The ethnicization of politics

"Mr. K did not think it necessary to live in any particular country. 'I can go hungry anywhere,' he said. But one day he went through a town occupied by the enemy of the country in which he was living. One of the enemy's officers came toward him and forced him to step down the pavement. Mr. K stepped off and became aware that he was furious with this man, and not only with this man, but even more with the country from which the man came. So that he wished it could be wiped off the face of Earth. 'Why' asked Mr. K, 'did I become nationalist for that moment? (...)."

In December 1989 Romania faced a wave of public protest movements against the communist regime. The movement was initiated in Timisoara, a city near its western border, and spread over the country. As in other cities, on December 21, the inhabitants of Targu-Mures, a county capital in central Transylvania with a population evenly divided between Romanians and Hungarians, occupied the main square of the city. The crowd called for the resignation of the president Nicoale Ceausescu, for free elections and the dismantling of the political police - the Securitate. One of the Romanian intellectuals who witnessed the anti-Communist protests recalls the scene:

The main square was filled with people. In front of the City Hall there was a line of special troops, soldiers wearing masks... and somewhere behind them there were armored vehicles, waiting for an order. I saw everything. I saw all sorts of scenes... I saw young people who unbuttoned their white shirts... you see, in the square there was a space between the soldiers and the crowd. And those young men, their shirts opened were walking toward the soldiers saying: "Shoot me, if you have to shoot somebody!".(...) The next day the people were there [gathered in the square] shouting, just like on the first day: "The army-our brothers", "The Army and the People!" An army major still calm (...) assured us that there would not be any bloodshed. But they [the army] actually did shoot [and kill] people in the square, on the previous evening.(...) Around 11 am there was an announcement on the radio that the dictator had fled with a helicopter. In that moment (...) many people embraced the soldiers. Well, all this story it is very euphoric, a story of sudden brotherhood (...).I saw a Hungarian man who was comforting a soldier wearing a black mask, saying, in a very picturesque jargon [in Romanian]: "You see, man, you see that we can be brothers?!" And the other, very reserved and sweating, muttering "Yes, yes." What else could he say? So... I saw joy in those moments, everybody was so happy... (Gheorghe, male, 39, Romanian, journalist)[1].

Three months later the main square was filled again, but the joyous exuberance of December had been replaced by the dark tension of suspicion and hate. The ethnically mixed crowd of December had become two distinct crowds, one Hungarian and one Romanian, separated by a tiny police cordon. At a certain moment, Romanian peasants brought by trucks from the neighboring villages stormed the Hungarian crowd, triggering 12-15 hours of street fighting between Hungarians and Romanians that resulted in 6 dead and more than 200 wounded.

One of my informants, a half-Hungarian, half-Romanian student who identified with the Hungarian position recalls the scene:

If you know how it was, there were two groups [Hungarians and Romanians separated in two crowds with only a tiny space between them.] ...and, they started to throw things, around 3pm... they threw eggs, tomatoes [at each other]. Later on a police cordon was formed [to separate the crowds] (...) I did not see them [the Romanian peasants] at the time, but I heard the people around me saying: "Here they come!" There were many women with us, from our work place. (...) I felt tension and panic.. and they arrived... it was like a scene in a Western, a cloud of dust.. and the peasants with forks and wooden sticks in their hands... shouting and running toward us [attacking the Hungarian crowd]. And as the women ran away, I made space for them, as a gentleman [he smiles], until I saw them approaching like 10 meters in front of me, and I got frightened and I started to run myself. And as I was running, there was a guy near me who turned toward them and he was hit by a bottle, right in his face... the bottle was filled with water... and in the next moment the guy was all blood... I got really scared and I told myself: "I don't care what's happening and who [what politician] gets replaced, I just want out of here!". And I ran away..

I was walking on the street [looking for his girlfriend] and I saw 7 or 10 young men, Hungarians, running towards me with bricks, no, with stones in their hands. (...) And I saw them, and they were shouting: "Fogjátok meg" [Catch him!- in Hungarian] .. (...) I was scared, but I knew three of them. We played soccer together so I said to myself: "Hell, if they'll approach, those three will defend me." [Soon] I realized they were running after a [Romanian] truck driver. The guy almost reached his truck and they were shouting: "Fogjátok meg" [Catch him!]... It was like in the horror movies, you have my word, [since, once there] the guy could not start the engine. And hrrr... hrrrr... hrrrrrr.. [Imitates the sound of the starter failing to set up the engine].. The guy had run at least 15 minutes to reach his truck... and the truck would not move. And... as the others arrived, they broke his windshield... there were plenty of construction materials around the park... and there were plenty of bricks... and they [struck the truck with them,] broke his windows, imagine, the poor guy was still in the truck, and the others 10-15 bombarding him with bricks as long as they could, until.. he finally managed to start the engine. They hit his head, they broke the truck's windows, everything. The guy finally accelerated, moving left. There were cars parked on that side of the street, he lost control of the steering wheel and crushed all the cars into the wall. Then, he backed and left... I don't know what happened to him afterwards... it was my second shock [of the day] and I began to cry nervously (Petre, male, 23, half-Romanian, half-Hungarian, engineering student).

What happened in the three months that separate the two instances recounted above? How did it happen that people who fought and died together in the hard days of the Ceausescu regime, with no trace of ethnic or national hatred, end up so soon thereafter fighting each other in the name of their different ethnic or national affiliations? This paper seeks to shed light on the process through which ethnicity and nationality became increasingly relevant in the public and private lives of the inhabitants of Targu-Mures. It addresses the ways in which ethnic or national[2] belonging was transformed from a culturally available but not powerfully salient factor into an prominent mobilizing force that ended up justifying actions unthinkable but a few months before.

I adopt a processual approach emphasizing the complex web of action, interaction and reaction that facilitated the widespread use and intensification of ethnic framing. I do not understand my paper as an explanation for ethnic violence and street fighting as such, taken as a discrete incident. Instead, I chose the protracted process of ethnicization as the object of my study. My paper provides a "thick" analytical description of the 3-months long process. In doing so, I do not deny the important role that structural factors[3] (such as inter-state interaction, decline of GDP per capita or economic competition) play in setting the context to the succession of events. I stress, however, the small-scale rhythms of the ethnicization process in Targu-Mures by highlighting its dynamic dimension and characterizing it in terms of multiple sequences of moves and counter-moves. I thus explore a 'how' question rather than a 'why' question, favoring an "eventful" (Sewell, 1996), processual analysis of the ethnicization of politics and everyday life.

The ethnicization process tended to become more and more public, broadening its audience over time. In this respect, it is crucial to emphasize the points of inflection through which ethnic framing rose to a higher level of intensity or expanded from a limited arena to other arenas of public life. My account of the ethnicization process focuses on these inflection points that mark qualitative shifts in the ethnicization process: rather than tracking the increasing salience of ethnic issues in the city under the influence of a gradual, monotonic development, I stress its discontinuous changes.

In the months following the fall of the Ceausescu regime ethnicity became a prominent dimension of public life in most of Transylvania's cities. Only in Targu-Mures, however, did ethnicization eventuate in inter-ethnic violence. My account of this particularly intense process of ethnicization focuses on two analytical elements, the first common to all Transylvanian cities, the second specific to Targu-Mures.

a) Institutional fluidity: After December 1989, against the background of sustained efforts aimed to create a new institutional framework (or to radically alter the institutional practices and structures that survived the political shift) a struggle for control of local institutions emerged. On the one hand, the democratization of public life allowed for new institutions and organizations to appear, such as political parties, cultural, religious and civic associations. In Transylvania, in particular, some of these organizations were structured along ethnic lines, for example the Hungarian youth union (MADISZ), or the Hungarian political party (UDMR) introducing thus a new dimension into the landscape of an emerging civic society[4]. On the other hand, the regime change necessitated restructuring within the institutions that survived the political upheaval. In some places, like Targu-Mures, ethnic affiliation became the main criterion by which those who might control this or that factory, school or hospital were chosen. It did so by replacing any other competing criteria such as professional merit, political loyalty, age, status or gender. In the aftermath of the revolution and in the context of institutional instability ethnic or national elements came to play a more significant role in the public arena than had been possible before 1989.

Many of the inhabitants of Targu-Mures have living memories of the period before the '70s, when Targu-Mures was essentially a Hungarian city. The collapse of the regime made it possible for them (and others) to imagine that Targu-Mures could again be, administratively and symbolically, Hungarian. The fact that the city's population was evenly represented by Romanians and Hungarians made the question of 'ownership' much more acute in Targu Mures than in other cities of Transylvania. Most of these cities were preponderantly Romanian or, in the Szekler-region adjacent to Targu-Mures, preponderantly Hungarian. Before 1989 bureaucratic appointment operated according to tacit principles of ethnic representation. In general Romanians held the top administrative positions, with a more or less even distribution among Romanians and Hungarians in local administrative positions. After 1989 the distribution of positions became an explicit and highly contested issue. This triggered intense behind-the-scene struggle for control of the local power. Thus, the struggle for political and/or symbolical control of the city in a context of institutional fluidity generated a spiral of increasing ethnicization.

The vast literature addressing issues related to ethnic violence cannot be easily surveyed for clearly defined theoretical oppositions do not structure it. As any overview of the literature would be partial and arbitrary, I will not discuss here the multiple dimensions of this theoretical corpus.[5] Such an attempt would go beyond the limits and the aims of my study. My case-study of the events leading toward ethnic violence in Targu-Mures between December 1989 and March 1990 aims to provide an analytically informed "thick description" (Geertz, 1973) of an extended process of ethnicization. As indicated above my study intends to re-construct the processual dimension of the rise of ethnicity as the main frame of social perception.

On this understanding, the main objective of this section is to identify analytical resources that my narrative might draw on. In doing so, I rely on concepts and mechanisms described by a wide array of authors, not subsumable under a single theoretical label and indeed not all belonging to the literature on ethnic violence. Addressing the processual dimension of ethnicization, instead of taking the ethnic group for granted, I focus on a perspective that aims to answer questions such as: "how ethnicity is institutionalized as a political and cultural form?" and "What makes the nation-evoking, nation-invoking efforts of political entrepreneurs more or less likely to succeed?"(Brubaker, 1996:16).

My interest in the conditions and mechanisms through which the process of ethnicization takes place made me turn my attention to the literature on transition. The events covered by my study happened immediately after the regime change, in the first months of political democratization in Romania. The success of nationalist political entrepreneurs in this period may be related to the context of institutional fluidity in which these new political claims are made. From this perspective, Snyder (2000) establishes the link between the liberalization and democratization of the public affairs and emergence of nationalist conflict through several comparisons between historical and contemporary cases. Snyder stresses the early circumstances of the democratization process when threatened incumbents use state agencies and influence a partially free press to set a nationalist agenda. For Snyder divisive nationalism is a "by-product of elite's efforts to persuade the people to accept divisive nationalist ideas" in order to justify its rule 'in the name of the nation' and thus to remain in power (Snyder, 2000: 32). Post 1989 Romania constitutes a good illustration of Snyder's argument, a case in which the reformist communists undertook the power and used a nationalist agenda to legitimize their stay in power.

A concept relevant to my discussion is provided by Stepan and Linz (1996) who highlight several key prerequisites for a successful transition to democracy (five major arenas of a modern consolidated democracy). Thus, a political community with competing definitions of its boundaries (separatism, irredenta) faces a "stateness" problem and will fail to develop a stable democratic system. Although the case of Targu-Mures did not present a "stateness" problem, it was partially perceived as such and the concept proves useful in my analysis.

In the same vein Offe (1995) covers the process of ethnification of politics in the context of post-Communist societies. He defines the ethnification as the resultant of causal forces that manifest themselves through the actions of rational strategic individuals. Offe underlines the key issue of simultaneity of several processes that occur in these societies: the triple transition in political, economic and social terms. The issue of simultaneity plays a large role in understanding the evolutions that brought about ethnic violence in 1990's Romania.

In the context demarcated by the literature on transition from authoritarianism to democracy, the literature on ethnic violence provides additional analytical tools to approach the case of Targu-Mures. Horowitz (2000) provides a way in which a processual approach can be combined with a concern for comparative, structural analysis. He focuses his attention on a specific kind of ethnic violence, "an intense, sudden, though not necessarily wholly unplanned, lethal attack (...) [where] the victims [are] chosen because of their group membership." His perspective emphasizes a structural "natural history" of ethnic violence, as a highly patterned phenomenon. Through the comparative study of a set of "deadly ethnic riots" the author searches and identifies the temporal and spatial rhythms of the riot, the precipitants of the action, and the attitude of the authorities toward victims and rioters. Horowitz's approach perceives ethnic violence as 'a process with a logic of its own, its traits depending on a succession of contextual factors, such as: the structure of opportunities for violence, the organization of participants and the nature of leadership, police behavior. The book favors an 'eventful' perspective on the nature of ethnic violence, devoting much attention to the significance of the contingent events and to their relevant consequences; the dynamic of the phenomenon being conditioned by the changes that occur as the events develop. From this perspective, the job of the social scientist approaching instances of ethnic violence would be to identify the rhythms, the small precipitants, the micro-mechanism that constitute switching points, directing the course of the events. It is in this instance that the work of Horowitz is extremely useful to my project: the work singles out essential moments in the 'natural history' of an ethnic riot, such as the 'competitive outbidding' of the ethnic leaders, the precipitant action of political and ceremonial demonstrations, of ethnic strikes, or of divisive elections. All these elements will prove relevant in structuring my account of the ethnicization process in Targu-Mures, as they constitute key moments of its development. In a previous work Horowitz (1985) states that in legitimacy is distributed unevenly among ethnic groups competing for political prominence (or political equality). He connects group legitimacy to the concept of 'ownership' (of the soil, of the country) understood not only in its physical but also in its symbolic sense. Claims related to the ownership of places and institutions are indicative to re-definitions of the power relations between ethnic groups. In this reading, the contested ownership of the city (de facto and/or symbolic) represents, in the case of Targu-Mures, one of the key dimensions of the ethnicization process.

Another dimension on which I draw in my narrative is constituted by the significant role played by emotions in the process of ethnic mobilization. According to Horowitz (2000: p. 23 ) events viewed through the lenses of "bitter feeling and accusatory thought" are perceived as having a greater significance than those viewed without those lenses. Also, Lake and Rothchild (1996) highlight the essential dimension of fear, as one of the most frequent causes of ethnic conflict. Emotional reaction to perceived injustices, connected to the fear that these injustices will be perpetuated in the future are among the strongest factors that brought about the increasing use of ethnic framing I cover in the following pages.

Another exemplary work that illustrates the stress on the process of ethnic mobilization and on the microelements that bring about ethnic violence is Tambiah (1996). Tambiah focuses on civilian riots, underlining their common traits (such as such riot's duration, their phases, the nature of violence, and the action of the security forces and of the state.) The book identifies a plurality of mechanisms that are conducive to appearance and development of ethnic violence. Tambiah emphasizes the role of rumors, public space performances, collective ceremonies and popular protests in the development of a collective action, topics that also find illustration in the case I am addressing.

In the same vein, Beth Roy (1994) uses the tools of ethnography and oral history to make sense out of the communal violence she covers in her case-study of a Bangladeshi village riot. She highlights the process through which being Muslim/Hindu became relevant and suggests that this transformation be linked to political and economical changes. Her book is a careful re-construction of the ways in which a seemingly inconspicuous event triggered a serious local conflict. Roy's discussion of the relationship between leaders and followers, as well as her complex and rigorous methodological approach in-formed the way in which I addressed my case-study.

My paper constitutes a case-study dealing with the process of ethnicization that bought about an instance of ethnic violence in post-Ceausescu Romania. By 'ethnicization' I mean the process through which ethnicity replaces other interpretative frames. Following Brubaker, Loveman and Stamatov (2000) I favor a cognitive reading of ethnicity as not about "what people are" but about "how people are classified." The aim of the researcher is thus to uncover to the extent possible the cognitive schemata and categorizing practices that are used by people in order to make sense of the world. An essential advantage of this perspective consists in its definition of schemata as quasi-automatic 'processors' that organize experience. These processors are 'content-poor' requiring the participants to fill their empty slots in order to adapt them to the circumstances, and are activated by contextual stimulus or cues.

In the same vein I approach the literature on "framing." Here I follow Goffman who defined frames as "schemata of interpretation" that enable the actor to "locate, perceive, identify and label" the events from its surrounding reality (Goffman, 1974: 21). Frames are understood as rendering the outside world intelligible, organizing experience and guiding action. Another concept, developed by Snow and Benford (1992), is the concept of "master frame", defined as a meta-frame, being to "specific movement specific collective action frames as paradigms are to finely tuned theories." Ethnicity can be seen thus as a master frame that provides the rules and mechanisms through which its derivative, subsequent framing mechanisms, invaded the patterns through which sense is made out of the surrounding world. Thus, the cognitive approach combined with the literature on ethnic framing proves able to provide several key mechanisms and concepts with the help of which my research question can be approached.

As indicated above, my paper takes the entire process of ethnicization as its object and asks a "how" question rather then a "why" question. Because I am construing my project in this manner, with a focus on an extended process, rather than a discrete event, I am not interested here in testing competing theoretical explanations for ethnic violence understand as a discrete event. Rather, I offer an analytically informed narrative of an extended process. The main objective of my research is to provide an alternative way of approaching the issues of ethnic violence, by switching the object of study from explaining ethnic violence toward understanding the process of ethnicization itself and by exploring a possible manner of doing this.

a) The analysis of a set of 40 interviews I conducted in Targu-Mures in the summers of 1998 and 1999. Andrea Moraru, who worked in Targu-Mures for a parallel project in 1998, was kind enough to share with me 7 of her interviews with prominent local intellectuals and politicians.

b) The analysis of the local Romanian language daily newspaper Cuvantul liber between December 22, 1989 (the day of its first issue) and April 5, 1990.

c) The official report released by an investigating committee of the Romanian Parliament, that investigated the case immediately after the events, in April-May 1990, as well as other official documents, including reports of human rights groups such as Helsinki Watch. I have also drawn on two detailed, though partisan, accounts by key participants: In cumpana lui martie (In the balance of March) by col. Ioan Judea, mayor of the city at the time, and Black Spring by Kincses Elod, one of the leading figures of the Targu-Mures Hungarian community.

My interviews were mainly with ethnic Romanians, since I do not speak Hungarian. Nonetheless I had 3 interviews with ethnic Hungarians (in Romanian). During my first research period (summer of 1998) I simply knocked at doors, aiming to sample a diverse pool of neighborhoods (in the city center and in the industrial suburbs). The interviews were accepted in 1 out of 3 cases in the central areas and in 1 out of 5 cases in the industrial suburbs. This method proved to bias my sample in two ways: on the one hand there were mostly better educated men that tended to accept the interview, and on the other hand only people who were relatively open to discuss this sensitive matter self-selected. During my second research trip at Targu-Mures (summer of 1999) I attempted to select my informants in order to correct for the biases introduced by my previous method. I did so by using the networks of some friends from the city or previous informants who introduced me to their friends. My set of 47 interviews still retains a slight under-representation of women and of less educated individuals.

In using my data I had to address the complex issues related to the type of data grounded on re-collections of events that occurred 8-9 years before the interview was taken. All the quotations I use in the paper represent themes that are to be found in most of the interviews and are not denied by any of them. Some subjects (intellectuals such as Maria and Gheorghe) express in a more concise and clear way their thoughts. Due to this fact I chose to quote them more often than others throughout the paper.

The personal stories that the interview encounter elicited from my informants re-create the atmosphere and the personal reactions toward the events. I used data available from other sources (Parliament Committee, Helsinki Watch reports) in order to structure and position the personal memories on a more rigorous chronological account. The interviews unveil elements of cognition and awareness, mechanisms as the re-evaluation of past through feed-back, moments of sudden awareness of the ethnic criterion as a crucial factor in everyday life. I attempt to reconstruct the process of ethnicization mainly among ethnic Romanians. My data for the most part, but not exclusively, comes from their narratives. Consequently the presentation of the case is shaped by a certain asymmetry that favors the focus on the 'Romanian community'.

Nevertheless I do pay attention to the Hungarian perspective and make use of the pertinent primary and secondary Hungarian sources. I still have more to say about the parallel mechanisms and changes that affected the perception and the behavior of the Romanians. Needless to say, the analysis of the case cannot be made without taking into consideration the interplay of factors that brought to the several tides of action/reaction between the two 'communities'.

1. Romania[6] was a predominantly agrarian country before WW II. The communist regime imposed by the Russian tanks brought to power a party (the Romanian Communist Party) that hardly numbered 800 members. The party thus had to secure at least the acquiescence of the population for its policies. In order to eliminate any opposition the party first crushed the rather small urban elite that before the war dominated Romanian society. It also sent tens of thousands of people[7] to prisons and forced labor camps. Romanian communists, once in power[8], put in practice radical social and economic changes such as the collectivization of land, the nationalization of industry. The result of these policies was the elimination of the former cultural and social elite and the creation, in time, of a new intellectual elite, produced and trained by the communist higher-education system, an elite that had strong reasons to be thankful to a regime that transformed the peasants' children into 'gentlemen' -doctors, engineers, teachers[9]. In this social landscape, the government of Nicolae Ceausescu[10] (1964-1989) was able to pursue with almost no opposition a policy of extreme homogenization of the Romanian society[11] that brought it to a degree of leveling that lumped together all citizens (party members or not), except the core of political rulers and leaders of the repressive forces. Ceausescu's regime claimed its legitimacy on a version of national-communism that hailed its 'Romanianess,' stressing the country's independence against Moscow's interference[12]. The success of this stance was in part grounded in the extreme unpopularity of the anti-national policies of the first (Soviet dominated) communist government[13] in a context of historical antipathy towards Russia. Romanian nationalism represented a useful tool of the Romanian Communist Party first invoked by under Gheorghe Gheorghiu-Dej and developed by his successor Nicoale Ceausescu. His version of national-communism contributed to Romania's 'uniqueness' in the socialist block.

In the 1980's most of the countries of Eastern Europe were experiencing various attempts at reforming the system. Romania followed a specific neo-Stalinist path, the regime focusing on extreme austerity measures to face the mounting economic troubles. These austerity measures were enforced through the intermediary of over-sized and omnipresent political police that controlled and crushed even the incipient civic forms of opposition to the regime. Given these conditions, unlike the transition path of the neighboring countries the Romanian transition from communism to democracy was violent and tainted with blood. Spontaneous popular mass mobilization against the regime faced armed repression, and more than one hundred people died protesting on the streets. According to Linz and Stepan, (1996: 54) the "exceptionalism" of the Romanian case can be attributed to the specific traits of the Romanian political regime of late 80's, a "sultanistic" regime characterized by clientelistic methods and "unrestrained personal rulership." Instead of a negotiated transition Romania faced a revolutionary rupture with the past. The system had been radically destabilized and almost everything had to be created anew, thus making Romania's transitions to democracy all the more complex and singular in the regional context. It was in this period of institutional fluidity that ethnic violence erupted in a medium sized town in central Transylvania.

2. Transylvania is one of the three historical provinces that together constitute modern Romania. Throughout history the Hungarian nobility and Hungarian, German and Jewish bourgeoisie dominated the province economically and politically, despite the fact that the majority of its population (at least in the times for which census data is available) was ethnic Romanian. The Romanians from Transylvania were discriminated against under Habsburg and, later on, Hungarian rule. They were heavily underrepresented in the Parliament and administration. The large majority of Romanians were peasants, while Hungarians predominated in the cities and also in industry and administration[14]. Most of Transylvania's Hungarians are members of the Protestant Calvinist church and of Roman-Catholic Church. The majority of Romanians are of Greek-Orthodox faith while 6% of them are members of the Romanian Greek-Catholic church (which belongs to the Roman-Catholic church, preserving nonetheless the Greek-Orthodox liturgical ritual).

After WW I, Transylvania, as a province of the collapsing Austro-Hungarian empire, was allocated by the peace treaty of Trianon to the kingdom of Romania by virtue of the fact that 2/3 of its population were ethnic Romanians. The unification with Transylvania brought some 1.3 millions ethnic Hungarians (as well as German, Jewish and other minorities) under the authority of the Romanian state, including a region (Szekler-land) in which Hungarians constituted 80-90% of inhabitants. This large ethnic enclave was situated right in the middle of the new Romania. In the years that followed unification, the Romanian state made considerable efforts to support the development of a Romanian bourgeoisie in Transylvania, and to alter the ethnic balance within its cities -where the Hungarians heavily outnumbered Romanians.[15] During the 20's 30's and 40's the process occurred slowly: the administration become more and more Romanian, economic incentives were granted to the emerging Romanian entrepreneurs and a strong accent was placed on the development of a centralized Romanian system of free and compulsory education. At the same time a widespread state education system taught in minority languages was available through high school level. There were, however, no higher education institutions in minority languages. Between 1940 and 1944, as a result of an agreement dictated by Hitler and Mussolini in Vienna, the Northwestern part of Transylvania (including the city of Targu-Mures and the neighboring, compact Hungarian, Szekler-land area) was transferred to the Hungarian state. At the end of the war the region was returned to Romanian authority. Although atrocities associated with withdrawing the borders were committed by both sides, the level and the intensity of ethnic violence was low (compared, say, to the levels of the wartime ethnic massacres in neighboring Yugoslavia).

The post-war communist regime was more open toward minority-related issues: a Hungarian autonomous region was created, comprised of Szekler-land and its neighboring areas, with Targu-Mures as its capital. As the center of this unique administrative unit of the Hungarians from Transylvania, the city enjoyed special treatment: it was run by ethnic Hungarians, its language of administration was Hungarian and it hosted several major Hungarian-language cultural and higher education institutions: an institute of medicine, a theater academy. At the same time, throughout Transylvania parallel, Hungarian-language, cultural and educational institutions flourished. These included a variety of cultural and civic associations, a network of Hungarian-languages theaters, publishing houses, national and regional newspapers as well as radio broadcasts. There was also a parallel education system in the Hungarian language, spanning from kindergarten to university.

After 1964 the communist regime became more nation-centered (that is ethnic Romanian) because of the Romanian nationalist ideology used by Ceausescu as a legitimizing principle for his policies. The Hungarian autonomous region was abolished. Hungarian cultural and educational institutions continued to exist but on a smaller scale. The central government in Bucharest reduced, over the years, the number of students that the Hungarian-language departments could accept, as well as the number of publications, radio broadcasts and books produced in Hungarian. The nominal Hungarian schools had to accept Romanian-language sections within their walls. Nonetheless, the vast majority of Hungarians continued to be educated in their language.

The 60's and '70's represented a period of forced industrialization and urbanization. The cities expanded all over the country and the newly-build factories provoked a major exodus from villages to cities. If before 1944, 85% of Romania's population lived in rural areas, by the end of 70's the villages accounted for slightly less than 47% of the country's population. Industrialization and urbanization introduced dramatic changes in the demography of Transylvania. Due to the fact that in most areas of Transylvania the cities (encompassing a sizable Hungarian minority) were surrounded by ethnic Romanian villages, urbanization altered the existing ethnic proportions: more and more Romanians moved into Transylvania's cities, which lost their previous Hungarian character. These changes occurred in a context in which deliberate policies of ethnodemographic engineering[16] were also pursued by the state, that aimed on the one hand to alter the proportion between urban and rural population and on the other hand to increase the number of ethnic Romanians in Transylvania's cities.

3. In 1961 the Hungarian Autonomous Region, as noted above, was dissolved and replaced by three counties, (Mures, Harghita, and Covasna.) Targu Mures remained the capital of Mures county: the language of the administration reverted to Romanian, and the county governor was usually an ethnic Romanian. The general rule was that ethnic Romanians were more likely to hold the top administrative positions. Professional white-collar workers, like the large mass of blue-collar workers, faced less ethnic segregation; the odds of getting hired or promoted were very rarely influenced by one's nationality. The fact that the access of Hungarians to top administrative positions was quasi-blocked, while their access to almost any other positions was largely unrestricted, made the decision making process for selecting candidates for administrative and economic positions seem almost ethnically blind. Therefore, the relevance of ethnic or national belonging as related to everyday life practical issues was perceived to be minimal. Nevertheless the ethnic issue was still more sensitive for Hungarians than for the Romanians.

During this period Hungarian continued to be spoken in the public institutions but in different conditions: the Medical Institute in Targu-Mures, for example, had to host two departments: one in which the language of study was Hungarian and another one in which it was Romanian. The same process happened in most of the schools and high schools: the Hungarian-only high schools had to accommodate parallel Romanian-language departments and vice-versa. These changes not only transformed the former Hungarian-only institutions into schools with ethnically mixed population but also represented the moment when the Hungarians lost the administrative control over these institutions: their new Romanian colleagues had to be co-opted in the managerial structures. Not surprisingly when the communist regime collapsed, in all the cities with Hungarian minorities, the education system faced a powerful challenge to re-institute Hungarian-only high schools and higher-education institutes.

The dramatic changes introduced in the 1960s and 1970's had a powerful impact on Targu-Mures. The central government made huge investments, building a large chemical plant and some other industrial factories, together with numerous apartment buildings to host the industrial workers that moved into town from the neighboring villages (mostly Romanian), searching for jobs at the new factories[17]. At the end of the 70's the city's population was roughly divided between Hungarians and Romanians. This distinguished Targu-Mures from all other Transylvanian cities, which were either preponderantly Romanian or, in the Szekler region, preponderantly Hungarian. Moreover, unlike other cities, in Targu-Mures the ethnic proportion change is of recent date. It occurred during the lifetime of most of its inhabitants, people who remember the days in which the city was still a Hungarian world[18].

In the following sections I attempt to cover the development of the ethnicization process in Targu-Mures in the first months of 1990 focusing on the inflection points that mark significant steps in this process. I begin with a section on the baseline of ethnicity in the city in the late 80's, introducing the background circumstances against which this process must be understood. The following section focuses on the 1989 anti-Communist mass movements and their lack of ethnic markers. The first inflection point in the process of ethnicization reveals the context in which local politicians appealed for the first time to ethnic criteria in their attempt to build the structures of the new political decision-taking process in the aftermath of the regime-change. The second inflection point stresses the appearance of Hungarian-only organizations (ethnically grounded elements of the civic society) as new components of the public landscape of the city. The third step in the ethnicization process is marked by the debate over the 'ownership' of the Bolyai Farkas high school when issues of institutional control were for the first time expressed and grounded in ethnic terms. The agenda of the local media was influenced by these developments, and in a subsequent transition point, ethnicity invaded local media discourse. The strike of the Hungarian students marks another moment in the process, stressing the point in which elements of the civic society (parents' associations, youth organizations) became involved in the school debate. Yet another essential transition moment is revealed by the analysis of the media. Recent past events began to be re-interpreted and re-considered in the light of the new public concern for ethnic related issues. The following step of the process refers to the moment in which the ethnic organizations of Romanians and Hungarians reach mass dimension, staging series of ethnic demonstrations and counter-demonstrations attended by tens of thousands of citizens. Another transition moment is represented by the inflammatory media campaign promoted by the local newspapers, campaign that used intentional distortions of the message aiming to provoke and intensify ethnic responses from the readers. A final stage in the line of inflection points is marked by the first instances of ethnic violence, which triggered citywide ethnic clashes that were settled only by the intervention of the army. The outburst of violence transformed ethnicity into a master-frame of reference, the main schemata of interpretation used by the inhabitants of Targu-Mures.

In order to understand the rise of ethnic or national violence in Targu-Mures in 1990, it is necessary first to sketch a baseline of ethnic relations in the city before 1989. The language, the foreign accent (when speaking the other's language) and the names (first and last) are powerful indicators of one's ethnicity in Transylvania. Arguably, ethnicity was more significant for Hungarians than for Romanians before 1989. The average Romanian inhabitant of Targu Mures before 1989 enjoyed a privileged position in comparison to his/her Hungarian neighbor: he/she could afford to be almost unaware of the ethnic color of social and institutional relations. The Hungarians had been able to preserve their language and culture due to the institutional framework that existed after WW I. Ceausescu's rise to power in 1965 represented a shift toward more nation-centered Romanian policies. These policies contributed to the Hungarians' perception of their lives under communism as marked by their ethnicity. Unlike Romanians, many Hungarians perceived his regime as pursuing a nationalist agenda whose goal was the assimilation of all Hungarians.[19]

Therefore, many Hungarians from Transylvania tended to cluster among themselves and avoided exposure to the Romanian media and culture, listening to, for example, the news from the radio stations of neighboring Hungary. The subjects' narratives illustrate the practices upon which everyday life was built. According to a Romanian subject the Hungarians "lived in a [self-imposed] cultural ghetto, there was no dialog between the Romanian and the Hungarian culture, everyone minded their own business and, over a coffee, politely, they met from time to time and chat a bit.[20] " The citizens of Targu-Mures developed in time mechanisms and practices to manage their interactions. These mechanisms underlined the distinctiveness and allowed for the survival of Hungarian culture. Another subject reveals an even more nuanced view of the everyday interaction:

At our high school there were small groups, that is, there was a group that was composed almost only of Hungarian teachers, without including the others. We would go, say, to a painter's workshop to have a glass of wine... and as I looked around, I would notice that there weren't any Romanians there. Step by step I was accepted [by Hungarians], perhaps also due to the fact that I taught Hungarian, yet sometimes I was accepted, sometimes I wasn't. ... I used to be accepted at the little parties we had at school, but I was not invited at parties that were organized at home. And I had the feeling that I was not [accepted] (...) because I was Romanian (Maria, female, 40, Romanian, teacher).

The distinctions are finely grained: nuanced practices and changes in behavior are connected to the ethnic affiliation of those who interact. The fragments above illustrate the subtle ways in which the Hungarians living in Romania managed to preserve a level of groupness and of distinctiveness under an authoritarian national communist regime. The Romanians noticed the level of keeping to themselves of most of the Hungarian (or Hungarian speaking) groups and few resented it, but not as a major issue. Despite the tendency of Hungarians to cluster among them, however, in public interactions involving members of both ethnic groups markers other than ethnicity were more salient: profession, status or shared hobbies. My questions introducing ethnic distinctions or asking my informers about the ethnicity of participants at some narrated events received strikingly similar answers:

I don't know... it did not matter (...) What can I say, it did not matter for me. Why did you ask me this strange question? (Ion, male, 21, Romanian, truck driver.)

If you were a 'Metronom-er' [a fan of a particular music radio broadcast] it did not matter what your nationality was. (Petre, male, 23, half-Romanian, half-Hungarian, engineering student).

Another fragment[21] reveals the reconstructed element of these narratives: after 1990 people saw spontaneous soccer games among children organized along ethnic lines, which had not happened previously. All my subjects, with no exception, stressed that before 1989 ethnicity was not the dominant category by which individuals were identified. Instead people associated with each other according to shared interests. Although ethnicity was present as a way of characterizing others, it seems that its significance in everyday life was limited.

The dictatorship of Nicolae Ceausescu lowered the standards of living of all citizens of Romania to wartime levels. Food was rationed, central heating was either stopped or brought to a minimum 14-16' Celsius (57-61 Fahrenheit) during the winter and electricity was limited according to a schedule of 4-6 hours per day.[22] The economic hardships faced by the population limited the range of interests to survival-related issues and focused its attention on the dictatorial couple, the two Ceausescu, who were considered to be primarily responsible for the situation.

The mass protests in December 1989 were united in their main goal: "Down with Ceausescu!" The anger and the frustration experienced by the people under a leader whose reign lasted almost 25 years identified him as the source of all evil. Some argue that the Hungarian minority was in certain respects more affected than the Romanian majority, yet the hardships and the political repression made the differences less visible. In December 1989 Hungarians and Romanians alike filled the streets, together. As illustrated in the accounts below, those who protested against the regime did not invoke ethnic distinctions in their behavior and discourse; moreover, they, at least temporarily, overcame existing ethnically charged issues.

He [Suto Andras, well-known Hungarian writer] said that everything is over, that we are going to be brothers from now on, and then they chanted, ... all the time the people chanted in Romanian only. And I noticed a state... a state of extraordinary identification of the Hungarians with the Romanian flag (Maria, female, 40, Romanian, teacher).

Indeed during these protests slogans were shouted only in Romanian, and only the Romanian flag was waved while Romanian patriotic songs were sung in the squares. Moreover, according to one of my subjects:

The guy who first sang the anthem... the one who sang first "Romanian, Awake!" -was a Hungarian. ...It happened near the statue of Avram Iancu, we were all gathered there, and he knew that that song ought to be sung, but he did not know.. [the lyrics]. So he sung [the first words of] "Romanian, Awake!" and the crowd continued (Petre, male, 23, half-Romanian, half-Hungarian, engineering student).

Singing "Romanian, Awake!" (the anthem of the national-liberal 1848 Romanian revolution) under the statue of Avram Iancu (leader of the 1848 Romanian peasants war against the Hungarian revolution's troops) would have been almost unthinkable for a Hungarian several days before - the young man did not even know the lyrics. It seems that in the case of the young Hungarian, even the existing ethnic or national distinctions submerged if not disappeared in the new context.

In order to survive the crisis, the Communist regime tried to mobilize ethnic support. The armed forces around Targu-Mures were ordered to crush an alleged rebellion of Hungarian paramilitary forces. A Romanian officer serving in a Targu-Mures armored unit recalls:

They called us and said: "At Targu-Mures, at IMATEX [industrial factory] there are groups of people gathered in the plant's yard, generally, groups of Hungarians! They are armed and they are threatening..." Several sub-units went out in the city, patrolling, ... already the masses were marching toward the center... I think that only at that moment (...) [we] realized... opened our eyes and... "Just a minute, but there aren't any... any paramilitary groups... there are simply the people who claim their rights!." (Stefan, male, Romanian, ca. 33, army officer).

Once again this is an instance in which an ethnically framed perception of the situation is replaced by alternative "ways of seeing." The radical shift from 'Hungarian paramilitary troops' to 'the people claiming their rights' (my Italics) illustrates the trend that was associated with the anti-regime movement: the further diminishing of the existing (weak) ethnic distinctions.

The next section examines the emergence of an increasing process of ethnicization. I first discuss the struggle for power in local institutions that favored the appearance of ethnicity as a frame of perception.

Local politics: the struggle for power and the first issues related to ethnicity

At the national level, the National Salvation Front (FSN) led by Ion Iliescu emerged as the dominant political force that assumed the government of the country after the regime-change. Second rank Communist officials dominated the Front, which was also backed by the army. It was the army and not the local administration (which was not trusted by the FSN leaders) that constituted the main tool in the hands of the government during the first months of 1990.[23] Already in January, the Front had to face mass mobilization led by the newly created political parties challenging its domination of the political sphere. The mass mobilization in the capital resulted in the partial co-optation of the representatives of the opposition parties in the government.[24] Nonetheless, the Front dominated by far the decision-taking process (as well as the administration and the national media) until its unsurprising landslide victory in the May 20 elections.[25]

The fall of the regime on December 22, 1989 left an institutional power vacuum throughout the country. The participants in the anti-regime demonstrations found themselves in the unexpected position of potential leaders of the city. The protest had been a spontaneous movement, lacking political leaders or an organizational structure.[26] The group of revolutionaries that occupied the local headquarters of the Communist Party set up a local revolutionary Council of the National Salvation Front. They selected the name and the nominal affiliation of their organization following the events in Bucharest -which they watched 'live' on the national TV channel. They ended up appointing themselves and the few people they trusted to local administrative positions. Since no procedural or organizational structures were in place to indicate who was qualified for what position, the bargaining process through which seats and authority were allocated brought in a multitude of divisive issues and opinions. A well-known dissident and Hungarian intellectual (Kiraly Karoly) rose unchallenged as the head of the revolutionary county council.[27]

It is at this moment when the (previously neglected) tendency of the Hungarians to 'keep together' began to be perceived in a new light. At these negotiations the Hungarian members seemed more organized, knew each other better than their Romanian colleagues. This situation made some Romanians feel uncomfortable. One of my informants was invited to be a member of the local council and remembers the surprise of her first encounter with several other (Romanian) council members:

I have been told: 'It is great to have you here, now we are going to be more Romanians here, because these guys [the Hungarians] want to outnumber us here, and to dominate us" (Maria, female, 40, Romanian, teacher).

Similarly, the Hungarian representatives in the council appealed to ethnicity as a relevant criterion to the issues related to the elections for the Mures County Council of the FSN. According to one of the Hungarian representative in the council, the first lists with members were rejected because they "did not reflect the real composition of the nationalities and would have included some servants of the old regime.[28]"

The relationship to the former regime, either as collaborator or as dissident was the main distinction that dominated Romanian politics in the early 90's. 'The real composition of the nationalities' was a specific concern of Targu Mures local politics. The first days that followed the regime change foregrounded what became major developments over the following months: ethnicity emerged as a resource able to mobilize and organize interests (Glazer and Moynihan, 1975). I note here a first inflection point in the ethnicization process: the local politicians appealed to ethnic criteria in order to create a new legitimacy and a new set of rules on which to ground the political decision-taking process. Several days after the regime change a political party claiming to represent the interests of the Hungarian community in Romania (UDMR) was created in Bucharest. As Targu-Mures was an important Hungarian cultural center, its branch of the party included several prestigious leaders of the UDMR. Moreover, Targu-Mures' unique position as a city with a population evenly distributed between Hungarians and Romanians, (and as the former capital of the Hungarian autonomous region) made it an important player in the strategy of the emerging Hungarian political party. It was often referred to as 'the last bastion of Hungarianess' and as one of the most important centers of the Hungarians from Transylvania. The increasing influence of the UDMR in Targu-Mures' local politics provoked the reaction of local Romanian political entrepreneurs, who turned initially toward the provisional government in Bucharest for support. They felt betrayed when they realized that the center was either unwilling or unable to help them and that in Bucharest the UDMR was taken to be a legitimate political force.[29] Consequently, a local Romanian nationalist movement (The Union Vatra Romaneasca) emerged in Targu-Mures and spread nationwide.

These political developments did not occur only behind closed doors. Politicians and public figures held speeches that hinted at the appearance of ethnic differentiation in the city:

In the main square, the first speeches ...were give in Romanian and then were translated in Hungarian. Then, because many leaders were Hungarian, gradually they spoke Hungarian and ended up adding just a short summary in Romanian. And then, of course, the Romanians felt neglected" (Gheorghe, male, Romanian, journalist, 39 years old in 1990).

During the first weeks of January ethnically framed political struggles reached the city's inhabitants in indirect ways only. Public discourse was polite and ethnically neutral. Even if occasionally the language switched from Romanian to Hungarian, the key issues remained the same: - the organization of local public affairs and the condemnation of the former regime.

The "agenda setting" effect of the local media reinforces the same conclusions. In the two local papers (Cuvantul Liber in Romanian and Nepujsag in Hungarian), the political struggles were covered in terms of persons being replaced or contested. The main criteria addressed in the local press at the time in reporting the numerous changes in the local council focused on professionalism or lack of credentials and not on alleged ethnic takeovers and counter-attacks.

The management of the major economic units constituted another arena in which ethnic framing could have been referred but was not. The active political participation that followed the success of the mass protests led to the creation of incipient forms of independent trade unions that challenged the legitimacy of former managers in literally every factory or public institution. The heated debates that accompanied the forced resignations of these managers and the election of new ones by employees could have created a fertile terrain for ethnic framing and ethnic markers. Yet, the report of the Parliamentary Committee that analyzed the ethnic clashes of March 1990 highlights the fact that the ethnic proportion of the newly appointed managers was roughly similar to those which had existed before 1989.[30] The fact that the ethnic distribution of these positions remained unchanged supports the thesis that other criteria than the ethnic affiliation of the candidates had been taken into consideration when the workers voted to re-confirm or to dismiss their managers.

Another aspect of public life in the city reinforces the same conclusion. Before 1989 official signs could be issued only in Romanian. In December '89 the City Hall issued new signs (marking the denomination of official institutions), in Romanian and Hungarian - with no initial reaction from the media or population.[31] This occurrence was noteworthy; after February the question of mono- versus bilingual signs divided the city to such an extent that even 10 years later the matter is still high on the city agenda.

In Targu-Mures, ethnically framed issues began to be used for the first time in the context of emerging political contestation. Targu-Mure's inhabitants did not share to any significant extent concerns related to national/ethnic issues although elements such as the language of the public speeches or the bilingual inscriptions were already cueing toward more ethnicized frames of perception.

During the first weeks of January the use of ethnic framing was limited to the level of political entrepreneurs and of local administrative structures while other arenas, such as media discourse, the education system or the work place remained unaffected. As will be seen below, the major inflection points in the ethnicization process in Targu-Mures relate, among others, to the arena of public education and to the mass media.

For the citizens of Targu-Mures the regime change represented the moment when the arena of the civil society ceased to be a forbidden space. The free press, public debates and civic associations play a key role in the process of democratic transformation. Yet the sudden emergence of a free (or potentially free) press and of competitive elections in a fluid institutional context may promote ethnicization (Snyder, 2000). Already in the first days after the revolution, people began doing things that were not possible before: talking freely about public affairs in the main square, reading tens of newspapers and spending endless hours in front of the TV sets. This was also a period when values and ideas were radically re-configured and seemingly solid truths vanished in a day or two; new ideas and ideologies made their triumphant entry. It was a time when people's perceptions of the world were transformed. There was also a sudden interest in civic life. People were eager to get involved, to participate in a realm that had been completely forbidden before. Various forms of associations and organizations were created, such as political parties, cultural unions, youth clubs and professional associations. Some of them were organized along ethnic lines, and their specific goals included the preservation and the development of Hungarian culture in Transylvania[32]. Such were the Democratic Union of the Hungarians from Romania -UDMR or the Association of the Young Hungarians in Targu-Mures-MADISZ. Also, associations and organizations linking people of the same faith were usually structured along ethnic lines since Romanians' and Hungarians' religions do not generally overlap. Thus, already in January 1990, most of the associative life was either ethnically neutral or Hungarian-based. The appearance of Hungarian-only organizations (a new element in the public landscape of the city) marks another inflection point in the ethnicization process. It provoked appeals in the public discourse to "eliminate chauvinism of any kind.[33]" The concern for 'chauvinism' rather than for 'communist practices' or 'generation differences'[34] points to the influence of these ethnically based associations in the emergence of ethnic framing.

The incipient structures of civic society introduced in the public sphere elements that were defined along ethnic lines. As a result the media reported for the first time events in which nationality appeared as a differentiating criterion. The first distinction between ethnic Romanians and ethnic Hungarians was published in the local Romanian newspaper on January 11, in the letter "Some are more equal than others?"

Humanitarian aid is received from abroad, consisting of medicines, clothes, food and other consumption goods. It is a gift for our wonderful nation, which has been facing so much trouble. It is painful to realize that these goods are preferentially distributed. There is a special preference for the Catholic church, the Calvinist church, etc. The Greek-Orthodox church, which has the largest number of members, does not receive an equitable share of these kind of goods (A letter from a reader published in Cuvantul Liber, Friday, January 12, 1990 p 2).

Although the distinctions are spelled out in apparently religious terms it is the ethnic subtext that is dominant: the author contrasts the Romanian church against two Hungarian churches. Yet, the article appears not as a text endorsed by the publisher, but as a "letter from a reader." At the time newspaper's staff was not ready to publicly embrace a stance that would introduce an ethnic subtext in the public discourse. Moreover, this seems not to be an isolated case: a couple of subjects used ethnicity as a frame when they mentioned the issue of the humanitarian aid.[35]

Thus, the existence of ethnic networks of distribution (mainly based on religious organizations) was enough to bring to the fore disparities or at least the doubt and the assumption that "some might get more assistance than others." In a timid way the we/they distinction is for the first time expressed in ethnic terms.

In response to the creation of the UDMR (and other Hungarian organizations), a group of 36 Romanian intellectuals attempted to create their own association Vatra Romaneasca aiming to promote the culture and the interests of Romanians from Transylvania.[36] One of the founders, the head of the Targu-Mures municipal library explained:

In the moment in which our Hungarian compatriots began organizing and harassing other Hungarians to make them join the UDMR, the idea appeared of creating a Romanian organization. (Constantin, male, 52, Romanian, librarian).

My informant perceived the appearance of Hungarian political organizations as the outcome of the actions of a minority of political entrepreneurs imposing on the average Hungarian to make him/her join the organization.

Thus, the civic structures that developed in the first weeks after the regime change introduced an ethnic dimension into public discourse and facilitated the spread of ethnicity as a principle that structured social perception.

As I have stated already, Transylvania has a long history of minority education in Hungarian (Livezeanu, 1995). In the history of the region, schools have been a crucial factor in preserving minority language's and culture's. In January 1990 a school-related debate marked the first major step towards ethnic polarization. Before 1961 the city hosted a high school in which the language of study was exclusively Hungarian. As mentioned above, there were several others high schools that had parallel, Hungarian-language and Romanian-language departments and some high schools where the language of study was only Romanian. From 1961 on, the previously Hungarian-only Bolyai Farkas high school was forced to host several classes of students who studied in Romanian.

In January 1990, at the request of several Hungarian teachers and students, the local administration of the city considered a proposal to transform Bolyai Farkas into an exclusively Hungarian high school, moving the classes of students who studied in Romanian to another high school. The proposal was passed in the local council on January 18, unleashing a heated debate between the professors, students and parents involved in the change. The protests of the Romanian teachers and students from the high school prompted the county council to hold a second meeting dedicated to the problem on January 19. The compromise solution was that the Bolyai Farkas was declared at once a Hungarian lyceum, whose Romanian students were to be allowed to continue to attend until the end of the academic year. The compromise did not resolve the issue. The fact that some students and teachers had to leave their work and study place only because of their nationality made some Romanians suddenly and acutely aware of the national/ethnic issue. The first group involved was that of the high school's teachers:

I remember the first debates among teachers on the topic of the Hungarian lyceum. There was, on one hand... a feeling of being offended (...) a sort of offence as between lovers. "Why, you don't believe me?" That was the message that the teachers were conveying to each other, they, who had been colleagues for years... the Hungarians seemed deeply affected and almost crushed at the idea that the Romanians don't understand them... and on the other hand the Romanians were saying: "We suffered together, we did not do any harm to you, why don't you discuss this over with us?" (Maria, female, 40, Romanian, teacher).

The Hungarian and the Romanian teachers from the Bolyai high school faced a new situation in which it was their ethnicity that mattered, rather than their age, friendship relations or any other criteria. For some people, especially for the Bolyai teachers, this was the moment when the we/they distinction took on ethnic colors. The confession of one of them reveals the micro-mechanisms of action-reaction that occurred in those circumstances:

After several days off (...) I discovered [at school] a very loaded, tense atmosphere. The teachers were already split along a U-shaped table, Hungarians on the left, Romanians on the right. I asked what is happening and I was told that it is requested that we should leave the school. I was amazed... I mean... people with whom I had such good relations before, who were my friends! ..and what hurt me the most...was a [Hungarian] chemistry teacher -(...) she was a good friend of my wife's - saying: "You get out of here!!!" (Dumitru, Male, 34, Romanian, teacher).

For these teachers the arena of personal relations was invaded by a new distinction that rendered meaningless their previous ties and their common past. Dumitru discovered that in his relation with the Hungarian chemistry teacher something new happened: his main characteristic was not anymore 'the husband of my friend.' Instead, another trait had moved to a central position: he is/become more 'a Romanian' than 'a friend.' Surely, he always had been a Romanian for her, among so many other things (say, man, tall, teacher, good storyteller) but before this was just one of his multiple characteristics. We must not fail to note that this is only one side of the issue - her Hungarianess became, in reaction, her main trait in his eyes, since she chose it as the main criteria defining their relationship. Ethnicity became the main criterion of social perception that positioned those who addressed it in opposite camps, transforming them into players of a zero-sum game. The image of the teachers' room evoked by the informant is revealing: the spatial disposition of the teachers, Romanians on one side, Hungarians on the other side of the U-shaped table rendered palpable the increased salience of national/ethnic distinctions.

In the arena of public discourse the school issue marked an inflection point in the ethnicization process. The ethnic subtext that only a week before was presented in convoluted terms and relegated to the "letters" section of the Romanian language newspaper rose emerged as the main issue of the day. On January 20, 21, 23 and 24 the Romanian language newspaper published at least one article dedicated to the ethnic separation of the schools. The issue of January 25 of Cuvantul Liber dedicated 5 articles to the topic. The 26th January issue devoted an entire page to the articles related to the problem and printed in large letters on its first page the national anthem "Romanian, Awake!"

Some Romanians interpreted the Hungarians' claims of having their own institutions in the city as a first step toward their complete control of Targu-Mures, and later on, of Transylvania. Since the region switched hands from Romanians to Hungarians during World War II (things that many inhabitants remembered) it was not unthinkable that the final aim of the Hungarian leaders was to actually unite again with Hungary. The schools issue, and the new political and civic involvement of the Hungarians in the public affairs were thus connected to a territorial and a stateness problem (Linz and Stepan, 1996). The association of the two themes: "Hungarians want to control some schools" and "Hungarians want to control Transylvania" (through the intermediary: "Hungarians want to control the city") was facilitated by an incident that occurred at the height of the school debate. In the same loaded issue of January 25^th Cuvantul Liber published a "Declaration" of the Romanian news agency:

"Romania received with surprise and concern the statement of the president of France, Francois Mitterand, claiming that Hungary was "deprived by two thirds of its territory" after the two world wars.[37]"

Since Mures county (and Transylvania) was a part of the "two thirds" of the Hungarian territory mentioned by president Mitterand, the statement could hardly have been issued at a worse time, despite the next day's announcement, stating that "the French government does not mean to question by any means the existing borders of Europe."[38] To add to these anxieties, in an appearance on national TV in the evening of January 25, the head of the Romanian state, Ion Iliescu expressed his concern about the Hungarian minority's "separatist ambitions". The issue of minority rights was thus connected to a perceived stateness threat, and the reasonableness of this association was backed by highly credible sources, such as the national news agency and a discourse of the head of the state. The issue of simultaneity (Offe, 1995) deserves to be stressed at this point: one has to remember that the inhabitants of Targu-Mures had to deal at the same time with radical changes in the political sphere, with the opening of the civic sphere, with a re-structuring of the grounds of the economic system and with a radically new configuration of the international relations.

It is in this framework that cues and stimuli such as the declarations of president Mitterand might have activated or reinforced a territorial reading of Hungarian's claim for minority rights.

The school debate had a snowball effect on the public and private lives of the inhabitants of Targu-Mures. If before ethnicity was relevant only within the restricted arena of local administration, the school controversy made ethnic framing widely available as a mean of making sense of public life. Whichever side one identified with, one could relate in an emotional manner to it. People tended to consider themselves members of a victimized minority, against which injustice was done. The Romanian side was upset that the children were thrown out of school "in their own country" and that nobody intervened to correct the injustice. The Hungarians felt that not even the hard won freedom and democracy could guarantee their rights and resented their minority status even more than before. At this point the role of emotions and violated sense of justice connected to the theme of "ownership." For many Hungarians, "getting back" the high school became an issue of crucial importance in their attempt to re-negotiate the status of their ethnic group in Targu-Mures. At the same time, some Romanians tended to perceive the matter as a first step in a process that allegedly aimed to restore the prominence of Hungarians in the city.

On January 30, unhappy about the compromise decision of the county council, the Hungarian students of the Bolyai high school went on strike, demanding the immediate transformation of the school into a Hungarian-only institution. Their slogan was "Itt es Most!" (Here and Now!). At the level of political arena, county leaders had either to radicalize their position toward ethnic issues or to be overshadowed by more daring colleagues. According to Kincses Elod, a leading representative of the Hungarians in the local council:

"On January 30 I went to my old school and I tried to convince the students' committee and the parents' committee [that organized the strike] that they should accept the January 19 [compromise] resolution... One parent there said to my face that I was a traitor to the Hungarians, and that I was unwilling to assert just Hungarian demands in order to further my political career. His words were approved by some present."[39]

From this perspective I consider well-gounded the Roy's (1994) critique of Horowitz, questioning the unidirectional relationship between elites and masses. Indeed, what this moment suggests is that leaders do not always lead and sometimes are forced to follow. What we notice here is that the parents (some of them) weremore radical than the ethnic community leaders. Another (unintended) effect of the struggle for school was the appearance in the civic society of exclusively Romanian associations. On January 25^thin Reghin (a small city in Mures County) the "Romanian Brotherhood" was created. One of its main goals was to make sure that "the Constitution designate a single official language in all domains of activity (...) and that to be Romanian."[40] Late in December 1989 several Romanian intellectuals attempted to build an ethnic-Romanian organization: the Union Vatra Romaneasca. Nevertheless, the founders of Vatra were a relatively limited group, which did not take any public action and therefore passed largely unnoticed in the first weeks following the regime change. Yet, in the context of the school debate, the Vatra Romaneasca Union held its second meeting on February 1, 1990. This meeting marked the split between the people who backed its initial (civic-cultural aims) and those who favored a more political, nationalist orientation meant to "defend the rights of the Romanians." According to one of the founding members:

Things changed, already at our second meeting, because of those who joined later on, and whose sole aim was to use the organization and to create a political party. In these conditions, most of the founding members canceled their membership. (Constantin, male, 52, Romanian, librarian)

If in January the exclusive-Romanian movement defined itself as a ethnic-cultural one and represented a restrained group of intellectuals, in the new context created by the school issue, Vatra Romaneasca became an organization with a mass audience and radicalized its goals and demands.[41] One of its leaders defined the position of many Vatra sympathizers when he said: "We are not reactionaries, but we react" [facing the Hungarian demands.] The new line adopted by Vatra involved mass demonstrations and a policy of force, aiming to impose Romanian control over the city and its institutions and to make sure that none of the Hungarians' claims would be met. In at least one case that I covered people known for their active participation in the anti-Ceausescu protests turned into organizers of Vatra. More than that, people made sense out of the new ethnic movement relating it to the heroic theme of the recent anti-Communist protests. Here the concept of 'schema' proposed by Brubaker, Loveman and Stamatov (2000) might be useful. I suggest that at this moment some of the inhabitants of Targu-Mures were facing a new phenomenon and that they made sense of it by appealing to a pre-existing and still salient schema, that of the December 1989 uprising, the use of which might have been triggered by cues such as intense collective feelings, crowds invading the streets and shouting slogans. As one subject recalls:

In Targu-Mures they used the chants from Timisoara [the birth-place of the anti-Ceausescu movement]. The first slogans were: 'Targu-Mures-Timisoara!" (Gheorghe, male, 39, Romanian, journalist).

The people who participated at the first Vatra demonstrations associated their involvement with the struggle against Ceausescu and this association seems to have been an active one for the Romanian participants. Likewise, for Hungarians, participation in a struggle for gaining minority rights was easily connected and seemed at least as justified as their previous fight for their civic rights against the Ceausescu regime.

The radicalization of both camps favored the promotion in the leadership of the Hungarian and Romanian structures of those individuals who held extremist and ethnic-exclusionist views about the future of Targu-Mures. The radicalization happened either through changes of personnel at the top level, or by promotion of a more 'national' oriented political line by those who were already in leading positions within the ethnic organizations.

At the same time, the accentuated ethnic framing of these current issues produced feedback interpretations of past events. A good case study is provided by a succession of three articles published in 'Cuvantul Liber'. During the anti-Communist uprising in December 1989, in certain areas inhabited by Hungarian majorities, some Romanian policemen or political police (Securitate) officers had been lynched and their houses burned. What at the time was perceived as people's revenge against the repressive forces of the dictator Nicolae Ceausescu was re-interpreted in late January as Hungarians savagely killing Romanians.

On January 17, the Romanian-language newspaper Cuvantul Liber printed an article signed by the mayor of a small town -Sovata- where the Hungarians constitute the majority. The article denied "the rumors concerning some Romanian-Hungarian conflicts" in the town. Two days later, one of the journalists of Cuvantul Liber checked the facts and published an article that mentioned a number of incidents that might have justified the rumors denied by the mayor: some hand-written 'black lists' of the unwanted Romanian doctors were posted on various spots of the city and two Hungarian doctors expressed 'chauvinist' views at a trade-union meeting. Yet, on January 30, during the peak of the schools debates, Cuvantul Liber revisited once again the "ungrounded rumors" mentioned by the Hungarian mayor's article two weeks before. The new article reflects another inflection point at the level of the media discourse since the newspaper produced a completely different story: the article was illustrated with 4 photos of burned houses of Romanian police and Securitate officers which Hungarian "pyromaniacs" set on fire in the days of the revolution. It is practically impossible to claim that the newspaper's staff did not know earlier about these houses -the previous article mentioned even few hand-written lists posted on 'several corners' of the small town: the burned houses must have been more conspicuous. Thus, by the end of January, with the salience of ethnic or national perception overcoming the salience of other principles/criteria structuring the perception, even the hated Securitate officers were re-considered as 'Romanian' victims of malevolent Hungarians.

At the beginning of February, the strike of the Hungarian students of Bolyai high school triggered a similar movement among the students of the Hungarian department of the Institute of Medicine and Pharmacy. Some of my Romanian informants claimed that the undergraduate Hungarian students did not request only a separate higher education institution in Hungarian, but also separate student restaurants. Even if this is not true, the persistence of the rumor concerning the segregated student restaurants reveals how ready the Romanian public was at the time to believe such a story. Here, as in other settings (Tambiah 1996, Horowitz 2000) rumors played a central role in propagating the new ethnic frame and in reinforcing the perception of those of another ethnicity as belonging to an organized, well-structured group pursuing dangerous objectives. One of the new leaders of the Vatra Romaneasca, Radu Ceontea, made political career out of his 'bon mots,' such as addressing the gathering of the Hungarian students with the words: "In what language are we eating today, gentlemen?"

This is the context in which Vatra Romaneasca made an electrifying appearance on the public and political level on February 8. To the Hungarians' sit-in, the Romanian students answered with a 'walk-out.' The day debuted with a demonstration of the Romanian high school students, which gathered in front of Bolyai high school and from there marched throughout the city shouting slogans such as: "Not even in communism / Did we face separatism!," "We are brothers / Do not separate us!" but also "Romanians in Romania, Hungarians out!"" It was for the first time when the city faced an ethnically motivated demonstration. The student march ended at the sport's arena meeting, organized by Vatra Romaneasca. They were joined by a great number of citizens who felt the need to express their position and ethnic loyalty. There was an enthusiastic mass approval for union's policies and there was not much opening for compromise. 'Tough' speeches were applauded and answered with chants such as "We have a home, Hungarians are renting!" or "Out with the Hungarians!" Extremist speakers rose as popular leaders. One of the leaders of the union requested to reduce the proportion of Hungarian representatives in the city council from 50% (the proportion of Hungarians in the county) to 7% (the proportion of Hungarians in Romania).[42]

Thus the struggle for the schools triggered a struggle for the city, reaching a point where the already accepted rules of the political game were questioned and reformulated -namely, the right of each ethnic community to proportional representation in the local council. Those who claimed to speak on behalf of "the Romanians" seemed less and less inclined to accept these rules. The exclusive claims originating in exclusive values ("Romanianess" or "Hungarianess") denied the possibility of trade-off or compromise. The leaders of the two ethnic organizations engaged in a mechanism of out-bidding (Horowitz, 2000) the other side, their claims and goals spiraling more and more toward ethnic extremism. This process had as an outcome the increase in the salience of ethnicity in the everyday life:

It was as if each community, Romanian and Hungarian, had fought the battle for the city. Men more or less touched by paranoia dedicated themselves to the salvation of their city and of their community, which was martyred by the other community, which, in turn, believed itself a martyr. And the more some people on this side listened to their messianic leaders, the more the people on the other side would listen to theirs. (Gheorghe, male, 39, Romanian, journalist).

The Romanian students' demonstration was followed by a major Hungarian demonstration that remained in the city memory as "the candlelight march." Ca. 40000 ethnic Hungarian inhabitants marched in complete silence on the streets of Targu-Mures, each one holding in hand a book and a candle. The effect on the Romanian and international opinion was powerful:

The candlelight march showed up for the first time the existence of a disciplined force that reinforced the feeling of fear within the Romanian community. I told myself: "Look how many they are and how disciplined they are!" (Maria, female, 40, Romanian, teacher).

The ethnic demonstrations mark another inflection point in the process of ethnicization: they facilitated the perception and the understanding of the other side as a coherent, homogeneous and organized group. The ceremonial and ritual traits of the candlelight march (Tambiah, 1996) marked a new level in the self-perception of the Hugarians from Targu-Mures, and in reaction, in the way they were perceived by the Romanian inhabitants.

The rest of February and March saw many demonstrations and counter-demonstrations of both communities, in a symbolic struggle over the 'ownership' of the city.

The demonstrations meant also a major inflection point in the media discourse: the Romanian-language newspaper abandoned any pretence of independence and began to present extensively the actions of Vatra Romaneasca. It did not cover the candlelight march. More and more the media discourse grounded the ethnic claims of both sides on appeals to history: local history in the case of Hungarian-language media, national history in the case of the Romanian one (Bodó, Cosmeanu, Mátéffy and Mărginean, 1995). The sense of legitimacy grounding the 'ownership' claims was based, for Hungarians, on their historical domination of the city, while Romanians appealed to the policies of the former Hungarian governments that barred the access of the rural Romanian majority to civic rights and political equality.

Political involvement along ethnic lines increased by the end of February. The debate moved out of newspapers and schools into the street. The Bolyai Farcas high school controversy and the subsequent struggle for the control of that institution triggered a spiral of increasing ethnicization of local politics, civic organizations and press.

Unlike those people who had an active interest in public affairs and had occupied the political and media spheres in the previous weeks (Hungarians and Romanians), some inhabitants of Targu-Mures were not deeply affected by the ethnic debate. According to some of the students I interviewed, participating in demonstrations was for many a great way of cutting classes and having fun in the city rather than a way of showing ethnic loyalty. Several informants mentioned that they tuned their radios to the national or other regional station wavelengths, avoiding the increasingly heated verbal confrontations broadcast by the Targu-Mures radio-station[43]. The post-revolutionary Romanian public discourse was structured (outside of Targu-Mures) around issues pertaining to the mechanisms of the democratic transition and to the increasing challenges to the Iliescu provisional government. Some inhabitants of Targu-Mures followed thus another sort of politics, different from the local ethnic politics that invaded almost all the arenas of the public life in the city.

Targu-Mures was neither occupationally nor spatially segregated along ethnic lines. Day-to-day interaction, life-long personal relations between people of different ethnicity proved sometimes to be effective shields against the cleavages introduced by the emerging ethno-nationally framed public discourse. Despite the existence of competing ethnic organizations (Vatra Romaneasca and UDMR), ethnic local politics, and inflammatory press, not all the links that united some of the people in Targu-Mures could be broken.

I have to tell you that among my friends, my colleagues... the Romanian-Hungarian relations did not get worse. Spectacular changes in the Romanians' attitude toward what was happening in Targu-Mures intervened only after the March events [the outburst of ethnic violence] (Maria, female, 40, Romanian, teacher).

The fragment underlines the resilience of the interpersonal relations, some of which survived the process of ethnicization of the public sphere relatively well. Only the eruption of violence accounts for the break up of the last ties that united the people of different ethnicity.

The first 20 days of March were characterized by a series of ethnic provocations: rumors and facts presented in the media with the obvious aim of eliciting ethnic reactions. Their common denominator was the deliberate manipulation of the message in order to amplify the ethnic reading of the facts. It is beyond the scope of this article to determine who was behind the orchestrated media campaign, or behind the manipulation of collective acts that stirred ethnic violence in 19-29 March 1990. I intend to illustrate the ways in which the manipulation succeeded (or failed) to elicit the response of the population.

The last days of February constituted a new transition point in the troubled city. An unknown hand profaned the statues of one of Transylvania's Romanian heroes, Avram Iancu, calling in Hungarian, for the statue to be dismantled. This was the same statue beneath which a Hungarian was the first to have had the idea of singing "Romanian, Awake!" during the anti-Communist civic protests in December. Yet, the inscription was written in ungrammatical Hungarian ("LE VED" instead of "VEDD LE"- "Down it" instead of "Take it down.") - a possible indication of a provocation. On March 16, 1990 the Romanian-language newspaper published a first page article on the incident: "A barbarian act!" in which the author designates the unknown culprit as "the hairy monkey." The language and the images used by the newspaper degrade the culprit (the Hungarian, of course) to a sub-humane level.

A more obvious element of an ethnically inflammatory media campaign was the publication on March 14, 1990 on the first page of the Romanian language newspaper of a manifesto of the Los Angeles Hungarian diaspora asking for international protection for Transylvania's Hungarians. The handbill dated from January 15, 1988, but the Romanian newspaper intentionally altered the date, in order to present it as a proof of contemporary Hungarian bad-faith propaganda and to reinforce the association between a stateness issue and the minority rights policies. The handbill presented the map of Hungary before 1918, (with Transylvania as a Hungarian province) and deplored the torments of 3 000 000 Hungarians under the Romanian yoke.

On the 19^th of March two buses of Romanian peasants from the neighboring villages arrived in the city. The peasants had received 'anonymous' phone calls[44] that warned them that the Hungarians are going to kill their children, many of whom studied at high schools in the city. According to some informants from the village, the church bells rang and mayor and the chief of police advised the peasants to take two buses and go to Targu-Mures to protect their children.[45] The peasants demonstrated in front of the City Hall and forced the resignation of one of the ethnic Hungarian county leaders. Afterwards they marched toward the headquarters of the Democratic Union of Hungarians from Romania and stormed it. The Hungarians leaders took refuge in the attic of the building. The police did not intervene and the army, despite repeated calls, did not respond. Late in the afternoon, the Romanian head of the city council arrived with a truck of soldiers. The few Hungarians who trusted him and left the attic were brutally beaten by the Romanian peasants. Suto Andras, the most important Hungarian poet from Romania, lost an eye in the incident.

The following day there was a huge demonstration of Hungarians protesting against the violence of the previous day and demanding that their representative be reinstalled in the county council[46]. There was widespread anger and indignation among Hungarians, whose sense of justice had been violated by the March 19 events. As time went by, a smaller crowd of Romanians began to gather on the other side of the main square. Many Romanians feared that the size of the Hungarian crowd in the square was a proof that they intended to remove the Romanian leadership and to take control over the city. The tension grew and the few policemen present attempted to separate the two camps. The army refused to take positions to separate the two camps. The rumors run wild among the participants, fantastic stories about paramilitary troops, a possible Hungarian occupation as well as the arrival of the Romanian miners heightened the spirits. By 3 pm the same buses with peasants arrived in the square. Repeating the previous day's events, the peasants (armed with axes and wooden sticks) attacked the Hungarians gathered in the square. Violence was unleashed and street fights continued all day and night. The Hungarian side received the help of the Hungarian peasants and of the Hungarian-speaking Gypsies from the suburbs so the battle ended with a Hungarian victory. Late in the morning the army finally intervened in an effective manner and cleaned the square and the city hall of combatants. The ethnic violence ended with 6 deaths (3 Romanians and 3 Hungarians) and two hundred wounded.

The "March events" mark the culminating stage of the process of ethnicization and increased salience of national/ethnic criteria in the everyday life of the inhabitants of Targu-Mures. The events almost completely destroyed the relations between the inhabitants of different ethnicity, although even during the events there were instances in which Hungarians and Romanians collaborated.[47] After the events the ethnic or national dimension unequivocally dominated inter-personal affairs:

[Before 1990] We were observing Easter twice... my colleagues used to come to celebrate together the Hungarian [Catholic] Easter and then the Romanian [Greek-Orthodox] Easter, but it does not happen anymore. After the March events there was such a... a rupture. I am telling you, I had friends, and I mean good friends, and I must tell you, it all left me with a bitter aftertaste... I had Hungarian friends who helped me when I was in trouble, we were together like the fingers of a hand... yet after the March events... [the relationship ended] (Ana, female, 60, Romanian, retired technician).

The street violence shattered almost all the remaining links between people. The salience of ethnic framing surmounted the level of the public arena and invaded the sphere of private life and interpersonal relations. This represents the final degree of intensification of the ethnicization process in Targu-Mures, which reached its climax in the days that followed the March events.

More and more individuals accepted in the ethnic framework as their master-frame (Snow and Benford, 1992) of social perception. For them the 'events' constituted the point where the ethnic dimension became one of the major features framing their perception of their everyday environment: even children's soccer teams were split along ethnic lines, in the fortunate case in which they played together.

An interesting feedback mechanism must be underlined: events that previously had not been perceived in ethnic terms were now reinterpreted as such. One of my Romanian informants who worked together with a Hungarian colleague since 1975 recalls:

Sometime in the 80's her colleague's son was admitted to the English department, yet, because of arbitrary changes at the level of the ministry of education in Bucharest, her son was re-located, against his will, to the Romanian department. Discussing the issue at work, the angry mother said: "I would have preferred him to study even the Gypsy language, but not Romanian!" At the time my informant did not take these words seriously.

The day after the ethnic clashes, meeting again at work, the Hungarian had an outburst of anger and shouted to my informant: "Kill me, kill me, you can do it now, you are more Romanians here! "I was very upset", my informant confessed, "that she could think that way about me. And only then, she claims, she understood that when her Hungarian colleague said that she cannot stand the idea of her son studying Romanian, it was not just a moment of anger, but that that she had disliked Romanians all this time (Elena, female, Romanian, ca. 42, nurse).

The ways in which interaction was defined before the events were radically re-considered and re-configured in the light of the ethnic framing provided by the March ethnic upheaval. For my informant, the current behavior of her Hungarian colleague -appealing to ethnicity as the main attribute of the social interaction- projected the new ethnic frame also to re-interpret the past events, which got re-defined along the lines of the new frame[48].

In several cases the events marked major rupture points in the inter-personal relations. Some of these have endured into the present, as one of my Romanian informant recalls:

[Before '89] my daughter would have played in the yard [with Hungarian children](...) now, my granddaughter. Yet, my granddaughter goes to play in the yard, and she meets other children, Hungarians. Do you think they play with her? (...) [A Hungarian girl from the yard] tells to my granddaughter: "Go away!" (Ana, female, Romanian, 60 retired technician).

The painful process of re-building trust and the shattered friendships between Hungarians and Romanians took a long time and it will not be addressed here. In the aftermath of the March events local and national elections contributed to the appearance of an institutional framework that proved able to successfully manage the potential sources of conflict. No instances of ethnic mobilization or ethnic violence were reported in Targu-Mures after 1990. The leaders of the Hungarian party (UDMR) were able, as legitimate players in the Romanian political game, to collaborate and negotiate with their Romanian colleagues, the main points in which the interest of the two parts seemed to diverge. Although there are still many issues to be solved, the array of institutional agreements that was set up after 1990 proved to constitute a solid ground for the future inter-ethnic interaction.

My approach intended to highlight the processual dimension of the changes that occurred in the arenas of local politics, administration, civic life and media discourse between December 1989 and March 1990 in the city of Targu-Mures. Instead of seeking to explain the outcome of these extended processes of ethncization -the two days of ethnic violence in which they culminated- my paper has taken as its object these processes themselves and has sought to provide a theoretically informed, analytical description of them. It has, therefore, not been structured as a test of competing explanations but rather as an extended analytical narrative addressing not a "why" question but a "how" question.

My narrative stressed the succession and the interplay of mechanisms of action and reaction that brought about more and more instances of ethnic framing in Targu-Mures contexts. It stressed the role of the competing political forces in the re-structuration of the public life and public discourse along ethnic lines. It was the sphere of the local political struggle that produced the first instances of ethnic framing after the crumbling of the communist regime. The development of the civic society entailed the appearance of structures and networks that were organized along ethnic lines. Thus, ethnicity as a way of making sense of everyday life migrated from the rather limited arena of local political struggles toward the larger public sphere. The emerging civic society, with its nascent civic, religious, non-governmental organizations proved to be an aggravating element (Stamatov, 2000). The creation or re-creation of associations whose membership was defined (or restricted to) ethnic Hungarians provoked the response of the Romanians, and the creation of "Romanians-only" organizations, which had as sole reason of existence the desire to counter-act and oppose the Hungarian initiatives. These developments induced a permanent note of tension and conflict in the public life. The birth of a partially free press proves to be another sensitive element of the case study I addressed here. The role and the importance of the local media cannot be underestimated in the analysis of the mechanisms and processes that led to ethnic mobilization and ethnic violence. During the process, the media switched from mere presentation of the events to partisan presentation and ended by manipulating the public opinion and making open appeals to violence

Another key element of the ethnicization process in Targu-Mures related to the institutional fluidity associated with the regime change. The struggles for the control of this or that institution implied the ethnic framing of the adversary, and the ethnic perception of groups that before were not perceived as such. The members of the Hungarian community from Transylvania felt the homogenizing efforts of the previous (national-communist) regime as oppressive. In the context of the political democratization process, the emerging political and cultural organizations of the Hungarians militated for autonomous and separate institutions of Hungarian language that would ensure the survival and the development of their culture. The practical aspects connected to creating (or reverting to) Hungarian-only institutions affected the ethnic Romanian population of the city. In the context of an evenly distributed ethnic proportion and under the frame of the democratic (elective) character of the political life, the 'natural' Hungarian claims were perceived as a threat. The issue of minority rights evolved into a battle for the ownership of the city.

The ownership struggle erupted when the ethnic Hungarian representatives raised the issue of the school separation. The practical implication of reverting to a Hungarian-only high school meant necessarily evicting the Romanian teachers and students. The issue raised a storm of public (and private) debates and divided the city along ethnic lines. The political struggle at local and national level had an aggravating role over the efforts of amiably solving the problem. Successive levels of action and reaction brought to the escalation of the tension. Local (and national) politicians made a powerful resource out of the school problem, in a never ending game of bidding and out-bidding their political and ethnic adversaries.

The paper illustrates the various spheres of public life in which ethnicity became salient and the inflection points in which the salience of ethnic framing increased (in degree) and expanded (toward other arenas) due to various local developments.

I underlined the endogenous process of ethnicization, in which the awareness and sensitivity to national/ethnic matters augmented through mechanisms of action/reaction that inter-related the Romanian and the Hungarian inhabitants of Targu-Mures. Added to this process, I noted the exogenous influences that provoked the conflict between the Romanians and Hungarians in Targu-Mures, conflict that ended up in violence. Thus, the interplay of oral testimonies, analyses of the local press and the official reports released by the governmental and non-governmental agencies (national and international) enabled me to illustrate the ways in which the process of ethnicization emerged.

Anderson, Benedict. Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso, 1993.

Armstrong, John A. "Toward a Framework for Considering Nationalism in Eastern Europe." Eastern European Politics and Societies 2.2 (1988): 280-305.

Banac, Ivo, ed. Eastern Europe in Revolution. Ithaca and London: Cornell University Press, 1992.

Benford, Robert D. "An Insider's Critique of the Social Movements Framing Perspective." Sociological Inquiry 67 (1997): 409-30.

Berindei, Mihnea, Ariadna Combes et Anne Planche. Roumanie, le livre blanc. Paris: Editions La Déecouverte, 1990.

Bodó, Zoltán, Marius Cosmeanu, Csaba Mátéffy, and Paul Mărginean. "Alter/Ego targumuresean" (Targu Mures Alter/Ego). Altera 1 (1995): 77-89.

Brubaker, Rogers. Citizenship and Nationhood in France and Germany. Cambridge, Mass.: Harvard University Press, 1992.

________. Nationalism Reframed: Nationhood and the National Question in the New Europe Cambridge: Cambridge University Press, 1996.

Brubaker, Rogers and David D. Laitin and Laitin. "Etnic and Nationalist Violence." Annu. Rev. Sociol. 24 (1998): 423-52.

Brubaker, Rogers, Peter Stamatov and Mara Loveman. Beyond Social Construction: The case for a Cognitive Approach to Race, Ethnicity, and Nationalism, paper prepared for ASA meeting, August, 2000

Câmpeanu, Pavel. De patru ori în fata urnelor (Four times in front of ballot-boxes). Bucharest: Editura All, 1993.

Deletant, Dennis. Romania Under Communist Rule. Bucharest: Civic Academy Foundation, 1998.

________. Ceausescu and the Securitate: Coercion and Dissent in Romania, 1965-1989. Armonk, N.Y.: M.E. Sharpe, 1995.

Fearon, James D. "A Cross-Sectional Study of Large-Scale Ethnic Violence in the Post-War Period" - unpublished manuscript, 1997.

Fischer-Galati, Stephen. Twentieth Century Rumania. New York: Columbia University Press, 1991.

________, ed. The Communist Parties of Eastern Europe. New York: Columbia University Press, 1979.

Gamson, William and A. Modigliani. "Media Discourse and Public Opinion on Nuclear Power: A Constructionist Approach." American Journal of Sociology 95 (1989): 1-37.

Geertz, Clifford. The interpretation of cultures; selected essays. New York, Basic Books, 1973.

Gellner, Ernest. Nations and Nationalism. Ithaka, NY: Cornelle University Press, 1983.

Glazer, Nathan and Daniel P. Moynihan, eds. Ethnicity : Theory and Experience. Cambridge, Mass.: Harvard University Press, 1975.

Gofman, Erving. Frame Analysis: An Essay on the Organization of Experience. New York: Harper, 1974.

Gurr, Robert T. Minorities at Risk: A Global View of Ethnopolitical Conflicts. Washington, D.C.: United States Institute of Peace Press, 1993.

Hitchins, Keith. The Rumanian National Movement in Transylvania, 1780-1849. Cambridge: Harvard University Press, 1969.

________. A Nation Affirmed: The Romanian National Movement in Transylvania, 1860-1914. Bucharest : Encyclopaedic Publishing House, 1999.

Horowitz, Donald L. Ethnic Groups in Conflict. Berkeley : University of California Press, 1985.

Ionescu, Ghita. Communism in Rumania, 1944-1962. Westport, Conn.: Greenwood Press, 1964.

Jowitt, Kenneth. Revolutionary Breakthroughs and National Development: The Case of Romania, 1944-1965. Berkeley: University of California Press, 1971.

Kligman, Gail. The Politics of Duplicity: Controlling Reproduction in Ceausesu's Romania. Berkeley: University of California Press, 1998.

________. The Wedding of the Dead: Ritual, Poetics, and Popular Culture in Transylvania. Berkeley : University of California Press, 1988.

David A. Lake and Donald Rothchild, "Containing Fear: The Origin and Management of Ethnic Conflict" in International Security, vol. 21, No. 2 (Fall, 1996): 41-75

Levy, Robert. Ana Pauker: The Rise and Fall of a Jewish Communist. (Berkeley: University of California Press, 2000) forthcoming.

Linz, Juan J and Alfred Stepan. Problems of Democratic Transition and Consolidation: Southern Europe, South America, and Post-Communist Europe. Baltimore: Johns Hopkins University Press, 1996.

Livezeanu, Irina. Cultural Politics in Greater Romania: Regionalism, Nation Building, and Ethnic Struggle, 1918-1930. Ithaca, N.Y., London: Cornell University Press, 1995.

Offe, Claus. "Ethnic Politics in East European Transitions." In Jody Jensen and Ferenc Miszlivetz, eds. Paradoxes and Perspectives. Szombathely: Savaria University Press, 1995.

Posen, Barry P. "The Security Dilemma and Ethnic Conflict." Survival 35.1 (1993): 27-47.

Rothschild, Joseph. Ethnopolitics: A Conceptual Framework. New York: Columbia University Press, 1981.

Roy, Beth. Some trouble with cows : making sense of social conflict. Berkeley : University of California Press, 1994.

Smith, Anthony, D. The Ethnic Origins of Nations. Oxford, UK, Cambridge, Mass.: Basil Blackwell, 1986.

Snow, David A. and Robert D. Benford. "Master Frames and Cycles of Protest." In Aldon D. Morris and Carol McClurg Mueller. Frontiers in Social Movement Theory. New Haven, Conn.: Yale University Press, 1992.

________. "Ideology, Frame Resonance, and Participant Mobilization." In Bert Klandermans et al., ed. From Structure to Action: Comparing Social Movement Research across Cultures. Greenwich, Conn.: JAI Press, 1988.

Snow, David A., E. Burke Rochford, Jr., Steven K. Worden, and Robert D. Benford. "Frame Alignment Processes, Micromobilization, and Movement Participation." American Sociological Review 51. 4. (1986): 464-81.

Snyder, Jack. From Voting to Violence: Democratization and Nationalist Conflict. New York, London: W. W. Norton and Company, 2000.

Stamatov, Peter. "The Making of a 'Bad' Public: Ethnonational Mobilization in Post-

Tambiah, Stanely, J. Leveling Crowds: Ethnonationalist Conflicts and Collective Violence in South Africa. Berkeley, Los Angeles, London: University of California Press, 1996.

Van Evera, Stephen, ed."Hypotheses on Nationalism and War." In Charles Kupchan, ed. Nationalism and Nationalities in the New Europe. Ithaca, N.Y.: Cornell University Press, 1995.

Verdery, Katherine National Ideology under Socialism: Identity and Cultural Politics in Ceausescu's Romania. Berkeley: University of California Press, 1991.

________. Transylvanian Villagers : Three Centuries of Political, Economic, and Ethnic Change. Berkeley: University of California Press, 1983.

* * * "News From Romania: Ethnic Conflict in Targu-Mures." News from Helsinky Watch, May 1990, p.1-8.

The collection of the daily Cuvantul Liber, from December 22, 1989 to April 1, 1990.

Dumitru, N.S. and Verestoy Attila. Parliamentary report concerning the Targu-Mures events.

[1] I use the age of the subjects in 1990. Fictional names are used throughout the paper.

[2] I will use throughout the paper the concepts of 'ethnicity' and 'nationality' interchangeably; 'ethnicity' is favored on the American continent, while European researchers more often use the term nation/nationality.

[3] For large-scale comparisons aiming to identify structural factors that account for instances of ethnicization and ethnic violence see Gurr (1993), Fearon (1997), Van Evera (1995) and Posen (1993).

[4] The Hungarian ethnically-based associations were not the only ones to appear at the time. Other ethnically-based structures were created, such as the associations of the Gypsies, Germans, etc.

[5] For a recent review of the topic see Brubaker and Laitin (1998)

[6] For concise accounts of modern Romanian history see Fischer-Galati (1991) and Georgescu, (1991).

[7] The Communist regime imprisoned the leading figures and rank-and file members of the Romanian political parties that activated before its advent to power, members of the administration, army officers and cultural personalities, the priests and bishops of the various churches, small owners and even peasants who happened to be influential in their communities.

[8] On Romanian communist regime see Ionescu, (1964), Deletant, (1998), Fischer-Galati (1979), and Jowitt (1971).

[9] On this and other aspects of the social transformations of the Romanian society under the Communist regime see Verdery, (1983), p. 1 and passim.

[10] On Nicolae Ceausescu and his reign see, Kligman, (1998) and Deletant, (1995).

[11] On homogenization see Kligman, (1998), p. 32-35.

[12] See Verdery, (1991), p. 167-209 and 302-309.

[13] See Levy, (2000) forthcoming.

[14] For an account on the ethnic and social structure of Transylvania see Hitchins, (1969) and Hitchins (1999).

[15] On the Romanian state cultural policies in post-1918 Transylvania see Livezeanu, (1995).

[16] These policies of ethno-demographic change were very negatively perceived by the members of the Hungarian community. At the same time they were a component of an industrialization and urbanization effort that represented a chance for upward mobility and better standards of living for many people, including ethnic Hungarians.

[17] For an account on the relevance of urban/rural dichotomy in the rise of nationalism in Eastern Europe see Armstrong, (1988).

[18] One of my informants, a Romanian journalist describes the city in the mid 50's, as he saw it when he moved in from a neighboring [Romanian] village to attend a vocational school: "The general atmosphere was pleasant at Targu-Mures, especially for the students. Although there were 700 students at our school, of which the majority was Hungarians (...) there were no problems of this kind [ethnic] among us. Yet, the city was completely Hungarian, one spoke only Hungarian, at all meetings, beginning with those of the Communist Youth Union and ending with those of The Romanian Workers' Party [the name of the Communist party] one spoke exclusively Hungarian."

[19] The objective pursued by the regime was more encompassing: the actual goal was not the assimilation of the Hungarian minority (although steps to diminish its influence had been taken) but to transform all the citizens of Romania according to the ideal of the "new man" and to the ideas of the "illustrious" ruler. For the social policies of Ceausescu's regime see Kilgman, (1998) p. 19-42.

[20] Gheorghe, male, 39, Romanian, journalist.

[21] A fragment of an interview with two friends, one Romanian and other Hungarian:

Paul: "[My friend] saw for the first time, after 1990, the kids playing soccer, Romanians versus Hungarians. (...) And, being from Targu-Mures, he... found it very strange." Ianos: "I had never seen something similar. When I was a child we split the teams according to blocks.. and that's how we played". (Paul and Ianos, males, Romanian and Hungarian friends, high school students, 18).

[22] For a detailed account of the everyday life in Romania in late '80's see Kligman, (1998) p. 148-206.

[23] For an account on post regime-change Romania see Kligman and Verdery, (1992).

[24] Under the pressure of the February 28 mass protests, the Council of the National Salvation Front -the organism that ruled the country- was re-structured as the Council of the National Unity, where half of the votes went to the representatives of all political parties created after the regime change (2 representatives for each party) and the other half remained in the hands of the members of the members of the Council of the National Salvation Front.

[25] In the first post-Communist Parliament the Front won 66% of the seats, while the other parties won one-digit percentages (the Democratic Union of the Hungarians from Romania won 7% and the National Liberal Party 6% of the seats). On post-1989 Romanian elections see Campeanu, (1993).

[26] Actually the people who found themselves in the local Communist Party headquarters hardly knew each other.

[27] Kiraly Karoly's political career did not stop at the county level. A few days after his election, on December 26, 1989 he was appointed vice-president of the Council of the National Salvation Front, the second position in the hierarchy of the country's political structure. He retained his position as the head of Mures county; however, the deputy-president, a Romanian army general (in reserve), filled his position in the interim.

[28] Elod, (1992.) p.35

[29] At the time the FSN seems to have been considering a wide array of possible future political alliances and UDMR was a possible option. Later on, the front moved toward a more nationalistic stance. Nonetheless, to this day, UDMR remains an important negotiating coalition member in the Romanian political system.

[30] See Appendix 1 - " Statistical data on the management structure of Mures county's institutions, according to the ethnic criterion."

[31] One of my informants recalls: "I know it for sure that the first bilingual signs had been posted immediately after Christmas 1989 and nobody cared about them. I mean, then, during the first days. I do not remember how long it took. One week, perhaps two?" (Gheorghe, male, 39, Romanian, journalist).

[32] Of course, not only the Hungarians organized along ethnic lines. Other ethnic groups, (Gypsies, Germans, etc.) created their own civic structures.

[33] "In the organization of the Mures county youth there are three main organizations: The Democratic Organisation of Youth, The Organisation of the Young Volunteers and the MADISZ [the Association of the Hungarian Youth of Targu-Mures]...We must preserve the unity that held the young people together in the fire of the revolution. The elimination of suspicions, of chauvinism of any kind (...) are and should be a common cause." Cuvantul Liber, January 11, 1990, p.2. It was the birth of MADISZ, which was not a generic youth organization, as the other two, but one restricted to the young Hungarians, that brought in the term 'chauvinism' here.

[34] Issues that were high on the agenda of national youth media during those months.

[35] "They received a lot of aid from abroad. We -as Romanians- did not receive anything.. In '90 I worked at a warehouse where the humanitarian aid was deposited. I was ...the only Romanian [working there] and I can state it" (Traian, male, 45, Romanian, driver).

[36] The idea was inspired from the XIXth century Romanian cultural associations (such as ASTRA) that successfully fought against the assimilatory policies of the Hungarian government of the time.

[37] Cuvantul Liber January 25, 1990, p.4

[38] Cuvantul Liber, January 25, 1990. p.4

[39] Elod, (1992.) p. 53-54

[40] Tuesday, 30 January Cuvantul Liber, "Chemarea la unitate," (The call for unity), p.1. The article mentions that the initiative was a direct response to the decision to create a Hungarian-only high school in Targu-Mures.

[41] The first four paragraphs from the "Declaration-Program" of the Vatra Romaneasca union state:

"a) the union is not a political party, b) the union demands to be consulted by the government in any issue that concerns the Romanians from Transylvania, c) The union intends to defend Romania's territorial integrity, d) The union aims to defend the rights of the Transylvanian Romanians". (Declaratia Program a Uniunii Vatra Romaneasca, Cuvantul Liber, February 6, 1990, p.3)

[42] Of course, this suggestion was never put into practice.

[43] As one of my informants recalls: "At that time [Radio] Timisoara and Bucharest [which were concerned with non-ethnic issues connected with the democratic "transition"] occupied... 90% of my interest in politics, so to speak" (Petre, Male, 23, half-Romanian, half-Hungarian, engineering student).

[44] The role and the responsibility of the political entrepreneurs in the initiation and organization of the March19-20 are difficult to assess. There was no judicial procedure aiming to establish the truth, and those involved are reluctant (at best) to provide information. In the following pages I will limit my account to the data I was able to gather.

[45] Three months later, the capital Bucharest was invaded, in a strikingly similar fashion, by a much larger contingent of miners from a remote mining area, Valea Jiului. The city center had been occupied for several months by a non-stop demonstration of the student leagues and civic associations, demanding for the resignation of the president Iliescu. The misinformed and manipulated miners stormed the University of Bucharest and the city, beating students and innocent bystanders. See Berindei, Combes, Planche, (1990).

[46] According to some of my informants, among the thousands of Hungarians who gathered there to protest peacefully there were some who carried home-made weapons (such as chains and metal balls).

[47] In several cases the people that live in the blockhouses near the main theater of the battle instituted mixed teams to guard the entries. They refused the access of the 'warriors' of both camps, speaking Romanian or Hungarian according to which side was holding that area for the time being. Some people watched the events from the height of their balcony, together with their Hungarian/Romanian neighbors.

[48] A Romanian informant provides another instance in which the same feedback mechanism is implicit: "My grandma, my parents told me who those guys are [the Hungarians.] yet I could not believe until in March, when I saw it with my own eyes. Only then I realized that my family was not talking non-sense: indeed they were right" ( Ion, male, Romanian, 21, truck driver).