approaches to Ethnic Group Solidarity: Towards an Explanatory Model. Ethnic
and Racial Studies 13:14771.
Scott, James C. 1990. Domination and the Arts of Resistance. New Haven, Conn.:
Yale University Press.
Semyonov, Moshe, Rebecca Raijman, and Anastasia Gorodzeisky. 2006. The Rise of
Anti-Foreigner Sentiment in European Societies, 18882000. American Sociological
Review 71:42649.
Sharp, John, and Emile Boonzaier. 1994. Ethnic Identity as Performance: Lessons
from Namaqualand. Journal of Southern African Studies 20:40516.
Shibutani, Tamotsu, and Kian Kwan. 1965. Ethnic Stratication. New York:
MacMillan.
Smith, Anthony D. 1981. War and Ethnicity: The Role of Warfare in the Formation,
Self-Images and Cohesion of Ethnic Communities. Ethnic and Racial Studies 4:
37597.
. 1986. The Ethnic Origins of Nations. Oxford: Blackwell.
Smith, Michael G. 1969. Institutional and Political Conditions of Pluralism. Pp. 2766
in Pluralism in Africa, edited by Leo Kuper and Michael G. Smith. Berkeley and
Los Angeles: University of California Press.
Sollors, Werner. 1986. Beyond Ethnicity: Consent and Descent in American Culture.
New York: Oxford University Press.
, ed. 1991. The Invention of Ethnicity. Oxford: Oxford University Press.
Starr, Paul. 1978. Ethnic Categories and Identication in Lebanon. Urban Life 7:
11142.
Stegmu ller, Wolfgang. 1985. Probleme und Resultate der Wissenschaftstheorie und
analytischen Philosophie, Bd. 2, Teil E: Theoriendynamik, Normale Wissenschaft
Ethnic Boundaries
1021
und Wissenschaftliche Rebolutionen, Methodologie der Forschungsprogramme oder
epistemologische Anarchie? Berlin: Springer.
Stinchcombe, Arthur L. 2006. An Interactionist View of Boundaries and Borders.
Paper presented at Great Divides: Transgressing Boundaries, 101st Annual Meeting
of the American Sociological Association, Montreal, August.
Streicker, Joel. 1995. Policing Boundaries: Race, Class, and Gender in Cartagena,
Colombia. American Ethnologist 22:5474.
Takezawa, Yasuko. 1995. Breaking the Silence: Redress and Japanese American
Ethnicity. Ithaca, NY: Cornell University Press.
Telles, Edward E. 2004. Race in Another America: The Signicance of Skin Color in
Brazil. Princeton, N.J.: Princeton University Press.
Thelen, Kathleen. 2003. How Institutions Evolve: Insights from Comparative
Historical Analysis. In Comparative Historical Analysis in the Social Sciences,
edited by James Mahoney and Rueschemeyer. Cambridge: Cambridge University
Press.
. 2004. How Institutions Evolve: The Political Economy of Skills in Germany,
Britain, the United States, and Japan. Cambridge: Cambridge University Press.
Thibaut, John. 1968. The Development of Contractual Norms in Bargaining:
Replication and Variation. Journal of Conict Resolution 12:10212.
Thomson, Gerald E. 1997. Discrimination in Health Care. Annals of Internal
Medicine 126:91012.
Tilly, Charles. 2006. Identities, Boundaries, and Social Ties. Boulder, Colo.: Paradigm
Press.
van den Berghe, Pierre L. 1967. Race and Racism: A Comparative Perspective. New
York: Wiley.
. 1991. The Ethnic Phenomenon. New York: Elsevier.
. 1997. Rehabilitating Stereotypes. Ethnic and Racial Studies 20:116.
Verdery, Katherine. 1994. Ethnicity, Nationalism, and State-Making. In The
Anthropology of Ethnicity: Beyond Ethnic Groups and Boundaries, edited by Hans
Vermeulen and Cora Govers. Amsterdam: Het Spinhuis.
Wacquant, Loic. 1997. Towards an Analytic of Racial Domination. Political Power
and Social Theory 11: 22134.
Wade, Peter. 1995. Blackness and Race Mixture: The Dynamics of Racial Identity in
Colombia. Baltimore: Johns Hopkins University Press.
Wallman, Sandra. 1978. The Boundaries of Race: Processes of Ethnicity in England.
Man 13:200217.
. 1986. Ethnicity and the Boundary Making Process in Context. in Theories
of Race and Ethnic Relations, edited by John Rex. Cambridge: Cambridge
University Press.
Ward, Robin, and Richard Jenkins. 1984. Ethnic Communities in Business: Strategies
for Economic Survival. Cambridge: Cambridge University Press.
Warren, Kay B. 1998. Indigenous Movements and Their Critics. Princeton, N.J.:
Princeton University Press.
Wasserstrom, Robert. 1983. Class and Society in Central Chiapas. Berkeley and Los
Angeles: University of California Press.
Waters, Mary C. 1990. Ethnic Options: Choosing Identities in America. Berkeley and
Los Angeles: University of California Press.
. 1999. Black Identities: West Indian Immigrant Dreams and American
Realities. Cambridge, Mass.: Harvard University Press.
Weber, Max. (1922) 1985. Wirtschaft und Gesellschaft: Grundriss der verstehenden
Soziologie. Tu bingen: Mohr.
. 1978. Economy and Society: An Outline of Interpretive Sociology, edited by
Guenther Roth and Claus Wittich. Berkeley and Los Angeles: University of
California Press.
American Journal of Sociology
1022
Williams, Brackette F. 1989. A Class Act: Anthropology and the Race to Nation across
Ethnic Terrain. Annual Review of Anthropology 18:40144.
Wimmer, Andreas. 1993. Ethnischer Radikalismus als Gegennationalismus:
Indianische Bewegungen im sechsten Jahrhundert nach Kolumbus. In 500 Jahre
danach: Zur heutigen Lage der indigenen Volker beider Amerika, edited by Peter
Gerber. Chur: Ru egger.
. 1994. Die ethnische Dynamik in Mexiko und Guatemala. In Ethnische
Dynamik in der auereuropaischen Welt, edited by Hans-Peter Mu ller. Zu rich:
Argonaut-Verlag.
. 1995. Transformationen: Sozialer Wandel im indianischen Mittelamerika.
Berlin: Reimer.
. 2000. Racism in Nationalized States: A Framework for Comparative
Research. Pp. 4772 in Comparative Perspectives on Racism, edited by Ter Wal,
Jessika Verkuyten, and Maykel Verkuyten. Aldershot: Ashgate.
. 2002. Nationalist Exclusion and Ethnic Conicts: Shadows of Modernity.
Cambridge: Cambridge University Press.
. 2003. Democracy and Ethno-Religious Conict in Iraq. Survival: The
International Institute for Strategic Studies Quarterly 45:11134.
. 2004. Does Ethnicity Matter? Everyday Group Formation in Three Swiss
Immigrant Neighborhoods. Ethnic and Racial Studies 27:136.
. 2005. Kultur als Prozess: Zur Dynamik des Aushandelns von Bedeutungen.
Wiesbaden: Verlag fu r Sozialwissenschaften.
. 2007. Elementary Strategies of Making Ethnic Boundaries. Manuscript.
University of California, Los Angeles, Department of Sociology.
. In press. How (Not) to Think about Ethnicity in Immigrant Societies. Special
issue no. 47: Immigrationssoziologie. Kolner Zeitschrift fu r Soziologie und
Sozialpsychologie, vol. 59.
Wimmer, Andreas, and Brian Min. 2006. From Empire to Nation-State: Explaining
Wars in the Modern World, 18162001. American Sociological Review 71:86797.
Winant, Howard. 2000. Race and Race Theory. Annual Review of Sociology 26:
16985.
Wintrobe, Ronald. 1995. Some Economics of Ethnic Capital Formation and Conict.
In Nationalism and Rationality, edited by Albert Breton. Cambridge: Cambridge
University Press.
Woldemikael, Tekle Mariam. 1989. Becoming Black American: Haitians and American
Institutions in Evanston, Illinois. New York: AMS Press.
Wolf, Eric. 1957. Closed Corporate Peasant Communities in Mesoamerica and Central
Java. Southwestern Journal of Anthropology 13:118.
Yashar, Deborah. 2005. Contesting Citizenship in Latin America: The Rise of
Indigenous Movements and the Postliberal Challenge. Cambridge: Cambridge
University Press.
Young, Crawford. 1976. The Politics of Cultural Pluralism. Madison: University of
Wisconsin Press.
Zelizer, Viviana A., and Charles Tilly. In press. Relations and Categories. In The
Psychology of Learning and Motivation, vol. 47, edited by Arthur Markman and
Brian Ross. San Diego: Elsevier.
Zolberg, Astride, and Long Litt Woon. 1999. Why Islam Is like Spanish: Cultural
Incorporation in Europe and the United States. Politics and Society 27:538.