In the 1990s, the Julian (Giulian) Region that includes the cities of Trieste and Gorizia and the Istrian peninsula attracted the renewed attention of scholars for its qualities as a space in which both cosmopolitanism and nationalist polarization had flourished in the late Habsburg era. Although a healthy debate exists as to the degree to which forms of interethnic tolerance remain a feature of everyday life in these areas, most historians agree that this region underwent a series of increasing nationalizations (Italian, Slovene, Croatian, and Yugoslav) that began in the nineteenth century and culminated in the partisan anti-Fascist uprising and massive demographic shifts after World War II, as the majority of Istria's “Italian” population (together with a significant number of individuals self-identifying as Slovene or Croat) migrated from the territory that passed from Italian to Yugoslav control.Footnote ¹ The historiography of the modern era in the Julian Region has thus confirmed many of the assumptions made by nationalist activists along this classic “language frontier” about the inevitability, exclusivity, and irreversibility of ethnonational identifications.

This historical tradition also proves part and parcel of a teleological tendency to write Balkan and East European history “in terms of escalating nationalist conflict that culminated in the violence of ethnic cleansing.”Footnote ² Admittedly, a distinct body of scholarship has focused on groups whose explicit hostility to nationalist mobilization led them to seek socialist or internationalist solutions, although even here scholars have emphasized the ways in which local and regional forms of socialism frequently spoke nationalist idioms.Footnote ³ The phenomenon of indifference to the nation that Judson and Zahra,Footnote ⁴ among others, have documented for other frontier zones of the former Habsburg Empire has warranted relatively little conceptual or empirical elaboration for the Julian Region.

This article examines indifference to the nation as expressed in the form of national indeterminacy during the supposedly “final” act in the nationalization drama: the period between 1943 and 1955 when an estimated 200,000–250,000 persons left the Istrian peninsula and surrounding areas in what were highly visible and politicized population movements. In doing so, I focus on local responses to the efforts by states and other actors to, as James Scott puts it, “make a society legible”Footnote ⁵ through the imposition of clear-cut categorizations of identity, as well as the employment of key technical instruments (such as censuses, maps, property registers, and fixed patronyms) for measuring and enforcing those identifications.Footnote ⁶ In particular, I explore one of the most dramatic situations in which state authorities sought “that people be understandable through the categories in which they fall”:Footnote ⁷ the option process laid out by Article 19 of the 1947 Peace Treaty with Italy, which awarded southern Istria (part of Italy from 1920 on) to Yugoslavia.

According to Article 19, all “Italians” resident in the ceded territories (which included not just southern Istria, but also the Dodecanese islands and parts of the Valle d'Aosta) on or before 10 June 1940 had the legal right (though by no means the obligation) to choose Italian citizenship. The principal requirements of Italian-ness in the case of the option were Italian as the lingua d'uso (language of customary use) and domicilio (domicile) in Italy on the determined date.Footnote ⁸ Opting for Italian citizenship required moving to the territorially reconfigured Italian state. Writing at the time of events, legal scholar Josef Kunz declared the option process “theoretically correct and apt to avoid difficulties.”Footnote ⁹ Determining Italian-ness on the ground proved no easy feat, however, either for Italian and Yugoslav governments deciding individual option cases or for the international organizations (United Nations Relief and Rehabilitation Administration or UNRRA and the International Refugee Organization or IRO) charged with determining whether individuals coming from the formerly Italian parts of Istria and Venezia Giulia were eligible for refugee relief. The national indeterminacy of some of these refugees—populations that had become increasingly illegible within the bureaucratic apparatus of states—made them quite visible to the staff of the international agencies trying to figure out who counted as an individual displaced outside his or her “home” country.

As internal debates among IRO staff about refugees from the Julian Region reveal, in practice individuals' linguistic capacities and affiliations often defied the clear-cut bureaucratic categories laid out by peace treaties and citizenship laws. Despite over a half-century of intensive propaganda efforts to turn Istrian peasants, fishermen, and town dwellers into Italians or Slovenes or Croats, some individuals continued to express and understand their identity in highly localized or “hybrid” ways. Others strategically used national identity in an effort to adapt to the prerequisites for eligibility for aid from either the Italian state or the international relief organizations. Over time, IRO personnel came to recognize officially the national indeterminacy of many such refugees from Istria and the larger region of the Julian March, creating the category of “Undetermined Venezia Giulians” and thereby rendering some of history's “illegibles” “legible.”

This article represents a first step in recuperating the history of these “undetermined” individuals whose motivations were often misunderstood or ignored by state officials and historians whose conceptual categories did not allow for such fluidity. Whereas Zahra contends that, “[u]nlike nationalism, national indifference has not left much of a paper trail,”Footnote ¹⁰ for cases like those of the indeterminate Julian refugees explored here, the documentation of international agencies like the IRO proves fairly extensive, suggesting new sources for historians interested in testing the usefulness of “national indifference” as a category of analysis.

Language and Nationalization in Istria: A Brief History

Many of the bureaucratic difficulties in sorting out “Italians” from (Yugo) “Slavs” according to the terms of the 1947 citizenship options stemmed from the Peace Treaty's lack of definition of “customary language.” A 1947 note from the Italian Ministry of Foreign Affairs to the Border Commission recognized the attendant difficulties, suggesting that “lingua d'uso” would seem to imply “the language normally adopted by a person in familiar and most common contexts.”Footnote ¹¹ Such language, however, remained a “question of fact and proof.”Footnote ¹² Aware of widespread bilingualism in the border region, the Italian Ministry of the Interior counseled in a 1950 memo that in practice “lingua d'uso” really should mean “lingua materna,” i.e., “the ‘native language,’ the ‘language of the patria,’ that is, the language of the nation to which one belongs.”Footnote ¹³ Italian observers worried that ethnic Slavs were using the option's language clause (sometimes by making false statements about their lingua d'uso) to “infiltrate” the border area around Gorizia and Trieste.Footnote ¹⁴ Here, opportunistic border dwellers figured as both duplicitous and potentially destabilizing. From the other side, the Yugoslav government frequently refused requests to opt, in part to prevent economic disaster should most of the region's skilled workforce leave.Footnote ¹⁵

The contradictions built into the option's model of identity reflected the earlier struggles by nationalist activists in the Habsburg Empire and, later, the state itself to use language as a key indicator of an exclusive national identification. Although the German liberals who framed the Austrian Constitution of 1867 did not intend the clause guaranteeing equality of language use to translate into demands for collective rather than individual rights, the end result demonstrated otherwise. As Judson notes, “Instead, every nationalist movement in post-1867 Austria organized its political agenda primarily around achieving linguistic parity for itself in education, in provincial administration, and in the judiciary.”Footnote ¹⁶

In Istria, the pattern of such language activism more or less followed general trends elsewhere in the empire, with the establishment of cultural associations preceding political mobilization and the later creation of ethnonationally defined parties. Some historians locate the first signs of “national” organizing in Istria in the 1840s, although more explicit political conflict dates to the 1870s. This reflects not only the response of nationalist entrepreneurs to the possibilities opened up by the 1867 Constitution, but also the rise of Italian irredentism with the creation of a unified Italy, which in turn further stimulated movements promoting “Croatian national rebirth” and “Slovene national awakening” already in progress.

Vanni D'Alessio's monograph Il cuore conteso (2003) takes seriously the insights of scholars of nationalism who have documented the assiduous process required to construct national identities.Footnote ¹⁷ D'Alessio focuses on contests between Italian and Slavic/Croatian activists in Pazin/Pisino, a town whose position as the “contested heart” of the Istrian peninsula he renders metaphor for the larger processes of political mobilization that ultimately fractured the multiethnic coexistence of Istria and produced a bifurcated society. “Pisino, or Pazin, respectively, became a stronghold of both Italian and Croatian identity, and was symbolically enacted both as an Italian ‘bulwark’ surrounded by a ‘Slavic sea’ and as the ‘heart of Croatian Istria.’”Footnote ¹⁸

In his periodization of nationalist mobilization in Pazin, D'Alessio views 1880–1915 as a time in which “the local population was not only exposed to political and ideological pressure during the election periods, but also forced to choose between evident markers of alternative forms of social interaction and to show an open belonging to either the Italian or the Croatian group. Social interaction and public life was indeed split.”Footnote ¹⁹ Although D'Alessio recognizes the complexities of lived identity and the frequent divergence between the public and private spheres, he ultimately concludes that “once a political fight had started and the bi-ethnic community had two separate networks of social interactions choices started to be unavoidable.”Footnote ²⁰ Such a conclusion neither admits many possibilities for alternative paths or choices, nor does it recognize that the nationally committed populations in Istria and the larger Julian Region were not merely two (a “bi-ethnic community”) but multiple, and included Germans.Footnote ²¹ D'Alessio further asserts that ethnic antagonism arose earlier than in other parts of Istria and proved more pronounced in Pazin, where political and social borders between groups “became rather thick.”Footnote ²² This image runs counter to Judson's contention that “most inhabitants of such [border] regions rarely viewed themselves specifically as ‘frontier people’ or their regions as frontiers between nations.”Footnote ²³

With his emphasis on the growing rigidity of social, linguistic, and political borders in nineteenth-century Pazin and Istria, D'Alessio voices a common view of Istria as an increasingly divided society in the nineteenth century. On this point, Croatian historiography has also generally been in line with Italian-language scholarship about the national polarization of Istria during the late Habsburg era.Footnote ²⁴ Might it be possible that the old political border between Venetian and Habsburg Istria, a line that ran near to Pazin, indeed made for a different situation—with fewer linguistic and national “amphibians” in Istria than in the Habsburg regions that scholars like Judson and Zahra examine?

Certainly, the distinct pre-1797 imperial heritages of Istria made for diverse patterns of linguistic concentration and interaction across the peninsula. Along Istria's western (Venetian) coast existed numerous towns that were relative “islands” (a metaphor that nationalists would later adopt to describe the threat posed by linguistic isolation) of speakers of Italic/Latinate variants (istriotto, istro-veneto). “Italian” represented the language of the political and economically dominant class identified with urban life, as well as a maritime lingua franca. This does not mean that speakers of Slavic variants did not exist in western Istria, but rather that there prevailed a situation of diglossia, that is, a pronounced hierarchy of languages that made for distinct patterns of linguistic competence. Elites who spoke some variant of the language of high culture, “Italian,” could choose to be monolingual (although not all did), whereas the populations that lived outside the city walls frequently spoke Slavic variants but also needed to communicate in Italian, the language of administration, trade, and social mobility.

In the interior and along the eastern coast of the peninsula (the old Habsburg territories), the development in the nineteenth century of a Croatian sphere of public culture directed by clergy and rural bourgeoisie (lawyers) helped mitigate assimilation pressures toward Italianization. Nevertheless, along the eastern coasts, some elites also spoke Italian and Venetian dialects.Footnote ²⁵ Most of the emerging Croatian national elites in Istria originated from the Kvarner Islands and Kastav/Castua. Although just a “little town on the eastern margin of the Istrian province,” after 1880 Kastav “became the centre of new local Croatian politics involving the masses through cultural associations and a network of financial supports for the peasants.”Footnote ²⁶ The Slovene “national awakening” in Istria took many of its cues from developments in Ljubljana and nearby Trieste, although some Croatian and Slovene activists also collaborated closely in Istria and formed a joint political party.Footnote ²⁷

Recognizing the reality of varied settlement and linguistic practices in the Istrian peninsula, however, does not mean that all or even most inhabitants of Istria perceived or worried much about symbolic borders—linguistic or national or otherwise. As scholars we must tread carefully, given that “ethnographic” patterns of language use (as well as other aspects of popular culture) offer the sort of “evidence” that nationalists frequently took to advance their own causes.Footnote ²⁸ The rural inhabitants of the interior, particularly the zone known as the Buiese, moved easily between linguistic variants, a reality that persisted at least as late as 1955, as captured by Fulvio Tomizza in his semi-autobiographical novel, Materada. Tomizza's protagonist Francesco Koslovic inhabits an impoverished rural world in which distinctions between Italians and Croats (and corresponding cultural spheres) are not at all clear. Tomizza, of course, enjoyed the artistic freedom opened up by literature to paint a rich portrait of linguistic and cultural intermixture.Footnote ²⁹ Historians of the region focusing on documents such as censuses and official statistics have not typically enjoyed this same luxury, with the consequence that such categorical and categorizing forms of evidence “have notoriously served to obscure bilingualism and national ambiguity in East Central Europe.”Footnote ³⁰ Censuses and population maps have also continued to fuel debate over the “real” ethnic makeup of territories like the Julian Region at key moments in time.Footnote ³¹

Although Habsburg administrators' aims in measuring linguistic identifications proved quite different from those of nationalists seeking accommodations or even independence from the state, the state's instruments to render peoples legible clearly proved useful to national projects. Kappus's comments on Istria hold for the empire more generally: “Not only nationalists, but the Empire's administrators as well classified the Istrian peoples in increasingly homogenized ‘national’ groups and it is, in fact, hard to determine if the national movements had informed the administrative categories or vice versa.”Footnote ³² Although state officials began to implement censuses in 1754 in selected border regions, systematic attempts to measure linguistic practice throughout the empire only came about with the 1846 and 1857 ethnographic surveys carried out by researchers at the Austrian Institute of Statistics.Footnote ³³ These surveys, focused on what state ethnographer Karl von Czoernig deemed “ethnographic statistics” or “statistical ethnography,” served in the construction of linguistic and ethnographic maps of the empire.

The methods for gauging language use and the very categories of language use employed in these surveys differed considerably from those used in the four censuses conducted between 1880 and 1910, as well as the 1921 Italian census of Venezia Giulia. Czoernig did not examine individual language use but, rather, predominant language use in each respective administrative region, data that chief administrators of the Länder supplied.Footnote ³⁴ Czoernig supplemented this data with “ethnographic history” that aimed to go beyond what he saw as the malleability of language practices to deeper truths about the migration of various tribes into every corner of the empire.

Czoernig's methods and categories admitted greater complexity of identity (linguistic and otherwise) than subsequent census forms, instituted in 1880, which took as their statistical unit of analysis the individual and that measured the “language of customary use.” The adoption of this problematic category, one that assumed the language of customary use to be singular, reflected larger transformations in the new science of statistics. Although a recommendation of the 1872 International Statistical Congress held in St. Petersburg urged census makers to adopt the category of “mother tongue” as the critical measure of identity, census makers in Austria did not heed this proposal and instead instituted the category of Umgangsprache (language of daily use) in the censuses of 1880, 1890, 1900, and 1910.Footnote ³⁵ Determining this “language of use” proved no easy task. As D'Alessio rightly points out for Istria, “Theoretically it was not clear whether language of use meant language of interpersonal relationships, working relationships, or even the one used inside the family. These languages were often not the same. The high number of bilinguals could offer different data and possible interpretations.”Footnote ³⁶

The Italian liberal regime that governed Istria briefly (1920–22) before the Fascist ascent to power retained the category of “language of customary use” (lingua d'uso) in the contested census of 1921. Subsequent censuses in 1931 and 1933 dispensed of the language category, a reflection of the dramatic shift in language politics with the Fascist regime's forced Italianization campaign.Footnote ³⁷ The regime prohibited the use of Croatian and Slovene in public life, including schools. Nonetheless, the regime continued to equate linguistic identity with a putative singular nationality, even as Fascists worried about a hidden world of Slavdom (and Socialism) circulating beneath the surface of an outwardly Italian public life. Although Fascism did not succeed in eradicating Slavic languages or identities in Istria or other areas of Venezia Giulia, it did render invisible an urban Croatian sphere previously centered in towns like Pazin/Pisino and Kastav/Castua. This forced the networks sustaining Croatian identity to retreat into the rural areas of Istria.Footnote ³⁸ Officials realized that their Italianization efforts had not succeeded when the results of a secret survey conducted in 1939 revealed that the numbers of “alloglotti” (speakers of languages other than Italian) had actually increased since the 1921 census.Footnote ³⁹

Most accounts of the Fascist era in Istria stress the ways in which the regime's efforts to stamp out Slavic languages and identities backfired, resulting in both mass emigration of Croatian and Slovene elites to Yugoslavia (and overseas) and the organization of armed resistance well before the outbreak of World War II. Although historians like Ivetic question a cherished Croatian narrative of the anti-Fascist struggle as constituting a “new Croatian [national] rebirth,”Footnote ⁴⁰ Ivetic does stress the widening gulf during the Italian period between an Italianized version of consumerist modernity in the towns and life in the nearby villages “which still lacked the essential modernisation of their customs.”Footnote ⁴¹ This economic and social contrast (which Ivetic largely reads as corresponding to an ethnolinguistic contrast), proved, in his words, “a crucial moment for the Croatian nation-building in Istria.”Footnote ⁴²

The violent executions carried out by partisans against Italians (Fascist and anti-Fascist) in 1943 and again at the war's end have appeared to many scholars as a final act in the nationalization of Istrian Slavs initiated in Habsburg times. Other observers, like Sandi Volk,Footnote ⁴³ have discerned a mirror process among the targeted population, as a sense of Italian identity coalesced (in some instances) or hardened in the face of persecution, marginalization, and displacement to Italy. More commonly, however, scholars treat Slavic and Italian identities and populations as already clearly delimited by the end of World War II. Between 1943 and 1955, runs a widespread narrative, Italians left the peninsula as a result of outright persecution and more subtle forms of intimidation (including Socialist confiscation of properties and forced participation in the local elections of 1950). Readings of these events as “ethnic cleansing” are increasingly prominent, particularly among Italian historians.Footnote ⁴⁴ Such an interpretation imputes nationalist motivations to the putative perpetrators of ethnic cleansing and implies that its victims belonged to a clearly defined and delimited ethnonational group. In my own fieldwork in the 1990s among Istrians who resettled in Trieste after World War II, this narrative proved pervasive, even when individuals' accounts of their motivations for leaving Istria complicated or contradicted the ethnic cleansing thesis. A powerful narrative of memory among these Istrian exiles or esuli, particularly those in Trieste, increasingly influences scholarly work on the Istrian past.Footnote ⁴⁵

Opting for Identity?

The techniques employed to render populations in the Julian Region legible correspondingly made illegible those persons whose understandings of identity did not fit the tight mapping of linguistic and national affiliation that figured prominently in the later Habsburg censuses, as well as instruments of statecraft like the 1947 Peace Treaty. Illegibility does not necessarily imply absolute invisibility, however, as the stridency of national activists suggests that the visible persistence of amphibious bilinguals and other insufficiently nationalized populations (such as those adhering to strong local or regional identities) generated both anxiety and resolve among nationalizers.Footnote ⁴⁷ As Joâo Biehl demonstrates for marginalized sufferers of HIV/AIDS in Brazil, the lack of bureaucratic recognition of such populations does not entail a mere absence but rather an active process of erasure (one that is always incomplete) through what he deems “techniques of invisibility.”Footnote ⁴⁸

Populations whose self-identifications did not match the categorical versions of identity laid out in the 1947 Peace Treaty—are you Italian or Yugoslav?—became a highly visible problem for agencies like the International Refugee Organization (IRO). How to deal with those individuals who fell out of the clear-cut categories of Italian or Yugoslav optant, as well as the clear-cut categories of recipient of aid from the Italian government (for “Italians”) and recipient of aid from IRO (for “Yugoslav” refugees)? Precisely because IRO spent a good amount of time worrying about how to classify refugees from Venezia Giulia, examining these debates illuminates the ways in which nationalization processes were far from complete in the region. Despite their visibility as a problem or concern for IRO staff, these “indeterminate” refugees have nonetheless been largely invisible in the historiography of the region and work on identity in the Julian “borderlands.” Likewise, although the option process has been studied in terms of its implementation by Italian and Yugoslav authorities, scholars have paid little attention to the ways in which international agencies like IRO attempted to deal with its consequences.Footnote ⁴⁹

The IRO came into being in 1948, during the same period in which residents of the areas of eastern Venezia Giulia, southern Istria, Rijeka/Fiume, and Zadar/Zara ceded to Yugoslavia by the 1947 Peace Treaty faced the choice of whether to opt for Italian citizenship (if they possessed the “right” to do so, according to the criteria of Article 19) or become, through choice or inaction, Yugoslav citizens. The “Law on Citizenship of Persons within the Territory annexed to the Federal Popular Republic of Yugoslavia on the basis of the Peace Treaty with Italy” restated the terms of Article 19:

To outsiders like those at IRO, the stress laid on customary language as a key marker of identity might have seemed odd, given that twenty years of Fascist Italianization policies in Venezia Giulia had further diffused fluency in Italian, long the commercial lingua franca of the region. The statesmen who negotiated the 1947 Peace Treaty with Italy, however, drew upon the history (outlined above) of using language as a measure of supposedly (exclusive) national identities. Many scholars have commented on the power of such state-imposed categories to constitute identities.Footnote ⁵¹ In his study of the population censuses in the Julian Region, for example, anthropologist Bjørn Thomassen emphasizes the declarative power of the census. “Taken as a ‘speech act,’” he argues, “national self-declarations are not only directed toward the state or the census collectors. As a technology of bio-power, its effects reside in the fact that the act of declaration eventually becomes directed toward the social community and toward the self. The knowledge of oneself becomes the truth of oneself.”Footnote ⁵² Yet the external ascription of identity categories—such as those laid out in the censuses of the post-World War II option—clearly does not always result in complete self-ascription.Footnote ⁵³ In her study of the Bohemian Lands, for instance, Tara Zahra demonstrates the persistence of ambiguous identities even in the face of forcible classification to transform “nationally indifferent Bohemians and Moravians into Czechs and Germans.”Footnote ⁵⁴

Historian Sandi Volk has argued that despite the Fascist regime's Italianization policies in Istria, at war's end there existed many individuals (at least in smaller centers like Gronjan/Grisignana) for whom “national identity was not yet definitely formed.” In Volk's interpretation, the mass migration occasioned by the option marked the moment in which those leaving “definitively chose also their national identity . . . the exodus was thus also a key moment of denationalization in which an imprecise number (in any case a minority) of Slovenes and Croats in Istria made the passage to Italianness.”Footnote ⁵⁵ It remains difficult, however, to prove that choosing Italian citizenship necessarily meant shedding one identity in order to embrace and internalize an Italian national identity—particularly for those who sought to emigrate abroad through the offices of IRO. As Pupo and Panjek contend, the option offered a kind of safety valve for general dissatisfaction with the Yugoslav regime in Istria, leading many of uncertain identity to declare themselves Italian. “More or less voluntarily, more or less uncertainly, many [optants] chose a nationality, others changed it, others yet declared one for the sake of convenience.”Footnote ⁵⁶

Into this complicated situation in Venezia Giulia came the staff at IRO, the successor to the United Nations Relief and Rehabilitation Administration (UNRRA), who labored to apply its criteria for eligibility to real-life situations like those on Italy's eastern border. IRO recognized greater numbers of persons as eligible Displaced Persons than had UNRRA.Footnote ⁵⁷ Stated a June 1947 memo:

Given that those individuals opting for Italy were said to be “Italian,” IRO considered these refugees ineligible for aid, since IRO's stated mission entailed aiding displaced persons outside of their home country. Previous citizens of Italian Istria, these optants for Italy were considered Italians who “remained” in Italy, despite the fact that Italy's border had moved and hence retaining Italian citizenship required moving with/to Italy. As a March 1949 “Memorandum on the Question of Refugees from Venezia-Giulia” put it:

IRO's Acting Director-General P. Jacobsen made explicit the questionable assumption that language proved a (fairly) reliable measure of “origins,” i.e., ethnonational identity. Recognizing that Istria was home to many different groups, including persons of “Austrian” and “Hungarian” background, Jacobsen contended that most of these groups would not be of Italian customary language. In his opinion, the Italian government had responsibility for all Italian speakers from the ceded territories. “[T]he problem of Italian speaking persons in Italy who have been Italian citizens only recently,” he wrote, “is at least as much a part of the Problem of the Italian population generally as the problem of the ‘Volksdeutsche’ is part of the problem of German populations.”Footnote ⁶⁰ In making such comments, Jacobsen ignored the opinions of Italy's IRO head, G.F. Mentz, who maintained in a letter to W. Hallam Tuck, Director-General of IRO, that comparison of the Venezia Giulia refugees to the Volksdeutsche was unwarranted, as the latter category “is a very particular and negative exclusion based on racial terms, [thus] to extend it to groups other than German would be very clear violation both of the letter and of the spirit of the IRO Constitution.”Footnote ⁶¹

In contrast to these “Italian” optants, IRO considered within its mandate those refugees from Venezia Giulia “whose customary language was Slovene [or another Slavic language], and who had lived on Yugoslav territory.”Footnote ⁶² The Preparatory Commission for the IRO Eligibility Office in Rome, in fact, made a key distinction between “Persons of customary Yugoslav language (Slovene, Croat, or Serb) and “Persons of customary Italian language,” with the former eligible for IRO assistance and the latter excluded. At that time, IRO acted as if the option's language criterion adequately mapped onto ethnonational identity, for persons of customary Yugoslav language “are to be considered as Yugoslav and cannot opt for Italian citizenship.”Footnote ⁶³ The Commission did recognize that some applicants were bilingual but still sought to identify a primary customary language; in these cases, the criteria to determine “customary language” included “house language, parents' language, family name, parish church, cultural and political associations, etc.”Footnote ⁶⁴ In practice, however, such bilingual applicants proved difficult to classify; by June 1949, IRO was reconsidering the applications of some 500 individuals—previously excluded from IRO's mandate—deemed bilingual or of Italian customary language. In May 1949 alone, IRO staff reported receiving 165 applications from refugees whose customary language was said to be “Istrian dialect” or who were bilingual.Footnote ⁶⁵

The reconsideration of these applications points to the difficulties that IRO soon ran into with its exclusion of so-called “Italian” refugees from Venezia Giulia, as well as with its general adherence to the linguistic criterion of identity. Displaced persons from the ceded territories included individuals who had not opted for Italian citizenship and therefore were considered de facto Yugoslav citizens but whose “customary language” appeared to be Italian. Initially, IRO's policy had been to exclude any “Italian-speakers,” even if they had not opted, as they were seen to be the responsibility of the Italian Government. IRO personnel soon recognized the problems with this policy.

In a “Report on Operations of the Eligibility Division in Italy Covering the 3 Months Period September – October – November 1948,” I. H. D. Whigham, Chief of the Eligibility Division, commented on the fact that many of these so-called Italians ruled ineligible were Italian only in terms of their language of daily use.

Having contended that many of these DPs were not of Italian “ethnic origin,” Whigham suggested that many likewise did not consider themselves Italian.

As a result of these discussions over eligibility, IRO changed its policies in early 1950 and began offering aid and assistance to this type of Venezia Giulian refugee, i.e., an individual who had not opted for Italian citizenship (and thus by law became a Yugoslav citizen), regardless of customary language.Footnote ⁶⁸ This decision also applied to individuals who had opted but whose option the Yugoslav government had not accepted. Some such individuals had provisional passports issued in Zagreb that had permitted them to cross into Italy; these provisional passports would prove to be a source of enduring controversy within IRO.Footnote ⁶⁹ IRO required these individuals to have “valid objections against return to Yugoslavia.” In addition, they could not be “firmly established in Italy,” the condition of which would negate the need for IRO assistance to emigrate.Footnote ⁷⁰

In revising IRO eligibility policies, then, the legal criterion of citizenship technically prevailed over that of ethnicity, though IRO staff continued among themselves to distinguish “Italian” from “Slavic” refugees. For statistical reporting on such refugees in camps or IRO Intake Centers, Alva Simpson, Chief of the Department of Health, Care and Maintenance, ordered that the nationality of such refugees should be put down as “Undetermined Venezia Giulia.”Footnote ⁷¹ Other IRO correspondence, however, reveals a system of symbols to indicate the ethnicity and customary language of such “undetermined” refugees.Footnote ⁷² IRO documentation evidences a broad range of positions concerning the ethnic provenance of these “undetermined” refugees and the nature of the territory in question. Some officials recognized the “ethnically” mixed character of Venezia Giulia and the noncoterminous nature of “customary language” and ethnic self-consciousness. In a November 1948 letter addressed to Sir Arthur Rucker, Deputy Director General of IRO, the Chief of the IRO Italy Mission, G. F. Mentz, maintained, “Venezia Giulia is a very mixed area, where not only Italians and Slavs are present, but many persons of various origins who are ethnically not Italian even if their customary language is Italian.” Mentz recommended, “Such persons, we feel, must be considered prima facie eligible for IRO assistance.”Footnote ⁷³ Here, then, some IRO members argued that language could serve neither as an adequate index of ethnic identity nor a criterion for determining eligibility.

The fairly nuanced views espoused by some IRO personnel, like Mentz, reflected a more widespread concern in IRO to avoid the racialized understandings of ethnicity that had brought about such suffering during World War II. At the same time, however, older vocabularies and thinking in which race blurred with ethnicity persisted. A report on eligibility in Austria carried out in December 1947, for example, made explicit how IRO staff used a term such as Volksdeutsche, which carried with it certain racial connotations. The report stated that Volksdeutsche indicated “German minorities or persons of German ethnic origin. It is officially a Nazi expression, but it is now used by the Allied Military Authorities as well as by the IRO in designating German minority groups.”Footnote ⁷⁴ The author continued, “As long as the term is clearly understood as referring to the German-speaking minorities, and not confused with racialist biological theories, there is no harm in its use. In practice, such theories lead to the absurdity of designating Americans of the eighth generation as Dutch, French, etc., because one of their ancestors came from Holland, France, etc., three centuries ago.”Footnote ⁷⁵

The author of this report further outlined the difficulties inherent to the concept of ethnicity:

This IRO staff member concluded, “In practice, what is meant by ethnic origin is a person's consciousness of belonging to a particular nationality group, or splinter group.”Footnote ⁷⁷

In light of this, then, at times in the Venezia Giulia debates, a seemingly constructivist view of identity (in today's terms) appeared to operate that recognized not just the importance of cultural affiliation (and its mutability), but also of ethnic self-identification. Just below the surface of this culturalist view, however, often lay an essentialist view that assimilation could change a person's language or culture but not the “ethnic” core of his or her being. A note on “Venezia Giulia” dating 26 January 1949 and sent to George Warren at the U.S. Department of State, for example, offers a view of identity in Venezia Giulia as shaped but not determined by language or culture:

I. H. D. Whigham, Chief Eligibility Officer, concurred with this opinion in a letter sent in the same month (21 January 1949). Here, again, we find recourse to a discussion of both “racial origins” and cultural affinity, neither of which is synonymous with an ethnic self-awareness rooted in political identity.

Whigham concluded, then, that:

Critical of culture as manipulable by authoritarian regimes, Whigham's response nonetheless raised the question: If one had to be careful not to give culture too much weight, what else constituted identity but presumed “racial origins” or “mixed origins”?

Other IRO personnel, such as Mentz, recognized the specificity of local and regional culture in Istria, even for those individuals who self-identified and were identified by outsiders as “Italians.” Mentz maintained that some inhabitants of Istria were both Italian-speaking and “predominantly Italian from the ethnical point of view” and yet “have never till now come out of Istria, and who, notwithstanding their relationship with the Italian race, have no connection in Italy and no possibility of resettlement in this country [Italy].”Footnote ⁸¹ With widespread unemployment in Italy and a pervasive view of Istrians as “foreigners,” continued Mentz, “it is quite unrealistic to accept as a principle that ‘all Italian speaking persons who have been Italian citizens are more likely to be easily established in Italy than elsewhere.”Footnote ⁸² In this opinion, Mentz argued against Acting Director-General of IRO P. Jacobsen. In a letter dated 5 November 1948, Jacobsen had insisted, “persons of Italian culture should not be considered to be within the mandate merely because they are indigenous Venezia Giulians with no connections in present day Italy.”Footnote ⁸³

The following year, Mentz restated his position:

Eventually, Mentz's opinion would prevail and IRO would offer aid to such refugees.

Other IRO staff members working on the ground in Italy likewise noted a local/regional identity that made even the Istrians identified as “Italian” distinct from other Italians in peninsular Italy. Arguing against the 1948 decision to consider Italian-speaking non-optants ineligible (a policy that was later reversed), Alva A. Simpson, Chief of the Department of Health, Care and Maintenance, echoed Mentz in her assessment:

Elsewhere, Simpson advised Intake Offices, “Equal consideration should be given to the feelings of local attachment, which inhabitants of Venezia Giulia undoubtedly possess, as to economic and other influences.”Footnote ⁸⁶ These debates came closest to acknowledging local and regional forms of identity that might remain indifferent to the stark choices of nationality.

To some IRO observers, however, the uneasy fit of Istrians or Venezia Giulians into categories of national identity spoke of something negative, unstable, shifty, and shifting, perhaps bordering on opportunism. The 1949 “Memorandum on the Question of Refugees from Venezia Giulia,” for example, read the indeterminacy of Venezia Giulians as indicative of their nature as a “pathetic border people who are dissatisfied with both sides and desperately want to emigrate overseas.”Footnote ⁸⁷ The unknown author of this memo portrayed these border populations as shifting their loyalties to suit their interests: “The people of Venezia Giulia and Istria, after having requested to be annexed to Italy in 1920 as Italian irredentists, are dissatisfied with Italy. Many of them are also opposed to Yugoslav rule, so that they regret the old easy times when they were Austro-Hungarians.”Footnote ⁸⁸

In his February 1949 memo detailing “Treatment of Refugees of Italian Customary Language now in Italy,” Director General Hallam Tuck exhorted field officers:

A case of a refugee from Venezia Giulia evaluated by the IRO Review Board in March 1949 demonstrates that IRO clearly used such presumed opportunism as a justification for declaring specific individuals ineligible for assistance.Footnote ⁹⁰ The petitioner, deemed a “former Italian form [sic] Istria (province Pola, foemerly [sic] Zone B or Venetia Giulia) whose ethnic origin may be Slav but hose [sic] customary language is Italian,” had not opted for Italy and thus became a de facto Yugoslav citizen. According to the IRO board, he had come to Trieste after the war and had obtained employment, together with an identity certificate and work permit (libretto di lavoro) from the Free Territory of Trieste. When questioned why he had not opted for Italy, “he answered that this was because he was awaiting the IRO decision on his eligibility. He went on to say that if [t]he had opted for Italy he could not have gone to America, and that he had heard that if a man is neither Italian nor Yugoslav, he may go to America at IRO expense.”

Here, illegibility in the eyes of the state meant legibility for the international agencies; indeterminacy of identity vis-à-vis the nation meant determinate rights as a refugee to be assisted by IRO. Having thus understood quite clearly that the refugee was by definition liminal, betwixt and between (neither Italian nor Yugoslav), this man lost his chances to emigrate as a result of his frank admission. Whether this admission was a sign of the man's actual indifference to nation or awareness of its strategic value (however unstrategic his actual declaration) proves unclear. IRO denied his request for assistance, classifying the petitioner as “not a genuine refugee but an emigrant who applies for IRO assistance for economical reasons.”Footnote ⁹¹

In evaluating individual cases for eligibility of Venezia Giulia refugees from 1950 on, IRO personnel also appear to have adopted Mentz's focus on local identity and autochthony as key markers of whether individuals were merely seeking to “exploit” IRO services or had legitimate reasons for not opting for Italy. One application noted that the petitioner, born in 1907 in Udine, had moved to Venezia Giulia in 1938 and remained there for only a year, though he formally maintained his residence there and opted in 1948, without receiving word from the Yugoslav authorities about the status of his option. IRO personnel R. L. Gesner and M. E. Wood determined that this petitioner remained outside the mandate of IRO. “Although petitioner has evidence of his residence in Venezia Giulia in 1940 [at the time prescribed for the option] he actually lived there for only one year; being ethnically Italian, within the country of his former habitual residence, he has a refugee status which is purely formal.”Footnote ⁹² Another petitioner found his request for IRO aid denied on the grounds that he was not firmly rooted in the region. Born in Brindisi in 1882, this man had moved to Zadar/Zara in 1930, from which he had fled in 1944. At the time of the IRO review, his option remained pending. The reviewers of this case concluded, “Although Petitioner is technically a Jugoslav citizen because his option for Italy is still pending, the Board does not consider him a genuine refugee because in his country of birth and since he is Italian by ethnic origin and natural affinity.”Footnote ⁹³ IRO staff deemed the applicant a native of the “'old provinces'“ who “does not show evidence of strong ties holding him to Venezia Giulia.”Footnote ⁹⁴

As these case decisions show, then, IRO staff explicitly distinguished between “indigenous” Venezian Giulians and newcomers. In a 1952 interview, R. L. Gesner, Chief Eligibility Officer for the Italian Office, made clear that IRO staff recognized “the native V.G. groups were those who had always lived there and their ancestors before them. These were different from the settlers in Italy before the war.” Gesner expressed admiration for the autochthonous populations of Venezia Giulia, whom he deemed “very industrious and ambitious.”Footnote ⁹⁵

IRO interviewers and personnel on the ground in Venezia Giulia thus acknowledged not just the legal formalities of the option/citizenship, as well as putative ethnicity, but also individuals’ degree of rootedness in the region. IRO staff charged with interviewing applicants for aid possessed considerable power in the actual decision-making process. A 1949 letter to the Director of the Fermo Camp in Italy noted the difficulties in classifying refugees from Venezia Giulia and the leeway accorded to interviewers in making the final determination. “This matter has been discussed with the Division of Eligibility as the problem of determining the proper origin of refugees should not be yours, but the interviewer's,” admonished the letter writer. “The definitions ‘Italian-speaking’ or ‘Slav-speaking’ used in Administrative Order No. 121.4 were written by the Division of Eligibility and were intended to correspond with the instructions given to the interviewers, i.e. to show the refugees on the CM/1s [forms] according to ‘customary language.’”Footnote ⁹⁶ Nonetheless, not all interviewers apparently devoted enough care to their judgments of the Venezian Giulian refugees. Joan Jamieson, in the Office of Statistics and Operational Records and author of the memo, regretted that “[n]ot all interviewers have taken care to classify Venezia Giulia refugees correctly, thereby creating considerable reporting difficulties.” Having sent out more precise instructions to staff conducting interviews in intake centers, Jamieson urged, “In the meantime, for the doubtful cases you now have, the ethnic origin of the refugees should be considered, rather than the spoken language.”Footnote ⁹⁷ In this instance, at least, a putatively knowable and classifiable “ethnicity” trumped customary language.

The power accorded to interviewers carried its own problems. Interviewers, chosen for their linguistic skills (among other qualities), could potentially alienate the applicants for assistance. IRO staff working in Treviso, for example, fielded complaints about an employee, Plamenac(h), an Intake Interviewer hired for his language abilities and knowledge of Venezia Giulia. The Association for Venezia Giulia and Zara protested to the Italian Ministry of Foreign Affairs, whose Liaison Office then contacted IRO. In the communication sent by the Ministry to Mentz at IRO, Baron Malfatti reported “the spirit of bitterness toward the Giulians and Dalmatians” and cited in particular the attitude of “a certain Plamenach, Slav from Belgrad” who reportedly considered refugees now opting in Italy as Yugoslav citizens.Footnote ⁹⁸ Members of the Association for Venezia Giulia and Zara called upon IRO to award the classification of “indefinite citizenship” to those refugees whose options had not been approved and to drop any pressure “for a declaration of Yugoslav citizenship as a condition for emigration to other countries.”Footnote ⁹⁹ Furthermore, the Association demanded the replacement of Plamenach with “someone more adaptable and tactful.”Footnote ¹⁰⁰ Plamenach “has been already observed for his nationalist and Slav attitude, contrary to the purely objective criteria by which the refugees should be screened.”Footnote ¹⁰¹

Regardless of whether Plamenac(h)'s accusers had any credibility, these incidents underline not just the political and practical challenges that IRO faced, but also the fact that refugees often proved fearful and suspicious. A belief that the international refugee assistance groups were helping Communist regimes, particularly in encouraging (sometimes forcible) repatriation back to the Soviet Union and Yugoslavia, proved widespread among refugees dealing with UNRRA, IRO's predecessor.Footnote ¹⁰² Refugees often did not know whom to trust either among their fellow refugees or UNRRA/IRO staff. IRO staff in Opicina determining eligibility for the refugee camps run by the Allied Military Government in Trieste soon learned, for example, “The eligibility interviewers should have a location which will allow a free-flow of conversation between the interviewers and refugee to obtain the fullest possible information without fear of being overheard by refugees waiting to be interviewed.”Footnote ¹⁰³ DPs resented or feared the screening process in which interviewers had the power to decide one's fate according to “[a] changed answer, a forgotten date, a charge of collaboration—any fact that nameless and faceless officials might seize upon became a nightmare for the DPs.”Footnote ¹⁰⁴ DPs understood quite well the dramatic consequences of such categorical bureaucratic procedures.

IRO documents reveal internal critiques of the screening procedure, as in the case of a memo complaining about the treatment of Venezia Giulia refugees in “a very bureaucratic manner, making applications difficult to submit.”Footnote ¹⁰⁵ Reflecting back on his work in the Rome Eligibility Division with M. W. Royse, Stephen Streeter concluded that IRO had adopted an overly legalistic approach. As an example, Mr. Streeter instanced two Venezia Giulian families living in the same room; the head of one family, having opted to retain Italian citizenship in Italy had not had his option approved and was accordingly found within the mandate; the head of the other family had opted in Yugoslavia to retain Italian citizenship and had come to Italy. His option having been approved, he was found to be not within the mandate.Footnote ¹⁰⁶ Added L. M. Hacking, IRO's Chief Historian and author of the interview with Streeter, “In Mr. Streeter's submission both families were equally refugees and in fact both families were treated identically by the authorities in Italy. The distinction between them, though it could be defended on logical grounds, made nonsense to the individuals.”Footnote ¹⁰⁷ Although too legalistic on the one hand, IRO personnel frequently made eligibility decisions on “basically emotional and impulsive” grounds, on the other hand, as Streeter found.Footnote ¹⁰⁸

Other members of the IRO staff, such as V. A. Temnomeroff of the Review Board, also admitted to errors in IRO's policies and practices in regard to the classification of Venezia Giulian refugees. In an earlier letter, Gesner had lamented that the IRO mission in Italy had previously accepted a number of Venezia Giulia refugees with provisional passports who the mission was now forced to exclude. Whereas Gesner sought to explain why the Italian Consular officials in Yugoslavia may have issued these passports, Temnomeroff adopted a much harder line.Footnote ¹⁰⁹ Though acknowledging, “It is to be regretted that IRO has admitted some refugees Within the mandate by mistake,” Temnomeroff nonetheless maintained, “it is difficult to justify establishment of a policy on the basis of mistakes.”Footnote ¹¹⁰ Temnomeroff thus pushed the legalistic approach criticized elsewhere by Streeter:

In addition, argued Temnomeroff:

IRO's reversal of some of its policies and the hard line adopted by some staff, like Temnomeroff, prompted numerous letters of protest by Giulian refugees. With the reclassification as ineligible of certain refugees previously deemed within IRO's mandate, a group of Giulian refugees in the IRO camp at Carinaro d'Aversa sent a letter to the IRO's Director-General.Footnote ¹¹³ They protested the change in policy as unfairly penalizing the refugees who “are at present in a critical material and moral position, for the prolonged stay in the camp has exhausted all their material and financial resources, because of the inadequate assistance.”Footnote ¹¹⁴ One Giulian refugee sent a protest letter to IRO's Geneva offices with the condemnation, “Your Organization lacks two things: COHERENCE and HUMAN FEELINGS, but it is full of useless and detrimental red tape.”Footnote ¹¹⁵ Another woman used more conciliatory language in her appeal to Geneva. She begged that she and her family be reclassified as eligible, despite having opted in Rijeka/Fiume, “as that was the only way of escaping from that hell.”Footnote ¹¹⁶ Applying to the IRO Office in Milan for assistance, the family was initially declared eligible. Later, however, “when we were waiting to be called to the camp, in September 1950, we were told that we would not be able to emigrate, because we had opted in Fiume.”Footnote ¹¹⁷

The IRO files contain numerous such letters that testify to the disappointment and anger felt by many Giulian refugees whose hopes for overseas emigration, sometimes raised by an initially positive eligibility decision, were ultimately dashed. In these cases, both legal technicalities (whether one had opted, provisional passports) and subjective evaluations (of whether they were “genuine” Venezia Giulians, together with their presumed ethnicity and reasons for opting/not opting) played into the decisions.