8


A Theory of Political Identities

David Laitin


On April 25, 1994, the Estonian state-run bus service dropped me off, after a four-and-a-half hour
trip from the capital city of Tallinn, in the historic Hanseatic city of Narva. This was my fifth
journey to this city of 80,000 people, 95 percent of whom speak Russian as their primary language.
The city is only a stone's throw across the Narva River, from Russia. For centuries, the river
represented a civilizational divide. A leading Estonian sociologist, Marika Kirch, puts it in stark
terms. “If one supposes hesitatingly,” she writes,


that the civilizational border between Estonia and Russia is anachronistic or negligible, one need
only stand on the bridge over the Narva river . . . and witness carefully the “overt civilizational
confrontation” of two cultures: on the Estonian side there is an historic fortress built by the
Swedes, Danes and Germans in accordance with the cultural traditions of Western Europe; on the
other [in Ivangorod], a primeval fortress as an exponent of Slavic-Orthodox cultural traditions.^
1^


But the Second World War altered this boundary. The Soviets shelled Narva heavily when they
occupied Estonia, and either killed, captured, or drove off virtually the entire population. After
the war the town was rebuilt, largely by demobilized Russian soldiers, most of whom believed, and
still believe, that they helped save the Estonians from fascism and merit honour for rebuilding the
city, which their own army had destroyed. From the early 1950s up until 1991, Narva and Ivangorod
formed a single Russian-speaking metropolitan area.

On my first visit to Narva, in July 1992, I met with a parliamentarian, Pavel Grigor’ev, who had
been quoted in the New York Times as a Russian activist standing up against the nationalizing
tendencies of the new Estonian state. After the interview, he invited me to his home in the seaside
village of Narva Jõesuu, just a few kilometers north of Narva, and the home of many Soviet writers
and intellectuals who had been rewarded by the state with private homes in peaceful communities.
The kilometers-long beach at Narva Jõesuu made it quite attractive to many Russians, especially to
well-connected Leningradians seeking restful summer dachas.

Grigor’ev lived there, not as a reward for artistic achievement but because he was a long-time
machinist in the nearby October fishing kolkhoz. He was born in Kingisepp (near Narva, but in
Leningrad region) in 1941 and started his career working in Murmansk, an industrial fishing city,
which he and his wife found depressingly dark. He found an opening at the October kolkhoz in 1966
and lived in a one-room flat with wife and son on the Ivangorod bank of the river. Two years later
they moved into a lovely three-room flat in a three-story building constructed for the kolkhoz on
the Estonian side of the border, where they still live today. Articulate, open-minded, and
uncannily able to get people of a variety of persuasions to believe that they share his vision,
Grigor’ev moved up the political ladder and was eventually elected to the Supreme Soviet of the
Estonian Republic.

In my second trip to Narva, in December 1992, I took Grigor’ev and his wife out to dinner, and he
graciously responded by helping me with my research. I returned for a two-week visit in April 1993,
and this time he invited me to live in his home while I tried (desperately, but with little to show
for it) to develop facility in Russian, a language I had been studying for a few years but without
great success. I got to know better his wife, Liuba, who pretended, as they used to say, to work in
a sanatorium while the government pretended to pay her. On my first trip to Narva, I had brought
packaged herbs and spices. She was appreciative and (being a superb cook) used them with great
ingenuity. From then on, I was always a welcome guest at their dinner table, and I eventually
taught her to cook on a wok, and in the Mexican style as well. I met their oldest child, Andreu,
who had been decommissioned from the Soviet army, much to his chagrin, when his country (the USSR)
disappeared. I also met their daughter Natasha, who had an excellent ear for languages, had
excelled since her early grades (so her elementary school teacher in Estonian, who herself was a
native speaker of Estonian, told me the following year), and was training at the Tallinn
Pedagogical Institute to become a teacher of Estonian in Russian schools. Finally, I met their
youngest, Roma, then in the sixth grade, whose goal then was to become a hockey player of note.

In the spring of 1993, Grigor’ev was a lame-duck parliamentarian. Because he was not an Estonian
citizen, he could not run for reelection to the Riigikogu, the Estonian parliament. But the
Estonian political establishment considered him a moderate, a man with whom they could do business.
A little-known (and rarely publicized) provision of the Estonian Constitution allowed the prime
minister to recommend citizenship to residents of Estonia for “service to the state.” Grigor’ev was
told that if he applied for citizenship, even though he understood hardly a word of Estonian, he
would be granted it. Indeed, he received citizenship, and Roma, who was under twelve at the time,
received it as well. Roma was studying Estonian at school for a few hours each week, and his
closest friend and next-door neighbor was Estonian, but he could hardly utter a grammatical
sentence in that language himself. His English, though primitive, was far better. And so Pavel and
Roma were Estonian citizens, but Liuba, Andreu, and Natasha remained “Soviet” citizens without a
country.

During my visit in April 1993, I told the family that I planned to do a year of field research in
Narva beginning the following fall. Pavel and Liuba worried about my safety and sanity and urged me
to live with them, in Andreu’s room, since Andreu was about to get married and move into his own
place. Despite many rules of anthropological thumb against becoming associated with any one side of
a community conflict, I consented, and moved in that September. The next month Grigor’ev ran
successfully for chairman of the Narva Jõesuu city council, something he could not have done had he
not become an Estonian citizen. Russian-speaking candidates could run successful campaigns in these
local elections, because all adult residents were eligible voters in these, but not in general
elections, where citizenship was necessary in order to vote. Since citizenship for service to the
state was granted only to the moderates among the Russian politicians, the Estonian prime
ministers, who had discretion on this matter, could assure their constituents that only a moderate
group of Russian-speaking politicians would get office in the initial years of the new republic.
(To be sure, the Estonians were not unified on this issue. Prime Minister Tiit Vähi gave the gift
of citizenship to people with whom he had good bargaining relations, much to the chagrin of the
opposition. But when the opposition leader Mart Laar came to power, he sought to cultivate “his”
Russians in the same manner, and Vähi, then out of office, criticized Laar for abusing the
practice. Nonetheless, neither Vähi nor Laar gave citizenship to potential fifth columnists.)

During the fall of 1993, I enrolled in a class in Estonian for Russian-speakers at the Narva
Language Center and began writing the family biographies, one of which is the basis for this
vignette. I returned home to Chicago for a few months in the winter and returned to Narva, as I
began this story, in April 1994.

Liuba was waiting for me at the bus stop. She was extremely agitated, and as we waited for nearly
two hours for Pavel’s official car (a late-model Lincoln Town Car, with a chauffeur ) to take us to
Narva Jõesuu, I learned that it wasn’t the extraordinary inflation that was bothering her the most.
Indeed, the bus fare to Narva Jõesuu was 90 cents (100 cents to the kroon; about 13 kroons to the
dollar) when I left in December and had increased to 3 kroons that April. Rent for their apartment
had jumped to 650 kroons a month from 330 a few months earlier. This inflation was harsh but not
threatening. Her principal concerns were deportation and the possible forced dispersal of her
family.

Her strategy for avoiding deportation was to become a full-time student of Estonian. She had just
paid 650 kroons for the course designed to provide citizenship-level competence in Estonian. Her
rush to learn Estonian was induced by a new requirement that noncitizens had to register for
temporary residency permits; a permanent residency permit might or might not be issued later.
Noncitizens who failed to register by July would be deported. Liuba was worried. Rumors abounded
that political criteria would be applied in the granting of permanent residency status. Unwanted
Russians would be deported. In Tallinn the minister of nationalities, Peter Olesk, was already
being called, in a bitter pun (vyselennia for natseleniia), the minister for deportation. Liuba
accepted the local notion (among Russians) that there was a small window of opportunity: if she
passed the citizenship language test by June, she would not need to register. Her application for
citizenship would serve as her propusk (permit), giving her all the rights of a permanent resident.
The situation for Russians, she emphasized, was very uncertain. The right of foreign travel, among
other things, was in the arbitrary hands of the Estonian authorities. Up till recently you could
get an empty Soviet foreign passport and have it issued by Estonian authorities for foreign travel.
But the Estonian government had run out of such passports and had no access to others, or at least
that was what Russians applying for such passports in Narva were being told. So Russian residents
of Estonia who wanted to travel abroad could not get papers without special intervention. Even
members of sports teams and other Russians with institutional ties to Estonian organizations were
having problems. To get the right to foreign travel, Estonia’s Russians theoretically could get
“Russian citizen” stamped in their Soviet internal passports-and indeed many did so – but that
strategy (despite official denials) was felt to prejudice future applications for Estonian
citizenship.

Liuba was near catatonic. Estonian is a Finno-Ugric language, with virtually no cognates in
Russian. She found its structure impenetrable. She was convinced that there was no way she could
reach the citizenship level in the time remaining. Indeed, the Narva Language Center has calculated
that to reach level B, enabling one to qualify for a job requiring only a low level of language
skill, 70 hours of instruction is necessary. Another 50 hours is required to reach level C,
qualifying one for most clerk and administrative jobs. Level D, for professional use, requires 120
additional hours. The citizenship level requires 60 more hours in the classroom. Each of these
courses involves considerable homework and additional computer-assisted grammatical drills (an
extra fee is charged for computer time). Liuba, before her course began, could get through a basic
greeting in Estonian, and she had named her cat Lumi, Estonian for “snow,” in a demonstration of
cultural accommodation to the country in which she lived. After that, as far as Estonian was
concerned, she was mute.

Liuba might have responded to the new language regime earlier. Indeed, a language law passed in
January 1989 sought to institute real bilingualism in the society and induce Russians to learn
Estonian. But the law had little bite, since under Soviet hegemony, the Estonians had lacked the
authority to enforce it. But after independence, the project of “naturalizing” some 500,000 mostly
Russian noncitizens (a large number of whom had been born in Estonia or had lived there for
decades) provided an opportunity to impose sanctions on Russians who had not learned Estonian.

As the precise terms of the naturalization process were laid down, the language test (as Liuba
herself experienced) became the most challenging of the hoops to be jumped through. In a new
citizenship law passed in January 1995, the Estonians added another civics examination to the
naturalization procedure for Soviet-era immigrants. Many noncitizens viewed this new requirement as
an attempt to slow the naturalization process, and their belief was reinforced when the government
delayed more than three months in issuing specific information on the new examination.

Moreover, the Estonians interpreted the noncitizens’ status to mean that all of  their Soviet-era
residency documents would have to be reprocessed. The Aliens Law passed in July 1993 caused a major
political crisis. Its administration was a bureaucratic nightmare. Many observers interpreted these
regulations and the slow processing of applications for citizenship as clear signals that the
titulars wanted Russians to leave rather than integrate. Although many Estonian nationalists openly
voiced that wish, the reality was that the great majority would remain in Estonia, and they would
have to come to terms with Estonian authority.

So Liuba studied Estonian in all her free moments, and even practiced pronunciation from her word
lists with her Estonian neighbor, who came over regularly to gossip (in Russian). Liuba all but
abandoned TV for study, though occasionally she sat with her word lists to watch the American soap
opera Santa Barbara on Ostankino, the most popular Russian TV station. Even young Roma’s interest
was sparked by his mother’s obsession, and he began to study with her, even engaging in Estonian
small talk when they sat down together in the kitchen after he returned home from school. Once I
showed off to the family that I had the latest Estonian grammatical exercise program on my laptop
computer. I conjugated a verb perfectly, and the program rewarded me with an electronic version of
“Merrily We Roll Along.” Liuba often borrowed my computer late in the evenings when I was through
working, and I’d be awakened in the wee hours of the morning to the tinkling strains of “Merrily .
. .” Liuba was possessed.

She was strategic as well. On my first morning back, she phoned an old buddy of  Pavel’s, whom I
had met when he was on border guard duty the previous fall, to see if he could help her get her
citizenship application forms without waiting in a long line in an office that was open only a few
hours a week. All this strategic activity took place in an atmosphere of considerable uncertainty.
The highly nationalistic government then in power – the Isamaa or “Fatherland” coalition – was
ambivalent about Russians’ learning the Estonian language. On the one hand, Isamaa leaders insisted
that race or civilization had nothing to do with citizenship. It was only natural for a country to
require immigrants to learn the language of the country before they could receive citizenship.
Indeed, as Pavel sardonically observed, “race” could not possibly be a criterion for the Estonian
national chauvinists. After all, he pointed out, his roots are in Ingerland, whose inhabitants are
closer to the indigenous Ugrics than most Estonians, who have much “foreign” (German, Swedish,
Danish) blood. On the other hand, many leading Estonian nationalists claimed that the percentage of
Russians in Estonia was so high that they would never assimilate. If two-thirds of them were to
“return” to Russia (though many of them had been born in Estonia), these nationalists claimed,
Estonian culture could survive. This ambivalence led to a program with contradictions. Language
centers were indeed created, but barely funded, and therefore tuition was high. (The Swedish and
U.S. embassies took the initiative to provide funds for these centers, which the Estonian
government was not pleased to accept but could not decline.)

Russian speakers who passed the language exam complained that their citizenship applications were
languishing (or disappearing) in the bureaucratic mill. One of my informants told me that three
times over the course of a year he had been told that crucial official documents were “missing”
from his file, although he knew he had submitted them. Each time he had to reply for citizenship.
Natasha, Pavel and Liuba’s daughter, faced the same Kafkesque nightmare. Pavel told me that she
passed the Estonian language exam without difficulty in Tallinn, where standards are higher than in
Narva, where informal dispensations are made, given the extreme difficulty of learning Estonian in
a city where hardly anyone speaks it, and where Estonian radio and TV reception is poor. Yet two
years later Natasha was still waiting for her citizenship papers. The Estonian government, he
reckoned, worked with an unannounced citizenship quota that violated international human rights
agreements.

Pavel is a realist. He told me that if he had the choice, he would vote for the secession of
northeast Estonia and for reintegrating it with Russia. But this would not be possible without
bloodshed, and he emphasized to me, and to conferees at a diplomatic panel in Tallinn, that a bad
peace is far better than a good war. Moreover, his desire to rejoin Russia would not involve
emigration, even though his mother lives in the Russian town of Kingisepp. His dacha keeps him in
Estonia. In a former swamp just south of the Baltic coast, thousands of Russian families have built
lovely summer homes from materials appropriated from the state, and in their gardens grow fruits
and vegetables that last most families for the entire year. Many men, including Pavel and Andreu,
spent all their free hours for years constructing those dachas. In fact, Pavel secured a special
visa for his mother, and she spends the entire summer managing the garden at the dacha. (I once
asked her if she spoke any Estonian, and she looked at me with astonishment, as if I had asked her
whether she had even been to Disneyland.) Property keeps Pavel a loyal Estonian.

Pavel’s realism is combined with a complete lack of prejudice. I have heard him make insinuations
about colleagues and public officials, but I have never heard him make an ethnic slur. Pavel used
to spend hours talking politics with his next-door neighbor, an Estonian, who died of a heart
attack a year before I first came to Narva. Pavel’s electoral “ticket” for the Narva Jõesuu city
council, when he successfully ran for chairman in October 1994, was ethnically mixed, without
tokens from either the Russian or Estonian communities. He had even named his youngest son after
the Estonian who founded his kolkhoz. A true Soviet man, he really did not see the world in ethnic
terms. In this way, he was quite open-minded about Estonian sovereignty.

On language, however, Pavel was less open-minded, although still more in tune with the times than
his mother. Pavel’s monolingualism was as natural to him as it is to nearly all third-generation
Americans. He had traveled as far east as Samarkand, as far south as Sukumi, as far north as
Murmansk, and could communicate with anyone in Russian. For Pavel it was as if the whole world
spoke Russian; what need did he have for a second language? This attitude infuriated Mart Rannut,
first head of Estonia’s State Language Office, who once complained that Estonia might cave in to
international pressure and extend citizenship to Russian-speakers who had only reached, in Rannut’s
phrase “dog level” proficiency in Estonian – in other words, the ability to respond to a small set
of commands. When I repeated this to Grigor’ev, he laughed uproariously. That much he knew he could
do. Pavel would often greet others with tere, the standard Estonian greeting, to show his socialist
internationalism; but that was just about the limit of his proficiency. Seven time zones, he was
fond of pointing out, and all you need is Russian. Sometimes he would stare, eyes glazed, at an
Estonian text, making believe he was giving it a gloss, but he rarely caught a word. To be sure,
the cataclysm of the Soviet collapse in 1991 had opened his eyes, and he was adjusting quite
rapidly to the new order. He took Roma out of the local school and enrolled him in Narva, where the
principal had given the school an international flavor, specializing in foreign languages. The
school, because it offered foreign language training, was quickly becoming more popular among
parents than the prestigious baccalaureate school. In his school, Roma was getting adequate English
and a touch of Estonian, but far more than he was receiving in Narva Jõesuu. Pavel, like his wife,
believes that Russian-speakers in Estonia should speak Estonian, but unlike Liuba, he is not
learning it himself. Still, he is making sure his children are equipped linguistically for the new
reality.

Andreu, like his father, had no yearning to learn Estonian. He remained more Soviet than his
parents and often accused Gorbachev of having destroyed the Soviet Union on behalf of U.S.
intelligence. In his long period of semi-unemployment after his demobilization from the army
-during which he had small jobs at the local TV station and elsewhere – he attained a level B
proficiency in Estonian, which qualified him for white-collar jobs. But when he landed a job as a
senior technician in a new insurance firm and had to reach level C, he copied the grammar program
from my hard drive and installed it on his own computer in the insurance office. He too began to
study assiduously.

Natasha’s proficiency in Estonian was legendary. Her primary school teacher in Estonian, who became
an official at the Narva Language Center in the 1990s, often used Natasha as the example of the
possibility for linguistic assimilation. In 1994, Natasha was living in Tallinn, studying education
in order to teach Estonian in a Russian-speaking school. She keeps her hair short – and I think
bleached – to make herself physically indistinguishable from her Estonian counterparts. This was
the situation as I left the field in June 1994.

The news of the Grigor’ev family in the Christmas letter that Liuba sent me in December 1994 was
joyous. Liuba had passed her language examination at level D and at the citizenship level as well,
and now had only a year’s wait for her official papers. Natasha had received her citizenship
passport from Estonia. Andreu would one day catch up.

It would be a mistake to see in this story the assimilation of the Grigor’ev  family from “Soviet”
to Estonian. They see themselves as Russians who have from practical necessity added a few Estonian
cultural practices to their own Russian repertoire. Yet cultural assimilation is like religious
conversion, and as the literature on religious conversion makes clear, what one generation
considers simple pragmatism the next considers natural. Thus children who  are brought up in a
religious community  will – egged on by religious authorities – castigate their parents for what
they see as their hypocrisy.^2 What we see with the Grigor’evs, then, is the beginning of
assimilation, not its end. Their experiences and those of their compatriots throughout the detritus
of the Soviet Union give us a glimpse of what it means to have “identities in formation.”


The Question of Identities

Stalin’s ideas on national identity continue to have a profound influence on the national identity
question throughout the former Soviet Union. For him, nations were the result of a common culture,
a common language, a common economic life, and a common territory. Scientific investigation could
determine true nations from mere ethnic or religious groups. Children are born into national
communities, and their national identification can be fixed, as it were, on the fifth line of a
passport.^ 3 Even after the collapse of the Soviet Union, and especially in light of the outbreaks
of grotesque nationalist violence in postcommunist lands, this Stalinesque view of nationality
marks quite strongly the understanding of nationality issues in postcommunist countries and
throughout the world. Analyses of ethnic conflict point to nationality groups as if they were
eternal actors on the stage of warfare. Books on the so-called new Russian diaspora worry whether
they will become a fifth column, return to Russia in a horde, or become loyal citizens of their new
republics. But the notion that they might assimilate or develop an identity other than “Russian” is
rarely even considered.^4 Reflecting this rather rigid view of national identity, Anatoly Khazanov
writes that he “was personally acquainted with Ukrainian, Belorussian, Daghestanian, Tatar, Kazakh,
Kalmyk, Buryat, Yakut, and Tuvinian scholars, including anthropologists, who cannot speak their
native language.” These people, he judges, were “doomed to acculturation.”^5 The notion that for a
Ukrainian, Ukrainian is “his” language suggests that he is not fulfilled as a person until he
recognizes his “real” identity and is doomed as an individual unless he develops the language
skills to become his real self. Even Robert Kaiser, who wrote a book sensitive to the Soviet
construction of nationality groups, writes that “the more expansive perception of homeland in
evidence among Russians [in the post-Soviet world] enhances the probability that international
conflicts will arise in border republics such as Kazakhstan, Ukraine, Belarus; and the Baltics,
where Russians live in concentrated settlements and where a Russian sense of homeland has developed
over time.”^6 He doesn't even raise the issue that if changed boundaries could affect national
identifications among Tajiks or Ukrainians in an earlier period, the changed boundaries of the
post-Soviet world might have similar effects on today’s “Russians.” The reversal of the tides in
Estonia, manifesting itself in a shift in the identity of the Grigor’ev family-from “Russian” to an
inchoate conception that includes an Estonian cultural component – all in the space of a few years,
belies the notion that the boundaries and social meaning of the Russian nationality are fixed.

What we need, if we are to place the study of Russian nationalism and identity into a plausible
theoretical framework, is a notion of “identity” and how it might be studied comparatively that
does not require us to consider people like the Grigor’ev to be anomalous. The definition of
identity I use is the one provided by the Freudian psychologist Erik H. Erikson a generation ago. I
build on Erikson’s definition of identity in order to formulate a microtheory of identity shift,
which also relies on the “tipping” model developed by Thomas Schelling. This theory can account for
both the depth of feeling and the capacity for change associated with identity politics. It allows
us to analyze the different types of identities that populations construct, such as conglomerate,
diasporic, transnational, and multiple – all within a simple framework.


Identity in Social Theory

There is a growing consensus among academic observers of identity politics that identities are not
inherited like skin color – which is the Stalinist view; its academic variant is called
“primordialism”^7 – but constructed like an art object. People, as they go through their youth, are
exposed to family, community, and national histories; they are brought up with a particular
repertoire of languages and speech styles; they may be given training in certain religious rituals.
Within their wider societies, others have adopted a variety of other social categories, local,
national, religious, linguistic. Usually people’s identities change with the level of aggregation:
within their community, they may identify themselves on the basis of socioeconomic background;
within their country, outside of their community, they may identify themselves with a brand of
politics; and outside their country, they may identify themselves with their nation. All societies
– perhaps especially today – have cultural entrepreneurs who offer new identity categories (racial,
sexual, regional), hoping to find “buyers.” If their product sells, these entrepreneurs become
leaders of newly formed ethnic, cultural, religious, or other forms of identity groups. As
individuals grow up they consequently feel pressure, in the phrase of Rom Harré, to organize
“identity projects”; that is to say, to choose the category that exemplifies them as individuals
and ties them to a social group. These identity projects carry with them, whether in religious
texts or social practices of past members, sets of “beliefs, principles and commitments.”^8
Although the choice of an identity may have had little to do with those beliefs, principles, and
commitments, by attaching oneself to such an identity project, one is expected by others hold to
them, and perhaps is motivated to do so by virtue of one’s own identification. Construction and
choice, rather than blood and inheritance, is now the standard story line about identities.

This notion of constructing an identity is modern. Although the ancients raised identity issues, it
was not until the nineteenth century, with Nietzsche and Hegel, that social theorists began
considering the transformation of identities and the emergence of new identity categories. Walt
Whitman articulated the revolutionary idea that each individual has within him – or herself a
nearly infinite set of identity possibilities. George Kateb suggests that this idea is
quintessential to the democratic age.^9

Yet twentieth-century political figures, from Woodrow Wilson to Adolf Hitler and Joseph Stalin,
continued to assume that social identities were primordially given. A school of anthropology gave
academic credence to such views. Indeed, some scholars still hold to the biological analogy of
identities and assume that they are like inherited characteristics. A 1993 study on ethnic identity
among Hispanics, relying on psychological theories of cognitive development, sought to find the
bases on which Hispanic youth brought up in the United States would have “correct ethnic labels"
and “more ethnic knowledge.”^10 In another psychologically based analysis, George De Vos saw
constructed identities as deviant. In his terms, “excessive instrumental expediency . . . betokens
inner maladjustment.” This “Zelig phenomenon;” he suggests, “occurs in what Durkheim termed anomic
social conditions,” and its presence forecloses strong emotional ties.^11

Nonetheless, prevailing social science research demonstrates that while there are many constraints
against the Zelig phenomenon, normal people do in fact engage in the construction of “identity
projects.” The motivating question in these studies is how to assess the sources of constraint. In
political sociology, the formative tradition on identity shift focused on previous societal shifts
in communication patterns. According to Karl W. Deutsch, the pathbreaker in this tradition, there
cannot be effective nationality shifts unless the probability of interacting with a person from a
different nationality is equal to that of interacting with a co-national. We are, in this
tradition, prisoners of our communications net.^12 The search for the social background conditions
for identity shift has continued with vigor. A post-collapse study of the “Yugoslav” identity found
that in the years after World War II, demographic conditions, urbanization, participation in the
Partisan war effort, and minority status within the separate republics all “predicted” the
declaration of oneself as “Yugoslav” in the state census. Even though the percentages who
identified themselves as Yugoslav were never large, the study did demonstrate the influence of
social conditions on changes in self perception of national identity. The authors were humbled,
however, by the complete collapse of the Yugoslav identity project and remain uncertain about the
persistence of newly constructed identities.^13

Proposing an alternative to the focus on social background conditions, several scholars began to
examine the role of the state (or state institutions) in manipulating the range of identities
available either by subsidizing or recognizing certain group identities and ignoring others. Others
focused on the cultural material and economic resources of the entrepreneurs seeking to empower a
newly formed identity category.^15 Still others focused on how strategies of exclusion and
inclusion by dominant cultural groups in a society tend to foster reactive identities.^16 A
compelling research tradition focused on historical legacies argued that the burdens of our
ancestors weigh heavily on who we are and who we can become.^17 Yet another research program
focused on how social networks – marriage ties, business dealings, neighborhood proximity – limit
but by no means preclude identity shift.^18 For all the focus on constraints, there is a shared
understanding, as Bhikhu Parekh puts it, that if identities are the products of history, they can
be remade by history.”^19 All of these studies promote a “constructivist” as opposed to a
“primordialist” paradigm, but they differ on the causes, constraints, and effects of that
constructing.


Contemporary Understandings of Identity

For all the debate between constructivists and primordialists, there is little agreement on what
constitutes “identity.”^20 A tried-and-true first cut at an answer to such a question is to observe
how the term is used in popular discourse. A survey of reports from the English-language press from
around the world helps us sort out these academic arguments and gives us a clearer notion of what
we mean by identity.^21 In the popular press, there is one realm in which writers insist that our
identities are primordial. Indeed, there is a clear notion of a personal identity (in the OED sense
of “the condition of being the same as a person or thing described or claimed”) in which “identity
projects” are either criminal or bizarre. These discussions about personal identities – usually
arising in legal discourse – are in a different discourse realm from that of social identities, in
which constructivist identity projects are considered permissible, though not always successful.
The resulting social identities are built on cultural materials coming from the family, the
community, and the nation, but they are not totally determined by these background conditions.
Since social identities are seen as constructed, they are always subject to reconstruction.
Following from this constructed nature of identity is a popularly accepted notion of a “crisis of
identity” when a person fails to fit easily or comfortably into any social category. A short
digression into the language of the popular press on “identity” questions should make this
distinction between primordial personal and constructed social identities clear.


Personal Identities

Personal identities are firmly entrenched in a primordial or genetic discourse realm. A person who
is x today will surely be x tomorrow. My name, my gender, the fact that I am the father of two
children, and my credit history have a DNA – like continuity to them.^22 While I (say, through a
sex-change operation) or others (say, by posing as David Laitin) may tamper with my personal
identity, such acts are considered bizarre or criminal. Indeed, press accounts of “stolen”
identities have all the appeal of the bizarre, like reports of sightings of space aliens.

“A titanic network of shared information gives each of us a credit identity but it has a major
flaw,” reports the Tampa Tribune (August 7, 1995). “A crook can steal your identity and swamp you
in so many bad debts that it could take months or even years to clear your record.” The Chicago
Tribune (June 20, 1995) reports in the same vein: “Authorities have charged Janetzke, 40, of
Streamwood with what amounts to the theft of another person’s identity. Police say he used the name
and credit history of a 35-year-old truck driver from Wood Dale . . . and even took out a telephone
number in his name. ‘He just took away my husband’s identity’ the truck driver’s wife said. ‘It s
just a big mess.’ “This practice can take on Gogolian proportions. USA Today (July 13, 1995)
reports that a Kenneth John took on the identities of forty-five dead souls, using them to kite
checks. “The growing crisis,” Cheryl Phillips writes, “costs the living billions of dollars – and
the dead their identities.”

 Legal discourse is replete with references to the genetic aspect of personal identity. The New
Jersey Lawyer (June 26, 1995) reports on the continued requirement of federal courts that the
public and the defendant “know the identity of the parties in public court proceedings in a civil
case for money damages.” The St. Petersburg Times (August 7, 1995) assures readers that
“authorities usually can enter it into a computer network and quickly learn the person’s true
identity. But if the alias has  not been used, the only way to confirm identity is by comparing
fingerprints.” Furthermore, “Even though criminals sometimes move in and out of the court system
without their true identities being discovered, Jaw enforcement officials think suspects . . .
under different names” eventually will get found out. On questions of immigration, the Federal News
Service (June 30, 1995) reports that “without documents, there is no place to begin an inquiry into
the true identity and the true purpose of an applicant for admission. . . . Every recent terrorist
act perpetrated by aliens was committed by  an alien who intentionally misrepresented his true
intention for coming to the United States or attempted to conceal some aspect of his identity,
using a claim of asylum.” The same problem occurs in Canada; the Ottawa Citizen (August 13, 1995)
reported that thousands of Somali refugees “cannot become landed immigrants because they have no
documents proving their identity.” On the issue of adoption as well, contemporary discourse accepts
personal identities as primordial. A program reported in the New York Times (June 18, 1995) allows
donor-inseminated offspring, when they are eighteen years old, to find out the names of their real
fathers. This is called the “identity-release” policy. This program is for young adults who want to
know who they “really” are, as opposed to what they have socially become.

    Everyday speech (as reported in newspapers and magazines) helps resolve a debate that consumes
the attention of social scientists. Identities are inalienable, at least when we are talking about
personal identities. Identities are also constructed, when we are talking about social membership.
To be sure, the languages of these realms are not totally distinct. We understand when Nader
Mousavizadeh means when he  writes in the New Republic (June 19, 1995) that for the Germans and
Japanese,  the “memories of war and defeat have been internalized as burdens of identity.” In a
sense, no German (of the postwar generation) can ignore that burden and  remain a German. We can
say, hardly requiring metaphor, that the subsequent generation of Germans (an identity constructed
through history) inherited that burden. So the two realms of discourse, personal and social, at
times overlap.

But this is not to deny a distinct realm for the legal notion of personal identity (which asssumes
a primordial quality). In a widely reprinted essay (see the Bergen Record, July 5, 1995, for one of
its printings), Thomas Sowell pointed out that “nothing polarizes the political left and right like
the idea of a national identity card.” But the objection to an identity card is not that
constructed, fictitious, or possibly reconstructed identities will be exposed. The objection (which
has also been raised in the United Kingdom) is that the government will know all too much about who
we really are. In this debate, as in other legal realms, identities are not in formation; they are
absolutely real.


Constructed Social Identities

Social identities are labels that people assign to themselves (or that others assign to them) when
they claim membership (or are assigned membership) in a social category that they (and others,
whether members of that category or not) see as plausibly connected to their history and present
set of behaviors.^23 It is further implied that this assignment has powerful emotional appeal, both
to its holder and to others in the society.

Social identities are distinct from personal identities, and they are built from available
categories that both divide and unite people in a society. People have inter alia national
identities, racial identities, religious identities, and hometown identities. Yet issues of social
identity become part of public discourse only when die categories themselves become fuzzy. Self
appointed boundary-keepers arise to redefine these categories so that rules of inclusion and
exclusion, as well as the behavioral implications of belonging to this or that category, can be
clarified.

One of the main reasons there is so much talk of identity in the press in our times is that the
boundaries and behavioral implications of many of our social categories are being contested. Gerald
Poyo reports for the Houston Chronicle (August 2, 1995) that recent immigration laws are forcing
“immigrants to conform to a mythical and narrow notion of American identity.” Digby Anderson,
writing for the Sunday Telegraph (July 23, 1995), sticks needles into the chief executive of the
United Kingdom School Curriculum and Assessment Authority for a similar reason:


He is undoubtedly right in worrying that the children may be learning a wishy-washy
multiculturalism, a sort of cocktail identity. But he is wrong in suggesting Britishness classes.
For the true identity of most of our children is not Britishness at all. It is Englishness. Indeed,
Britishness is almost as artificial and newfangled an imposition as multi-culturalism. The vast
majority of the British are English . . . not British. Scottish people tell me they think of
themselves first as Scottish and only second as British. English people just do not go about
thinking of themselves in this explicit sort of way. . . . Englishmen travel, it is the same. An
hour from arrival at Kennedy Airport, as the passengers scratch their dried-up ballpoints over
their crumpled immigration forms, the attentive British Airways staff have to tell countless of
them their identity lest they get it wrong: “No, no, it’s not England. You have to put UK.” Who on
earth thinks of themselves as a UKer? There is no such thing as a British village. So, if any
identity is to be taught in the schools, at least in English schools, it is not Britishness but
Englishness. To teach Britishness would take about 10 minutes, for all there is of it. What does
Englishness consist of? . . . If the aim is to educate, to inculcate manners and identity – the
manners and identity of England – then it must be cricket: and old-fashioned cricket at that.

But Mike Marqusee, in an earlier article in the Guardian (July 4, 1995), on the implications of a
Canadian playing on the English tennis team, had trouble finding even Englishness. He reveals that
“the truth is that the ‘English’ national identity was problematic long before immigrants from the
West Indies and South Asia be an arriving in the fifties. In fact, he admits, “Outside war zones,
sporting teams are today the main visible bearers of national identity.” Capturing an “English”
identity is like grasping a wet bar of soap. High levels of immigration lead education authorities
,and other officials to feel the need to grasp that bar. Correspondents, who feel rather more
secure than the officials, are freer to report on these issues with a smirk.

In Eastern Europe, of course, talk about national identity does not have this ironic tone. In an
article about Serbs and Croats, Michael Ignatieff in the Ottawa Citizen (July 2, 1995) frets that
“nationalism is a fiction of identity, because it contradicts the multiple reality of belonging. It
insists on the primacy of one of these belongings over all the others. So how does this fiction of
the primary of national  identity displace other identities? How does it begin to convince? Here we
begin to reach for theory.” He writes, “Globalism brings us closer together, makes us all
neighbors; it destroys boundaries of identity and frontiers between states. We react by insisting
ever more assiduously on the margins of difference that remain.”

For secure communities – such as the English in Britain – identity talk has an iconic tone. Like
East European minorities, however, race and gender groups in North America and Western Europe take
their identity projects with resolute seriousness. The St. Louis Post Dispatch (June I8, 1995) had
a headline, “A Respectful History of Gays and Lesbians,” and the article reflects on the 100-year
history since the 1st articulation of a “homosexual identity.” Newsweek (July 17, 1995) reported
that in the 1990s bisexuals are “now claiming their own identity.” The Bisexual Resource Guide, the
magazine reports, lists 1,400 groups throughout the world, including “Bi Women of Color.” In a
sense, a choice of identity is not, Zelig-like, completely free (it would be outrageous, at least
in some circles, for a homosexual white man to claim he is a bi woman of color); but in another
sense, identities are constructed (it is hard to imagine anyone claiming the identity of bi woman
of color before 1980).


Crises of Identity

Because social identities are constructed from the available repertoire of social categories,
misfits are inevitable. Some people cannot find a label that adequately represents their
identities. Or they may not like the identity they have chosen or were compelled to go by. Consider
the case of Maria Maggenti, who in the Village Voice (June 27, 1995) reported that throughout her
“adult life, I have called myself a lesbian. A dyke, sapphist, muff diver, lover of women. In this
identity, I found a home for my desire, my politics, my upside-down sense of humor.” But to her
horror, she fell into a heterosexual relationship. She was in shock: “To me, calling myself a
bisexual versus claiming a lesbian identity is the difference between Muzak and Mahler.” She is
facing, in our everyday language, an identity crisis.

The Houston Chronicle (August 2, 1995) diagnoses incidences of this sort as “gender identity
disorder.” The newspaper reports that “gay and lesbian teen-agers often find society unsympathetic
to their situation, and this can provoke psychological problems, many of which are also symptoms of
gender identity disorder: depression, attempted suicide, alienation and what psychiatry calls
‘borderline personality disorder.’” Readers were assured that at least in California, such people
would not be hospitalized, since “insurance companies just wouldn’t pay for it.”

Ambiguity of identity indeed represents a “problem,” or more commonly a “crisis.” In the United
States, Newsday (July 13, 1995) asks whether black children raised by white parents develop a
positive sense of self and a strong racial identity. The Orlando Sentinel (July 9, 1995) reports an
initiative by the U.S. Census Bureau to include a “multiracial” category in order to “help children
of biracial marriages with their self identity.” To go back to the sardonic article about British
identity (Sunday Telegraph, July 23, 1995), Digby Anderson further remarks that “crises of identity
are for people who haven't got one, like the Belgians, or who have got too many, such as the
Italians.”

This notion that lack of a clear identity can lead to a “crisis” is strongly contested. In Britain
it pits conservatives against liberals. In one formulation (reported in the Guardian, July 20,
1995), “The liberal believes that a man, once stripped of his national and cultural identity, will
become Everyman – citizen of the world. The conservative knows that, in fact, he will become
bewildered, schizophrenic, unhappy and lonely.” Liberal thinking remains strong in the United
States. An editorial in the San Francisco Chronicle (June 29, 1995) argues that “racism requires
the destruction of an individual’s confidence in his own mind. Such an individual then anxiously
seeks a sense of identity by clinging to some group, abandoning his autonomy and his rights,
allowing his ethnic group to tell him what to believe.” And an article in the Village Voice (June
20, 1995) reported that Georgia O’Keeffe “detested being considered a woman artist.
Identity-related adjectives attached to the noun ‘artist’ always demean.” These protests, in
support of the rights of individuals against outsiders’ attempts to label them, demonstrate the
power of these identity categories to subsume and even colonize individuals. In a sense, the
protests demonstrate the power of identity categories in spite of their arbitrariness and
constructed nature.

Intellectuals who understand that identity categories are constructed yet wish to fight for
opportunities for people with whom they identify face a problem the reverse of O’Keeffe’s. How can
they purposefully reify categories, giving people with complex pasts a single dominant label, when
they know those categories are constructed? Anthony Appiah addressed this problem with great
sensitivity while advocating the politicization of an African American identity. Gayatri Spivak
articulated the concept of “strategic essentialism” to address this issue. As Lisa Lowe puts it
Spivak argues that “it is possible to utilize specific signifiers of ethnic identity, such as
Asian-American, for the purpose of contesting and disrupting the discourses that exclude
Asian-Americans, while simultaneously revealing the internal contradictions and slippages of
Asian-American so as to insure that such essentialisms will not be reproduced and proliferated by
the very apparatuses we seek to disempower.”^24 On the one hand, identities such as African or
Asian American can mobilize thousands of adherents; on the other hand, these identities, when
careful archaeological work is done, are revealed as fabrications.

Both academic and popular analyses of the concept of identity (and especially  national identity)
are thus “both awed by its power and dumbfounded by its  weakness.”^25 By emphasizing the
constructed character of identities, they tend to underrate the power of identity attachments to
guide behavior, to drive people into incredible acts of heroism and terror. But when these acts of
heroism and terror are reported, suddenly the language of primordialism (the Stalinesque categories
of membership) is revived. Thus reports of the Croatian and Bosnian ethnic wars in the wake of the
Yugoslav collapse often refer to “ancient hatreds,” hatreds that for some reason did not stand in
the way of the high levels of intermarriage in previous generations between the combatants of the
mid-1990s. These analyses elide the historical facts of constructedness and change. Clearly we need
a better understanding that accounts for both the constructed nature of social identities (the
current conventional wisdom) and the power of these identities to seem natural to those who hold
them. The search for this better understanding takes me beyond everyday discourse and on to
approaches developed in psychoanalysis and game theory.


A Definition of Identity

The popular notion of a “crisis of identity” comes from the pioneering psychoanalytic work of Erik
H. Erikson. A reading of his work sheds new light on these debates on identity and helps us address
the problem of explaining both the constructedness and the power of social identity categories.
Erikson attributes his focus on identity to a cryptic autobiographical passage in Freud’s corpus,
when he spoke about his “consciousness of [his] inner [i.e., Jewish] identity” Erikson went far
beyond the master in developing a psychoanalytic notion of identity, even seeking to explain the
sources of the Protestant Reformation in one man’s identity crisis.^26

For Erikson, here relying on William James, identity is “a voice inside which speaks and says:
‘This is the real me.’” Finding that identity is often a lifetime quest, and failure to find it (as
Erikson suggests was the case with Hitler) can have a damaging impact on oneself and others. In
light of this quest, which is both personal and social, Erikson sees “identity formation [as] a
process ... by which the individual judges himself in the light of what he perceives to be the way
in which others judge him in comparison to themselves and to a typology significant to them; while
he judges their way of judging him in the light of how he perceives himself in comparison to them
and to types that have become relevant to him.”^27

This formulation elegantly captures what I have elsewhere called the “Janus-facedness of
culture.”^28 One face of culture reveals identities to be real and given, to be something that can
be searched for and discovered. Theories of culture that rely on primordialist imagery see only
this face of identity. Social solidarities are built on real foundations. While we may lose our
bearings, our true identities are there for each of us to find.

But the second face of culture – and here I focus on the quest rather than the goal, the “real me”
– is not primordial but instrumental. This face of culture reveals identities as constructed and
reconstructed as social opportunities change. Ernesto Laclau is discussing this face of culture
when he argues that “once the obviousness of social identities was put into question,” it was no
longer possible to imagine people “discovering or recognizing their own identity.” Rather, the
problem today is to think about “constructing’ an identity with the “explicit assertion of a lack
at the root of any identitv.”^29 Gustave Flaubert understood this face of culture well. L’Heureux,
the clever draper, he wrote in Madame Bovary, was “born in Gascony, but a Norman by adoption.” In
this way he could marry his southern verbosity with a northerners’ cunning. Identities from this
point of view are adopted, or constructed, according to how well they serve individual purposes and
reconstructed to take advantage of new opportunities.

It is here that Erikson is so useful. Primordialists and constructivists live in their separate
intellectual universes, each deriding the blindness of the other. Neither side comprehends that
each is looking at only one face of Janus. But Erikson positioned himself to see both faces. He
understood, with the primordialists, that not any identity will do. People are limited in their
senses of self by their families, their communities, the prevalent typologies of identity that
surround them, and what he called the “identity possibilities of an age.”^30 In this sense,
identity is given. Yet individuals also seek to adjust their identities to the judgements of
relevant others and tend not to settle on an identity for much of their youth. In this period of
search, individuals look to themselves and others, trying out new identities to see how they feel,
both to themselves and to others judging them, before adopting one permanently. In this sense, for
Erikson (and for Whitman’s democratic citizen), identity is constructed.

Identities are therefore categories of membership that are based on all sorts of typologies-gender,
race, class, personality, caste. People are limited by, but they are not prisoners of, their genes,
their physiognomies, and their histories in settling on their own identities. And if powerful
social forces motivate identity exploration – as they seem to do in our age – it is the
constructivist face of identity that seems the more real.


How Identities Change

A compelling model of identity must not only define the concept in a coherent way – as I believe
Erikson has done – but also be able to account for both the impressive power of identity groups to
give their adherents a sense of natural membership and the equally impressive power of individuals
to reconstruct their social identities. Relying on a model developed by Thomas Schelling, I propose
to interpret identity shift in terms of a “tip” or “cascade.”^31 Tips and cascades are common
features of social life. Consider the case of one or two African Americans who buy homes in a
stable “white” neighborhood. Suddenly the white families, fearing that they will be the last whites
in the neighborhood, all seek to sell out at the same time. But only African Americans are willing
to buy. Very quickly the neighborhood “tips” from stable white to stable African American.

In the late 1980s, protest cascaded across Eastern Europe in a similar way. Societies which street
protesting was nonexistent experienced sporadic demonstrations that were not quickly put down.
Suddenly, protest grew to literally revolutionary levels. Thoroughly hopeless and demobilized
societies suddenly became highly mobilized and active. In 1988 political protest seemed impossible;
in 1989 it was normal.^32

Such cascades occur because people’s choices about their actions are based on what they think
others are going to do. If I think none of my neighbors will sell his house if a few African
American families move close by, I have no incentive to sell mine. But if I think many others will
– or better, if I think many others will think that many others will – then I have an interest in
selling my house before those others do, that is to say, before property values plummet. Or in the
case of protesting: if I think that no one will be out picketing in the streets, I know I will be
an easy target for the police. But if I think that others will be out – or if I believe that many
others will be sure that many others will be out – suddenly prudence no longer dictates that I
remain at home.

Both of these situations have two stable equilibrium outcomes: an all – “white” or all – “African
American” neighborhood; and streets with no protesters or streets filled with protesters.

Identity shift can also cascade. In order to keep the discussion and the model focused, I limit my
remarks to one aspect of identity – one’s language community.^33 Like almost all social identities,
one’s language community (or mother tongue, as it is popularly understood) often has a near
mystical quality conferring membership in a category of similarly endowed people. Yet language
repertoires, like social identities, are subject to rapid intergenerational shift. Therefore, we
hear stories of language retention despite all efforts by a state to erase a language,^34 and
stories as well-for example, of Lutheran children in nineteenth-century Ohio mocking their
ministers who insisted on speaking German – in which mother tongues get lost within a generation.
Both story lines are accurate. When my grandparents came to New York in the late nineteenth
century, they knew that other children of Yiddish speakers would be learning English, and it would
be irrational for them to seek to maintain the intergenerational transmission of Yiddish.
Meanwhile, when Russians moved into the “virgin” lands of Kazakhstan at that very time, they fully
expected other Russians to maintain the linguistic repertoires they had in the Russian heartland.
Here we have examples of opposite and extreme equilibria. In New York after a generation, hardly
any monolingual Yiddish speakers were left. In Kazakhstan after a generation, very few descendants
of Russian immigrants were even bilingual in Russian and Kazakh.

The tipping model can account for both the constructed nature (to those who study it) and the
naturalness (to those who live it) of social identities. At any equilibrium, it appears to actors
that the world is completely stable. In this situation, identities are not under question. There is
(by the definition of equilibrium) no incentive for anyone to explore new identities. It is obvious
to people who in fact they are. A point of coordination, in which there is a tacit understanding
among all people in a community that this is an aspect of their identity (for example, that
Russians address all others living in Kazakhstan in Russian), is an example of what Thomas
Schelling calls a focal point.^35 Cultural and political elites of a group in equilibrium, by
giving meaning to the equilibrium – that is, by providing it with the “beliefs, principles and
constraints” that Harré identified – make it into a focal point. In this way they expand their
authority and gain the legitimacy to speak on “their” group’s behalf. In Max Weber’s terms, these
elites will seek to ascribe ultimate value to the focal point identity – that is, give it “value
rationality” -in order to defend it against all pragmatic considerations.^36

Exogenous events (such as the independence of Kazakhstan), however, can nonetheless bring some
instability, with certain people exploring new identities (such as Russian-Kazakhstani). At such
times, cultural entrepreneurs of the once-stable identity emerge and try to stem any tide away from
the old equilibrium, and seek to naturalize, or essentialize, the status quo ante. Those claims
will appear compelling, for the very fact that there was coordination at a particular equilibrium
over generations does, indeed, make that equilibrium look like a law of nature. Other cultural
entrepreneurs, more forward-looking, will seek to induce a cascade toward a new equilibrium, and if
they are successful, the change will be thought of as natural, or inevitable. The tipping game
therefore shows why identities are powerful “focal points” of coordination, yet are also subject to
change.

One might argue that adopting a new language does not automatically mean one has adopted a new
identity. Yet if Liuba Grigor’ev had felt like a traitor or a fool for taking Estonian language
lessons, her motivation would have been sapped. Her identity was becoming a “Russian who has
accommodated to the realities of Estonian sovereignty.” This was the real ‘Liuba.” But these
microadjustments in identity – a nuance Erikson did not consider – alter the identity possibilities
of a following age. In this sense, Liuba’s quest to keep her family intact lays the foundation for
a constructed Estonian identity for her grandchildren. In Schelling’s terms, she is in her
microactivities moving her family, and the Russian-speaking community of Narva, toward an identity
tip.

One might also object that the tipping game, with its emphasis on binary choices, does not capture
the fact that multiple identities are common in social life. Many people, for example, want to be
Catalans, Spaniards, and Europeans all at the same time. Some, I am sure, see themselves as
lesbians, workers, Catholics, and Hispanics and alter the emphasis to fit the context. A resident
of Harlem might identify himself as a black in the context of New York politics but as an American
in the context of international affairs. People’s “identity projects” are clearly more nuanced than
making either/or choices between matched pairs of identity alternatives.

Multiple identities, however, can coexist within a person only insofar as choice is not necessary.
Yet when the actions or behaviors consistent with one identity conflict with those of another
identity held by the same person, as they do when the two identities represent antagonistic groups
on the political stage, people are compelled to give priority to one identity over the other. A
person who sees herself as both a “Russian” and an “Estonian” may one day have to choose – say, if
there were a border war – which identity is dominant. In this case, the tipping game is a powerful
analytic tool, as surely one s choice is affected by the number of people (of like multiple
identities) who have given priority to their “Russian” or their “Estonian” identities.^37

Consider the rise of a politicized homosexual identity in the United States and its implications
for an African American whose sexual preference was heretofore a private matter. Perhaps his
partner or some of his past partners have been mobilized into action as homosexuals. Now they are
in daily political alliance with whites and Hispanics, and their former political identity group,
the African American community, becomes less prominent as a basis for their political information
and mobilization. It is useful to model this man’s choice situation and do an accounting of the
payoffs for “coming out” politically as a homosexual based on how many other black homosexuals have
chosen to come out, and the social benefits and costs of either decision, again depending on the
percentage of African American homosexuals who have reoriented their political activity. This is
not to deny that this person’s identity is multiple; rather it is to emphasize that the everyday
reality of identity politics forces us to weigh alternative presentations of self, keeping in mind
how others, like ourselves, are representing themselves. This is a basic dynamic in identity shift.

A final objection to the tipping motif has to do with the rational choice framework itself. How can
one calculate the costs and benefits of identity shift, especially under conditions of trauma and
uncertainty?

In response to this skepticism about rational choice models, I should like to offer two preliminary
counters. First, it is correct that a variety of identity projects can be offered to a population
in crisis. In this book we will observe two of them, toward a titular and toward a conglomerate
identity. Before people can strategize, they need to know what the choices are. Much of the “work”
of identity choice, in consequence, precedes the tipping dynamic.^38 In times of crisis, then,
people may be playing more than one game at a time. Nonetheless, the tipping model neatly
encapsulates people’s strategic dilemmas once the game has begun. In this sense, the tipping game
is but a partial rendition of the overall cultural dynamic.

Within the tipping game, however, and somewhat counterintuitively, it is not the case that trauma
and uncertainty undermine rational calculation. Despite the powerful arguments of Ann Swidler to
the contrary, in unsettled times, people – at least those families with whom I and my fellow field
researchers interacted – feel compelled to calculate and coordinate their calculations with others.
In this sense – as the ethnographies in this book make clear – uncertainty is the breeding ground
for coordination dynamics such as those evident in the tipping game.^39

^

National Identities

National identities, like social and neighborhood identities, have cascade qualities. National
projects – such as Catalan, Flemish, or Mayan – may seem quixotic and antiquarian at one point in
time, yet suddenly burst onto the historical stage as if by spontaneous combustion. In the modern
age, national projects have usually involved the reinsertion of a folk language as part of the core
identity of people who are descendants of speakers of that language, most of whom rely principally
on a more cosmopolitan state language. Language movements, with the goal of restoring languages in
desuetude as a tool to create modern nations, have been a constant source of identity politics in
the modern age.

Nationalist politics involve two interrelated identity issues. First is the issue of a “national
revival” in a relatively homogeneous region in a culturally heterogeneous state. Consider Estonia
in its period as a Union republic of the Soviet state. Estonians who wished specialized higher
education and occupational mobility found the inclusion of Russian in their language repertoires to
be of great value. Within a  generation, it became normal to rely on Russian in a variety of
professional and political domains. This is why families such as the Grigor’evs had no need to
learn Estonian. In the late 1980s, radical Estonian nationalists, fearful that the massive
immigration of Russians into their republic had set in motion a long-term threat to the viability
of the Estonian nation, pressed for the elimination of Soviet-period  immigrants from Estonian
political life. This move would make possible the sole use  of Estonian in all political,
educational, and administrative domains. From their point of view, such policies would forestall a
tip toward Russian-language dominance in Estonia.^40 These nationalists were seeking to undermine a
Soviet-inspired identity project that emphasized the merging of nations. Nationalists in other
republics, where Russian was beginning to replace the titular languages in many social domains,
were also seeking to reverse a linguistic tide. The key for regional nationalists, whether stemming
or reversing a tide, is to induce their followers to abjure the central language. An important
element for success is to get people from the regional culture to believe that all their fellow
regionals are already beginning to switch to a regional-dominant language repertoire. To the extent
to which people who identify ethnically as members of the regional culture rely principally on the
regional language for family, work, and cultural affairs, we can say that the nationalists have
successfully induced a tip in the national-revival game.

 A second issue in nationalist politics involves the “assimilation” of members of minority groups,
or immigrants, into the new national culture. To the extent that the Estonian national revival is a
success for example, people such as the Grigor’evs,  who relied principally on Russian, must work
to add Estonian to their repertoires and to seek education for their children in Estonian. They
will, of course  keep a careful eye on the choices made by fellow Russian-speakers. If all
Russian-speakers feel that all others will remain monolingual in Russian, they will see little need
to learn Estonian. But if they fear that many others are already adjusting to the new language
regime by learning Estonian, they will feel pressure to join the cascade.

In both national revivals and assimilation cascades, there are political pressures to alter one’s
“identity project.” I shall now explicate these two processes more formally from the perspective of
the tipping game.


National Revival

In his explication of the problem of consolidating a national revival, Ernest Gellner referred to
the revivalist region as “Ruritania” and the central state as “Megalomania.”^41 He assumed that the
political leaders of Ruritania are fully bilingual in both the language of state power and the
language of the region. As Megalomania consolidated power, he explains, the great mass of the
regional population became at least partially assimilated as well. But there was resentment, and
Ruritanian leaders sought sovereignty for their cultural group in a national revival.^42

Most portrayals of political movements that are dedicated to a region’s political, economic, or
cultural autonomy focus primarily on the conflict of interest between the power at the center and
the united national movement at the periphery. What these portrayals often miss is the conflict of
interest that exists within the regional nationality population. The major problem for regional
revivalists is to induce a tip within their own constituencies.

Let us suppose that in Megalomania, of which Ruritania was a part, access to wealth and power
required fluency in the language of the state. In fact, residents of any distinct region would be
able to communicate with residents from any other distinct region only through the state language.
Usually, under these conditions, people who live in culturally distinct regions become
“diglossic.”^43 This is a special form of bilingualism, where the state language is used mostly for
“high” functions (such as trade, high culture, and contacts with state authorities) and the
regional language is used for “low” functions (for intimacy, and for celebrations of folk culture).

If Ruritania, whose population is diglossic, is to achieve a national revival, its leaders must
reverse the functional domains of the two languages. Many people who consider themselves
Ruritanians will feel uncomfortable using what they see as a folk language in domains of high
culture or technology. They are likely to believe the myth that “their” language is not capable of
expressing complex or modern thoughts. Those Ruritanians who are members of the local bureaucracy –
those whose jobs and advancement required facility in Megalomanian – will have an interest in
maintaining the linguistic status quo.

The national revival tipping game – displayed in Figure I.I – illustrates these constraints. Here,
for illustrative purposes, I leave Gellner’s fantasy world of Megalomanians and Ruritanians for the
actual world of the republics of the former Soviet Union. On the x axis I have plotted the
percentage of the regional nationality (whom I shall call here, with the Soviet case in mind,
titulars) in full compliance with the laws and principles of the nationalizing movement (the
function TT) and the percentage of the regional nationality continuing to rely on the language of
the former center (the function RR). On they axis is the payoff an individual receives for his or
her linguistic choice. The payoff for an individual linguistic choice depends on how many other
individuals made the same choice.



Figure I.I. National revival game: Percentage of titulars who comply fully with nationalist
language laws


The dilemma portrayed in this diagram reflects practical decisions that real people face. Those
school principals, directors of industrial plants, newspaper publishers, retailers, and
professionals who operated in environments where Russian was  the dominant language need to decide
whether to comply with new state laws requiring the monopoly of official use for the newly
legislated national language. Parents (of the titular nationality) need to decide whether to send
their children to titular-language schools. If the children have already been attending
Russian-language  schools, the cost of change will clearly be high, and the change will be worth it
only if other enterprises (or parents) adopt it as well. Therefore, no organization (or parent) has
an incentive to move first, even if all agree that the payoffs for all would be  higher at 100
percent indigenous than they are at, say, 20 percent indigenous, which is the point at which many
revival movements begin. This is why many titulars will vote for a nationalizing program, believing
in it theoretically, but then subvert it in their own private and professional lives.^44 When
people personally subvert the goals of the very movement they have given their elected leaders a
mandate to promote that movement will fail. It will fail not only because of the nefarious
interference of the declining center (although the revivalists will want to blame their failure on
the center’s lust for reintegration) but also because of the rational linguistic strategies of
already partially assimilated members of the nationality group in whose name the revival is being
promoted. Cultural entrepreneurs directing a national revival movement must somehow induce a
language tip, and they can do so by convincing key  members of the titular population that other
key members are already in the process of changing their linguistic practices. The success of this
effort cannot be predicted from the size of the vote in favor of new nationalizing language laws.

With such heavy constraints facing regional revivals, how do their leaders  push societies toward
the desired equilibrium state? This is clearly the question  regional revivalists everywhere face.
As we shall see, historical patterns in the way regions have been incorporated into centralized
states tell us a good deal about the possibilities for success of regional revivals. Furthermore,
the degree of success of regional revivals is an important clue to the power of competitive
assimilation, the second crucial game in the drama of nationality politics.


Competitive Assimilation

Let us now consider the situation of the members of an immigrant nationality (let us, following
Gellner, call them Ruritanians) in a state (Megalomania) in which the dominant language is
different from their own. Let us further suppose that this immigration was economically, not
culturally, motivated. These Ruritanians very much want to maintain their language and culture and
to pass it on to their children, who will, they hope, subsequently pass it on to their children. If
the entire Ruritanian community thinks more or less in this way, they will be able to demand from
Megalomania a certain degree of cultural autonomy, such as the right to maintain
Ruritanian-language schools, and to have local administration and legal proceedings conducted in
Ruritanian.

Suppose, however, that at the time these people immigrated there were no schools, no local
services, and no entry-level middle-class jobs in which Ruritanian was used. Further suppose that
those who wanted those jobs would need to be literate in Megalomanian. It would then be rational
for an immigrant to send her child to a school that ensured rapid training in Megalomanian. A
Ruritanian child attending such a school would have a competitive advantage in the upwardly mobile
job market. But if it is rational for any parent to do this, then all parents will think that other
parents will probably do it, and that therefore they should get their children competent in the
language of their new society (i.e., move toward linguistic assimilation) before the middle-class
job market is saturated.

Under these conditions, which are more or less what immigrants to America have faced for over a
century, we get rapid assimilation into the national language even if all parents agree that it
would be better if all immigrant families held together and nurtured their home culture in the new
environment. In the case of the Russians in the titular republics, however, as represented in
Figure I.2, the rate of change may be slow at first until the number of people who switch begins to
increase. As more Russians learn the titular language, others will perceive the trend and calculate
that the payoffs for not speaking the titular language will, before too long, be lower than those
for learning it. As this process unfolds, and the feeling spreads that the direction of change is
toward the language of the national state, the rate of change will rapidly increase.




Figure I.2.Competitive assimilation game: Percentage of Russians who speak titular.


When are such cascades most likely to begin? Because this book focuses on the plight of the Russian
immigrants in the near abroad, this game gets special attention. As a preliminary matter, it should
be clear from a perusal of Figure I.2 that at the far left of the x axis, it is still irrational
for Russians to study the titular language. Something must occur to change the payoffs for at least
a few pioneers. From a macro point of view, changes in world trade patterns, interstate relations,
immigration possibilities, and state policies on education and administration have discernible
effects on individual incentives and can certainly induce a few Russians to shift their linguistic
repertoires. A micro perspective focuses on how such changes actually happen, and under what
conditions the tipping point (k) will be reached. From this micro perspective, several
(un)favorable conditions can be enumerated as initial hypotheses. First, assimilation cascades are
likely to occur when the expected lifetime earnings of a young person are substantially greater
when that person is  fluent in the language of the state in which the family now resides. Second,
assimilation cascades are likely to occur when the immigrant community is itself divided and puts
few constraints on its members to remain a tight-knit community. Put another way, assimilation
cascades can be halted if cultural entrepreneurs within the immigrant community can raise the
status of people who refuse to give up their cultures, or lower the status of those who mimic the
practices of the majority culture. Third, assimilation cascades are likely to occur when members of
the majority culture accept as one of their own (on the marriage market, in social affairs) those
immigrants who have attempted to assimilate. I call these three factors expected economic returns,
in-group status, and out-group status. Calculations about these returns,  I hypothesize, will have
implications for the likelihood of a linguistic tip, with a concomitant change in the Russian
population’s social identity; perhaps after the tip they will see themselves as “Bilingual Russian
Titulars.”


Assimilation, Diasporas, and Conglomerate Identities

Throughout this book, I refer to the Russian population living in the states of the former Soviet
Union as a diaspora – although, since they acquired that status because the borders of the Soviet
Union receded, rather than because they dispersed from their homeland, it is perhaps better to
think of them as a beached diaspora. Yet we should not forget that these Russians are being pressed
not only to assimilate but also to consolidate as part of a conglomerate identity group. calling
them a diaspora tempts one to forget about the social pressures for assimilation. Calling them a
conglomerate identity group tempts one to place them in the same category as Hispanics or Asian
Americans in the United States and appeal to the literature on reactive identity formation. The
naming, as it were, presupposes the category of analysis.^45 One of the advantages of the tipping
model I propose is that it allows us to analyze the identity situation faced by Russians without
presupposing the genre of group they have become.

In the social science literature, many attempts have been made to distinguish assimilation from
other forms of cultural and political incorporation into dominant society.^46 But the tipping model
allows us to cut through many of those distinctions, and to talk simply about rates of assimilation
in a variety of contexts. For our purposes, assimilation can be defined as “the process of adoption
of the ever changing cultural practices of dominant society with the goal of crossing a fluid
cultural boundary separating [minorities] from dominant society.”^47 From the point of view of the
tipping game, assimilation can be thought of as a successful switch, in a variety of cultural
realms, to the practices of dominant society. To the extent that in a range of these cultural
realms the minority population crosses the tipping threshold, we can say that societal (as opposed
to individual) assimilation has occurred. Note well that even with apparently complete
assimilation, there will always be those “half forgotten poets and lonely philologists” whose
expected returns for holding on to languages and rituals in desuetude are larger than the returns
for assimilating.^48 These cultural elites will always be ready, in the hope that social conditions
will someday allow a tip back to the status quo ante. In this sense, assimilation; according to the
tipping game, is never completely settled.

What, then, is a “diaspora”? In the contemporary literature, the attempt to distinguish “diasporas”
from immigrants, expatriates, refugees, guest workers, exile communities, overseas communities, and
ethnic communities has led to a plethora of distinctions that make little difference.^49 William
Safran recognized that originally “diaspora” referred only to Jews, and it was therefore somewhat
redundant to theorize about diasporas as a category of communities. Defining it merely as a
“segment of people living outside the homeland” (as Walker Connor does), however, dilutes it of all
its meaning, as it would then include all immigrant communities. Safran therefore suggests the
following criteria,


1)they, or their ancestors, have been dispersed from a specific original “center” to two or more
“peripheral,” or foreign, regions; 2) they retain a collective memory, vision, or myth about their
original homeland – its physical location, history, and achievements; 3) they believe that they are
not – and perhaps cannot be – fully accepted by their host society and therefore feel partly
alienated and insulated from it; 4) they regard their ancestral homeland as their true, ideal home
and as the place to which they or their descendants would (or should) eventually return – when
conditions are appropriate; 5) they believe that they should, collectively, be committed to the
maintenance or restoration of their original homeland and to its safety and prosperity; and 6) they
continue to relate, personally or vicariously, to that homeland in one way or another, and their
ethnocommunal consciousness and solidarity are importantly defined by the existence of such a
relationship.^50

^

Jews remain the paradigmatic case, but Armenians, Turks, Maghrebis, Palestinians, Cubans, Greeks,
and maybe the Chinese meet at least some of these criteria.

From the perspective of the tipping game, no such subtle distinctions, or criteria of inclusion,
are necessary. A diaspora is a population living in a society distant from the homeland that its
leaders claim as their own, and to which they expect one day to return. By emphasizing the group’s
diasporic qualities, identity entrepreneurs seek to raise the probability in the minds of their
constituencies of a successful return to that homeland. To the extent that they are successful – by
inculcating a sense of nostalgia among people who have never lived there, for example^51 – they
will be able to stem the tide of complete assimilation, for the members of their identity group
will want to cultivate intergenerationally the linguistic skills, cultural knowledge, distinct
names, or at least a sharp memory of ancestral belonging that would become essential if they or
their descendants were in fact to return. This is a form of in-group policing, giving social status
to people who invest in the possibility of a future return.

A “conglomerate” identity is a category of membership that is a common denominator among a set of
identity groups that share some characteristics that are distinct from those in the dominant
society in which they live. One must not think of conglomerate identities as false by definition;
after all, most of today’s nationalities – note the earlier discussion of Britishness -are
conglomerates of historically separable elements. Conglomerate identities often form when members
of dominant society  refer to a set of distinct groups in a common way (as South African Boers
would talk about “kaffirs”); but conglomerate identity groups can also arise when the social
boundaries separating a set of related groups all living on a foreign soil (as is the case of
Ukrainians and Russians now living in Kazakhstan) are relatively weak. Under  conditions of a
“conglomerate identity” group, there is often no credible traditional elite to protect the group’s
boundaries or to punish defectors who seek to reidentify themselves as members of the dominant
society. Conglomerate identity groups may therefore be demographically large but politically weak.
Lisa Lowe recognizes this in her analysis of an Asian American identity. “As with other diasporas
in the United States,” she reasons, “the Asian immigrant collectivity is unstable and changeable,
with its cohesion complicated by intergenerationality, by various degrees of identification and
relation to a ‘homeland,’ and by different extents of assimilation to and distinction from
‘majority culture’ in the United States.”^52 Theoretically it makes no difference whether Asian
Americans (or Hispanics) are diasporas or not; the important point is that by lumping Japanese and
Chinese within the same identity group, or Cubans and Mexicans – as a conglomerate identity – the
forces of dominant society easily overrun attempts by Asian American or Hispanic activists to
protect the integrity (or separateness) of constructed Asian American or Hispanic cultures.

What, then, are the Russians in the near abroad? Are they a diaspora like the Palestinians, forming
a bomb about to detonate? Or are they the forgotten people in the drama of decolonization, like the
pieds noirs, likely to be evacuated from their republics and ignored in their supposed homeland?
Maybe the analogous case is the English Rhodesians, a settler community that will continue to
maintain its privileges, even though they will never develop a true Zimbabwean identity. Still
again, as the Russians, Ukrainians, Belarusans, Poles, and Jews develop a sense of a common plight,
as a “Russian-speaking population”, the analogous case may be the Asian Americans or Hispanics, in
the development of a docile conglomerate identity.

My answer to this question is that the analogy should not precede the analysis, but rather should
follow from it. No category – neither settler, diaspora, nor conglomerate group – precisely fits
the situation. The real question is the extent to which cultural entrepreneurs in the former Union
republics will be able, like the leaders of the Palestinian Liberation Organization, to lay cogent
claim to a lost homeland that will one day be returned to them. To the extent that this move
succeeds, the analogy of diaspora will be self-fulfilling. To the extent that Russian leaders feel
compelled to categorize themselves as part of a larger group, involving non-Russian
Russian-speakers, they will have a more difficult time making claims about a common religious
identity, or a common homeland that one day will be returned to them. I do not seek to fit Russians
living in the near abroad into a particular category. My purpose is rather to understand the
dynamics of identity shift and the implications for the kinds of states they will be living in, and
for the degree of conflict they are likely to experience in their relations with the titular
populations.


Preliminary Statement of Findings

This book provides a coherent explanation of why Liuba Grigor’ev has already made some major
cultural adjustments in the direction of assimilation. It does so in a way that differentiates her
from Russian-speakers in Ukraine and Kazakhstan, who face different pressures and have different
opportunities. This book takes several hefty jumps into the rarefied atmosphere of theory –
especially as the tipping game is developed – but it does so in a way that is grounded in the lives
of real people.

More specifically, this book shows that each of the titular republics was incorporated into the
Russian and Soviet states in one of three ways (Chapter 3). The key to distinguishing these three
macro patterns of state incorporation of peripheral territories is the degree to which elites in
the incorporated peripheries had mobility prospects in the political center. To the extent that
they were treated as equal to the elites in the center – and Ukraine is the exemplar – the mode of
incorporation was  what I refer to as most favored lord. To the extent that mobility prospects were
virtually blocked – and Kazakhstan is the exemplar in this study – the mode of incorporation was
“colonial.” To the extent that mobility prospects were partially blocked but rather rapid within
the republic – Estonia and Latvia fit this picture – I shall refer to this as “integral”
incorporation. These macro patterns help us understand the setup of the linguistic tipping games
that were unleashed with the breakup of the Soviet Union.

The micro data presented in this book suggest two fundamental trends concerning identity shift in
the republics of the former Soviet Union. The principal finding (reached in Chapter 9) concerns the
prospects for assimilation by Russians into the national cultures of the states in which they were
beached. I shall show why the prospects for Russians’ assimilation into the titular society
(despite cultural distance and anti-Russian policies and practices by the titulars) is greater in
the Baltic states than in Ukraine or Kazakhstan. Furthermore, I shall show that the prospects for
Russians’ assimilation are greater in Latvia than in Estonia, despite the higher expected economic
returns for learning Estonian. From both ethnographic and survey data, I shall also show why
Russians’ assimilation is more problematic in culturally proximate Ukraine than in the culturally
distant Baltic states.

A secondary movement in the former Union republics is acting alongside and against the pressure to
shift from Russian to titular (elaborated in Chapter 10). Most of those people who fall outside (by
virtue of language) the nationalizing projects of the republican governments in the states of the
near abroad have begun to form (or be formed into) a single identity group. They have begun to see
themselves – in conglomerate terms – as a “Russian-speaking population.” A major finding of this
book is that the development of such a conglomerate identity – although its membership is quite
different in the different republics – is the principal countertrend to assimilation.

With these two trends identified – that of assimilation and that of the creation of conglomerate
identity – I shall (in Part 5) analyze the implications of these findings to address two
outstanding questions on the post-Soviet agenda. First is the question whether violent conflict is
likely between the Russian inhabitants and the titular nationalities of the nationalizing states.
Some analysts, such as my collaborator Marika Kirch, rely on Samuel Huntington’s notion of a “clash
of civilizations” to suggest that a Russian/titular divide is inevitable, no matter how fluid the
actual cultural scene. My response is that a micro theory of identity shift, such as the tipping
game, reveals conflicts within civilizations which diminish the cogency of a model that focuses
principally on claims to broad cultural difference as the fountainhead of post-Cold War conflict.

Other analysts, relying on models of interstate conflict, study the issue of ethnic conflict as if
it were a game between a team of Russians and a team of titulars. Ignoring civilizational divides,
these analysts portray ethnic conflict as purely an issue of the security dilemma. My response
(provided in Chapter 12) is that there is a crucial difference between internationality and
interstate conflict. In the former, the boundaries of group membership are always subject to
redefinition. Political leaders of ethnic groups must constantly worry about defections from their
own group, as people move into other identity groups. This situation, I argue, puts great pressure
on self appointed representatives of nationality groups to rely on coercion to assure group
solidarity. The sources of a significant part of interethnic violence, I contend, are to be found
in the intragroup politics of internal policing and boundary protection. The politics of identity
itself-of cultural elites seeking to protect the boundaries of the groups they purport to
represent-creates even more incentives for violence than the tensions between identity groups do.

Second, there is the debate about whether the “nationalizing states”^53 of the former Soviet Union
will become nation-states (on the order of France, Germany, and the United States) or retain their
multinational character (on the model of Belgium, Switzerland, and the Soviet Union). Elsewhere I
have shown that both international conditions and the nature of state-building itself in the
present era make the construction of nation-states quite unlikely today, even if there are
significant pressures for assimilation of minorities.^54 In accord with this perspective, I shall
show (in Chapter 13) the considerable constraints that leaders in the republics of the former
Soviet Union face in fashioning nation-states. Yet, in contrast to. my projections in Language
Repertoires, I show how Soviet rule helped to undermine those constraints, leaving the road open
for successful nationalizing projects in the Baltics and Kazakhstan (but not in Ukraine).

I do not return to the theses concerning violence and the nation-state until Part 5, in the final
two chapters of this book. In the remaining two chapters of Part I, I provide a macrohistorical
framework for understanding the nationality situation that Russians faced as the republics they
lived in consolidated their nationalizing programs. In Part 2 I then provide an extensive
ethnographic description of the identity scene among Russians in the former Union republics after
the collapse of the Soviet Union. These chapters provide the flesh and blood that past renditions
of the rather skeletal tipping game lacked. In Part 3 I introduce survey data and a sociolinguistic
experiment to probe further the question whether Russians can become titulars. In Part 4 I address
the question whether Russians in diaspora may develop a new conglomerate identity, neither titular
nor “Russian.” I ask repeatedly throughout: Who were the Russians and what are they becoming? The
answer to this simple question – with data from Kazakhstan, Estonia, Latvia, and Ukraine – consumes
my attention for the bulk of this book and helps provide a coherent yet novel set of answers to the
questions concerning what type of state and what sort of ethnic violence we can expect in the
post-Soviet world.


Notes

^1 Marika Kirch, ed., Changing Identities in Estonia: Sociological Facts and Commentaries (Tallinn
Estonian Science Foundation, 1994), p. 12.

^2 The intergenerational aspect of conversion is developed in the classic work of A. D. Nock,
Conversion (London: Oxford University Press, 1933).^

^3 Joseph Stalin, Marxism and the National Question (New York: International Publishers, 1942).
Stalin was an assimilated Georgian whose published views and political programs denied the
possibility of nationality shift. See also Iu. I. Semenov, Sekrety klio: Szhatoe vvedenie v
filosofiiu istorii (Moskow: Moskovskii fiziki-mekhanicheskii institut, 1996), p. 60.^

^4 See Vladimir Shlapentokh et al., The New Russian Diaspora (Armonk, N.Y: M. E. Sharpe, 1994), and
Paul Kolstoe, Russians in the Former Soviet Republics (Bloomington: Indiana University Press,
1995). But also see Pål Kolstø, “The New Russian Diaspora – An Identity of Its Own?” Ethnic and
Racial Studies 19, no. 3 (1996): 609-39, for an open-ended approach to Russian identity shift. This
paper helped inspire the approach I take in this book.   ^

^5 Anatoly M. Khazanov, After the USSR (Madison: University of Wisconsin Press, 1995), pp. 15, 7.^

^6 Robert J. Kaiser, The Geography of Nationalism in Russia and the USSR (Princeton: Princeton
University Press, 1994), p. 372.^

^7 For this view, see Edward Shils “Primordial, Personal, Sacred, and Civil Ties; British Journal
of Sociology 8 (June 1957): 130-45; and Clifford Geertz, “The Integrative Revolution;” in his
Interpretation of Cultures (New York: Basic Books, 1973), chap. 7.

^8 Rom Harré, Personal Being: A Theory for Individual Psychology (Cambridge: Harvard University
Press, 1984). The quotation is from John D. Greenwood, "A Sense of Identity: Prolegomena to a
Social Theory of Personal Identity,” Journal for the Theory of Social Behavior 24, no.1 (1994): 28.

^9 See Tracy B. Strong, ed., The Self and the Political Order (New York: New York University Press,
1992), for a series of essays that cumulatively lead to a social history of identity. The reference
to Whitman is from George Kateb’s essay in that volume, “Walt Whitman and the Culture of
Democracy,” pp. 208-09.

^10 George P Knight et al., “Family Socialization and Mexican American Identity and Behavior; in
Martha E. Bernal and George P. Knight, eds., Ethnic Identity: Formation and Transmission among
Hispanics and Other Minorities (Albany: SUNY Press, 1993), p. 123.

^11 George A. De Vos, “A Psychocultural Approach to Ethnic Interaction in Contemporary Research,”
in Bernal and Knight, Ethnic Identity Formation, pp. 235-68. Zelig is the character in Woody
Allen’s film of the same name, who, chameleon-like, can change his personality and physiognomy to
fit unobtrusively in different social environments.

^12 Karl W Deutsch et al., Political Community and the North Atlantic Area (Princeton: Princeton
University Press, 1957). For a sympathetic critique of the “caught in the net” determinism of this
model, see William Foltz, “Modernization and Nation Building: The Social Mobilization Model
Reconsidered,” in Richard L. Merritt and Bruce M. Russett, eds., From National Development to
Global Community (Boston: Allen & Unwin, 1981), pp. 25-45.

^13 Dusko Sekulic, Garth Massey, and Randy Hodson, “Who Were the Yugoslavs? Failed Sources of a
Common Identity in the Former Yugoslavia; American Sociological Review 59 (February 1994): 83-97.

^14 Ernst R. Haas, The Uniting of Europe (Stanford: Stanford University Press, 1958); David Laitin,
Hegemony and Culture (Chicago: University of Chicago Press, 1986); Melissa Nobles, “Responding with
Good Sense: The Politics of Race and Censuses in Contemporary Brazil” (Ph.D. diss., Yale University
1995).

^15 Yen Espiritu, Asian American Panethnicity (Philadelphia: Temple University Press, 1992); Joane
Nagel, “Constructing Ethnicity: Creating and Recreating Ethnic Identity and Culture;”  Social
Problems 41, no. 1 (1994): 152-76; Ran Greenstein, “Racial Formation: Towards a Comparative Study
of Collective Identities in  South Africa and the United States,” Social Dynamics 19, no. 2 (1993):
1-29.

^16 Espiritu, Asian American Panethnicity.

^17 Greenstein, “Racial Formation”; Ian Buruma, The Wages of Guilt: Memories of War in Germany and
Japan (New York: Farrar, Straus & Giroux, 1994).

^18 The landmark work in this tradition is Harrison White, Identity and Control: A Structural
Theory of Social Action (Princeton: Princeton University Press, 1992). Also see Craig Calhoun, “The
Problem of Identity in Collective Action,” in Joan Huber ed., Macro-Micro Linkages in Sociology
(Beverly Hills, Calif.: Sage, 1991), pp. 51-75.

^19 Bhikhu Parekh, “Discourses on National Identity,” Political Studies 42 (1994): 503-4.

^20 This section owes a great deal to long discussions with my colleague James D. Fearon. He shared
with me his lecture notes on the concept of identity for his course “Nationalism and International
Conflict”, and the formulations herein owe much to his subtle analysis of the issue.

^21 My teacher Hanna Pitkin introduced me to the possibilities of ordinary language analysis.
Frederic Schaffer, another student of Pitkin’s, gave me the idea of using the Lexis-Nexis database
for such analysis. Schaffer also directed me away from some misguided distinctions I made in an
earlier draft of this chapter. In the Lexis-Nexis database for 1994-95, I retrieved 904 stories
that had at least seven uses of the term “identity” or “identities.” This section is based on the
first 100 of these stories (all from June-August 1995).

^22 That there is currently an articulate social movement in the United States to move gender from
the primordial to the constructivist discourse realm is confirmation that the distinction proposed
here is indeed part of popular understanding. That the stakes are high is seen in an article on the
front page of the March 14, 1997, New York Times, “Sexual Identity Not Pliable After All, Report
Says,” by Natalie Angier. Angier reports the outcome of a famous case in which a child born with
superficially ambiguous genitalia was raised as a girl. It had been expected that socialization as
a girl would in fact produce a girl, but the child had always preferred “boyish” activities, and as
an adult opted for a masculine identity. That gender perhaps really is primordial is first-page
news for this politically correct newspaper.

^23 Plausibility is a key requirement in conventional understandings of group attachment. The
discovery in Australia that two whites, one a novelist and one a painter had taken Aboriginal
pseudonyms to increase the market for their artistic works was met with outrage. That these
assumptions of Aboriginal identity were implausible (once the truth was revealed) subjected the
claimants to the charge of “fabrication” and the threat of legal action. Conventional opinion had
it that they had no “right” to assign themselves Aboriginal identities. See Clyde H. Farnsworth,
“Two Exposed Artists, Neither Aboriginal (Nor Original) After All,” New York Times, April 2, 1997.

^24 Kwame Anthony Appiah, In My Father's House (New York: Oxford University Press, 1992); Gayatri
Spivak, In Other Worlds (New York: Routledge, 1987), p. 205; Lisa Lowe, “Heterogeneity, Hybridity,
Multiplicity: Marking Asian American Differences,” Diaspora: A Journal of Transnational studies I,
no. I (1991): 24-44.

^25 David Laitin, “The Game Theory of Language Regimes,” International Political Science Review 14,
no.3 (1993): 228.

^26 On the references to Freud, see Erik H. Erikson. Identity: Youth and Crisis. (New York: Norton,
1968), pp. 20-21. On the Reformation, see Erikson, Young Man Luther (New York: Norton, 1958).

^27 Erikson, Identity, pp. 19, 22-23.

^28 In my Hegemony and Culture (Chicago: University of Chicago Press, 1986) I discuss the
primordialist approach associated with Clifford Geertz and the instrumentalist critique associated
with Abner Cohen. See Geertz, “The Integrative Revolution,” in his Interpretation of Cultures, and
Cohen, Two-Dimensional Man (Berkeley: University of California Press, 1974).

^29 Ernesto Laclau, ed., “Introduction,” in The Making of Political Identities (London: Verso,
1994).

^30  Erikson, Identity, p. 36.

^31 Thomas Schelling, Micromotives and Macrobehavior (New York: Norton, 1978).

^32 Timur Kuran, “Now out of Never: The Role of Surprise in the East European Revolution of 1989,”
World Politics 44, no.1 (1991); Susan Lohmann, “Dynamics of Informational Cascades,” World Politics
47, no. 1 (1994): 42-101.

^33 For more on my preoccupation in this book with language, which I acknowledge to be only one
element of a person’s complex social identity, please see the appendix.

^34This is a principal theme of Toivo Raun’s Estonia and the Estonians (Stanford: Hoover
Institution Press, 1987). He cites the nineteenth-century poem of Jaak Peterson – “May not the
language of this land / On winds of song and / Rising m the heavens / Seek eternity?” – to show how
commitment to a Finno-Ugric identity could overcome historically powerful Germanic and Russians
efforts to stamp it out.

^35 Thomas Schelling. The Strategy of Conflict. (Cambridge: Harvard University Press, 1960).

^36 I owe this last point to Martin Riesebrodt.

^37 Roger Petersen, in “Rationality, Ethnicity, and Military Enlistment,” Social Science
Information 28, no. 3 (1989): 563/98, relied on the tipping game to do just this, using the case of
Japanese-Americans’ enlistment behavior during World War I.

^38 I owe this point to Prasenjit Duara.

^39 Ann Swidler, “Culture in Action: Symbols and Strategies,” American Sociological Review 51
(April 1986): 273-86. I develop more fully the notion that uncertainty breeds rationality in
“Identity Choice under Conditions of Uncertainty” (paper presented at the faculty seminar of the
Department of Political Science, University of Michigan, Ann Arbor, November 1996).

^40 Raun, Estonia and the Estonians p. 210, cites three documents from the early 1980s in which
Estonian intellectuals expressed “the fear . . .that the Estonian language and culture were in
danger of losing their leading role in Estonia as Russian increasingly became the language of
administration and was emphasized in education.”

^41 Ernest Gellner, Nations and Nationalism (Ithaca: Cornell University Press, 1983). In Chapter 9,
I elaborate on, criticize, and extend Gellner’s theory.

^42 Ernest Gellner, Thought and Change (Chicago: University of Chicago Press, 1964).

^43 Charles Ferguson, “Diglossia,” Word, 15(1959): 325-40.

^44 In “Laitin, The Game Theory of Language Regimes; International Political Science Review 14,
no.3 (1993): 233, I call this the “private subversion of a public good.” This strategy is similar
to that caricatured by Will Rogers when he said that Oklahomans would continue to vote for
prohibition as long as they could stagger to the polls.

^45 Rafael compares nicely across categories. His comparison of Filipinos living under Japanese
colonial rule with diasporic populations is extremely productive, but his approach does not lead to
a genuine comparative analysis. He cannot offer a general calculus, for example, on the relative
probability of assimilation in different settings. See Vicente L. Rafael, “Anticipating Nationhood:
Collaboration and Rumor in the Japanese Occupation of Manila,” Diaspora: A Journal of Transnational
Studies I, no. I (199I): 67-82.

^46 The most sophisticated in this genre, where “integration,” “acculturation,” and “assimilation,”
are shown to be distinct, is that of R. A. Schermerhorn, Comparative Ethnic Relations (Chicago:
University of Chicago Press, 1970). See also, John W Berry, “Ethnic Identity in Plural Societies,”
in Bernal and Knight, Ethnic Identity Formation, pp. 271-96.

^47 David Laitin, “Marginality: A Microperspective,” Rationality and Society 7, no. 1 (1995): 35.

^48 David Laitin, “Language Games,” Comparative Politics 20 (1988): 289-302.

^49 This is the point of Kachig Tölölyan, “The Nation-State and Its Others: In Lieu of a Preface,”
Diaspora: A Journal of Transnational Studies 1, no.1 (1991): 4, writing as the founding editor of
the journal.

^50 William Safran, “Diasporas in Modern Societies: Myths of Homeland and Return,”  A Diaspora: A
Journal of Transnational Studies 1, no. 1 (1991): 83-84.

^51 This point is nicely developed by Hamid Naficy. “The Poetics and Practice of Iranian Nostalgia
in Exile,” Diaspora: A Journal of Transnational Studies I, no.3 (1991): 285-302.

^52 Lowe, “Heterogeneity, Hybridity, Multiplicity,” p. 27. On the growing political weakness of the
Cuban diaspora in the United States as the Cubans become part of a “fluid diaspora of twenty-five
million” Latinos, or Hispanics, see Roman de la Campa, “The Latino Diaspora in the United States:
Sojourns from a Cuban Past,” Public Culture 6 (1994): 293-317.

^53 For a sensitive definition of what “nationalizing state” means in practice, see Rogers
Brubaker, “National Minorities, Nationalizing States, and External National Homelands in the New
Europe,” Daedalus 124, no. 2 (1995): 107-32.

^54 David D. Laitin, Language Repertoires and State Construction in Africa (New York: Cambridge
University  Press 1992).


In: Laitin, D. Identity in Formation: The Russian-Speaking Populations in the Near Abroad. Cornell
University Press, 1998, pp. 3-36.