1. Collectivization began in Romania as early as 1948, in the form of “associations” (intovărăşiri). The drive to collectivize fluctuated during the next several years and was pursued with vigor only from 1957. Collectivization was declared complete across the country in 1962 but had been concluded in many villages well before this date. In the text to follow I use the date 1959, when the collective farm in Aurel Vlaicu (the community of my research) was officially inaugurated.

2. State farms (IASs) were run as state enterprises with salaried labor and an appointed director; neither employees nor the director had any necessary relation to the land the farm worked. Much state farm land came from expropriating large landowners, political prisoners or “war criminals.” Collective farms (CAPs), by contrast, were formed from “voluntary” donations of land by villagers who thereby became their members and their labor force. Members were not paid a fixed salary but various forms of remuneration in cash and kind; the rates were determined variously at various points in time. Although a CAP might have members who had given no land to it, the bulk of its members—and quite possibly its president—would have close ties to the land they worked since it had once been theirs. IASs tended to be favored in the distribution of state resources and to specialize in capital-rather than labor-intensive farming; CAPs did both, the latter particularly in the form of members’ “private plots.”

As of 1980, 61% of Romania's agricultural land was in collective farms (including private plots), 30% in state institutions and 9% in individual private farms. If we look at arable land rather than all agricultural land, the figures are 74%, 21% and 5% ( David, Turnock, The Romanian Economy in the Twentieth Century [New York: St. Martin's Press, 1986], 184 Google Scholar).

3. I thank M.-R. Trouillot for a stimulating discussion that informs my own here.

4. Information obtained from discussions with various specialists and citizens of the respective countries. To date, there is little published literature on decollectivization. Frederic Pryor's The Red and the Green: The Rise and Fall of Collectivized Agriculture in Marxist Regimes (Princeton: Princeton University Press, 1992) is chiefly a history, not about post-revolutionary achievements. Papers by scholars such as Kideckel, Hann and Creed helpfully discuss the earliest moments of decollectivization in Romania, Bulgaria and Hungary, but in relatively brief compass; somewhat more detailed is Agócs and Agócs. See David Kideckel, “Once Again, the Land: Decollectivization and Social Conflict in Rural Romania,” and Christopher, Hann, “Property Relations in the New Eastern Europe: The Case of Specialist Cooperatives in Hungary,” both in The Curtain Rises: Rethinking Culture, Ideology, and the State in Eastern Europe, eds. DeSoto, Hermine G. and Anderson, David G. (Atlantic Highlands: Humanities Press, 1993 Google Scholar; Gerald, Creed, “An Old Song in a New Voice: Decollectivization in Bulgaria,” in East-Central European Communities: The Struggle for Balance in Turbulent Times, ed. Kideckel, David (Boulder: Westview, forthcoming)Google Scholar; Peter Agócs and Sándor Agócs, ‘ “The Change Was But an Unfulfilled Promise': Agriculture and the Rural Population in Post-Communist Hungary,” East European Politics and Societies 8 (Winter 1994).

5. This provision was not included in Law 18 but was passed later. Because of it, there has been much less argument in Romania than in neighboring Hungary over how to ensure that people receive land of a quality comparable to their previous holdings.

6. These qualifications apply to land distribution in Armenia and Albania, for instance.

7. —most conspicuously the head of the National Peasant party, Cornelius Coposu.

8. Stelian Tănase, “Simple note,” Sfera politicii 1 1 (November 1993): 4.

9. The idea of property restitution was obviously important to the historical parties for not only its moral justice but also its potential to create a material base for the parties’ members. For some who advocated it, it may also have been motivated by revenge—by the desire to destroy all that had happened under the communists, who had ruined these people's lives and fortunes.

10. According to the 1948 agricultural census, 6.6% of all farms exceeded 10 hectares, but the figures do not indicate how much land they contained (see Henry L., Roberts, Rumania: Political Problems of an Agrarian State [New Haven: Yale University Press, 1951], 374 Google Scholar). I am uncertain how the 10-hectare figure was chosen. One participant in the process said it came from calculating how much land would have to be made available so as to give something to landless village residents. There may also have been other kinds of political calculation, however.

11. I am indebted to Sorin Antohi for this observation. Antohi is convinced that this was indeed the aim of the 10-hectare limit; I believe its determination was more complex.

12. The names of Romania's main parties changed several times in confusing ways. The initial National Salvation Front split into two parlies, the splinter retaining the old name and the other faction becoming the Democratic National Salvation Front. Each then changed its name again, the former becoming the Democratic Party and the latter the Party for Romanian Social Democracy (confusingly close to the Social Democratic Party of Romania), of which President Iliescu is a member.

13. I had spent several days in the village during each of the preceding three summers as well. For a social history of this community, known before 1926 as Bintinp, see my Transylvanian Villagers: Three Centuries of Political, Economic, and Ethnic Change (Berkeley: University of California Press, 1983). The name of the village is pronounced vlAiku; its inhabitants are Vlaiceni (vlaichEni); the commune (Geoagiu) is djoAdju. According to commune statistics, Vlaicu's population in 1992 was 915, in approximately 270 households.

14. Other reasons why my discussion does not apply readily to all of Romania are Transylvania's higher population density, which means there is less arable land available to resolve conflicting claims, and the longer history of peasant ownership there than elsewhere—that is, Transylvanians’ sense of private ownership was more highly developed. Elsewhere in the country, peasants became proprietors in large numbers only after the land reform of 1921 broke up latifundia and distributed them to villagers.

Vlaicu, for its part, is not wholly “typical” of Transylvanian villages—as an uneasy county official hastened to remind me. It is multi-ethnic and has a high percentage (about a third) of inmigrants, and its location subjected it to more than the normal number of border adjustments. Believing, however, that atypical cases are especially revealing of social processes, I am not daunted by these aberrations but find Vlaicu an excellent vantage point for examining property restitution.

15. A land-rental law passed in 1994 alters the situation, enabling people with land in IASs to repossess it after five years if they wish.

16. This was in 1993; in 1994 the figure was to be raised to 600 kg./share. The wheat prices used are those set by the state, not those of the free market.

17. See the discussion below on the property regime of socialism.

18. Once people had submitted their petitions for land and been awarded an amount, the land commission measured this amount for them in the appropriate number of fields and gave them an affidavit stating how much they had received. This affidavit is a provisional document; unlike a property deed, it cannot be used as collateral with a bank nor as the basis for selling one's land. The amounts on affidavits are subject to revision if others challenge allotments in court. Because the affidavit gives only a person's total land area and does not state where the land is or in how many pieces, people who do not realize that some of it may count as IAS dividends (rather than as actual workable surface) can be genuinely confused about what they are turning over to the association.

19. All across Romania, agricultural associations were formed following liquidation of the collective farms. Their raison d'ětre was that people who received land from the CAP did not also receive farm implements, hence most of them were unable to work their new acquisitions. The task of the associations is to arrange for the cultivation of its members’ lands, paying the members a percentage of the yield for each season. Far from all households give their land to the association: those who have managed to buy tractors or who find themselves allergic to the idea of communal farming arrange to work it themselves. In Vlaicu, about half of all landowning families have at least some land in the association.

20. I heard rumors that some people in the governing council of the association were engaged in a comparable scam: they would tell the association that they were giving it their land, then assign the same piece to someone in a sharecropping arrangement and collect produce from both sources. This scam and the one discussed in my text could work only because of the chaotic state of property restitution at the time: because the association rarely knew with precision where fields were that had been turned over to them, if they found someone working land they thought was theirs they might simply move over a few meters and proceed.

21. For helpful discussions of this process in other socialist contexts, see István, Rév, “The Advantages of Being Atomized,” Dissent 34 (1987): 335–50Google Scholar; and Vivienne, Shue, “Taxation, ‘Hidden Land, ’ and the Chinese Peasant,” Peasant Studies Newsletter 3 (1974): 1–12.Google Scholar

22. None of my informants could recall exactly the percentages taken, but all remembered that there was a big break around 4 ha. For some figures on the magnitude of quotas in Hungary, see Martha, Lampland, The Object of Labor: the Commodification of Agrarian Life in Socialist Hungary (Chicago: University of Chicago Press, in press)Google Scholar.

23. Cf. Rév, op. cit.

24. Gail Kligman, who has worked extensively with Romanian population statistics, finds a large number of similarly “elastic” categories there.

25. See, for example, my “Theorizing Socialism: A Prologue to the ‘Transition, '” American Ethnologist 18 (1991).

26. I use the word “yoke” to translate the Romanian iug ăr [cf. jug, yoke]. This form of measurement is widespread throughout the world, indicating the amount of land one can reasonably expect to plow in a day with a yoke of oxen.

27. Martinovici, C. and Istrati, N., Dicionarul Transilvaniei, Banatului şi celorlate ţinuturi alipite (Cluj: Ardealul, 1921 Google Scholar. Data in this work are based on the 1910 Hungarian and the 1920 Romanian censuses.

28. One was the nominal listing of individuals with their land, animals and agricultural implements, known as the Borderoul Agricol de Produc ţie, or BAP. I consulted this listing in the state archives in Bucharest (Institutul Central de Statistică, Recensămîntul agricol şi al populated din ianuarie 1948). The second listing is in the state archive branch in Deva, Hunedoara county ( “RecapitulaUa generală a tarlalelor din hotarul comunei Aurel Vlaicu.” Fond Camera Agricola Deva, dos. 49/1948, Aurel Vlaicu). It shows fields, rather than individuals; each field is listed with the persons who hold parcels in it and the cultigens planted on each parcel.

29. These figures appear in an unnamed file kept in the offices of the former CAP in Vlaicu.

30. From the 1988 cadastre for Vlaicu, kept in and used by the OCOTA office in the county capital (Registrul cadastral al parcelelor, Aurel Vlaicu, 1988).

31. This figure is from a file in the commune archives, dated 1 January 1990; the file is entitled “Situatii Centralizate Rapoarte [sic] Fond Funciar: Documente in Sprijinul Aplicării Legii 18/1991.” Various tables in this file give contradictory figures for the amounts of land brought into the collective. While I was copying these figures, a lively discussion went on around me as to how unreliable they were.

32. The figures 599 and 759 hectares are from the topographer in charge of measuring and redistributing the land in Vlaicu; the 631-hectare figure is from the source in note 31.

33. The motive for this “gift,” I was told, had been Gelmar's inclusion in the collective of neighboring Geoagiu, a village with very little and very poor-quality land. In order to increase the prospects for its success, the fertile soils of Vlaicu's floodplain were included in the production plans for Gelmar.

34. Source as in note 31.

35. “Se ‘evaporă’ hectarele?” Expres 31 (9–15 August 1994): 10.

36. See Daniela, Păunescu, Drept cooperatist (Bucureşti: Universitatea din Bucuresti, Facultatea de Drept, 1974), 46 Google Scholar. This “volitional” character was in most cases, of course, a fiction: the coercion applied to get people into the collectives was a favorite topic of village discussion even before 1989. I note that in Romanian, collective farms are referred to as cooperative (cooperatives), not collectives; I prefer the latter since it is more commonly used in English.

37. Ibid., 161.

38. Ibid., 56, 83. Local officials observed this difference in practice quite variably but recognition of it was not wholly absent. For example, I asked one former mayor of Geoagiu whether he had observed any distinction between state and collective farms during his tenure as mayor; he replied that he had never considered that he could impose solutions on the collectives as he could on the state farms. Even if this answer does not conform with his actual behavior, it indicates his awareness of the distinction mentioned in my text.

39. Ibid., 127.

40. Indeed, when I discussed this problem with a member of the local land commission, he acknowledged that he had not realized the implications of the housegarden problem.

41. The law specifies that undisturbed use of a piece of land for a certain length of time entitles its user to claim it as owner. I am uncertain what that length of time is: the two judges with whom I discussed this matter gave me two different figures, 20 years and 30 years.

42. The decree specified that the exchanges were “obligatory for all owners whose lands are subject to commassation” (Republica Populară Rominia, Legislaţia civilă uzuală II [Bucureşti: Ed. Ştiinţifică, 1956], 101).

43. Ibid., 102.

44. Those who joined the early associations and collectives were usually the poorer peasants rather than the better off.

45. The money paid by most state farms to their “shareholders” is much less than the income a person could receive by working the land, although the shareholding arrangement does have the advantage of getting one's land worked for one.

46. Despite considerable effort, I was unable to find out how big this property is—as was the woman who was trying to claim restitution of a part of it. A crucial file from the Vlaicu Land Register is missing; moreover, because the estate straddles the border with Gelmar, some of its area would be listed only in the Land Register for that village, to which I did not have access.

47. Their insistence on the former sites precluded rationalizing the property structure through a new commassation, however. In Cigmău, the topographer told me, the average size of newly established holdings is 3–4 hectares; but the average number of parcels per holding is 20–30 and the average parcel size thus no bigger than the old “private plot” of .15 ha.!

48. How can the claims be insatiable? Law 18 specified that land could be claimed by any relative up to the fourth degree. This means that cousins, aunts and uncles, as well as the original owners plus their children have the right to claim land. Many of these people would not have received it had normal inheritance practices prevailed; their claims thus reduce the surface area that would have entered into a “reserve” for distribution to the landless.

49. Party considerations seem to have played no role in this. Geoagiu's mayor and vice-mayor as of 1993 were from the Democratic Agrarian Party, whose members are largely the managers and technical staff of state and collective farms, not village farmers.

50. I never witnessed bribe-taking but villagers allege that it occurs. See the discussion below on local-level abuse of power, however, for arguments that might call this allegation into question.

51. In a neighboring village I found a much more rigorous procedure: the commission met in a room into which one person at a time would enter, the door would be locked to prevent others from coming in and the complaint would be pursued to resolution. Clearly the modus operandi of commissions varies widely, a function of the probity and degree of organization of their members.

52. In Hunedoara county the figure was 28%. Figures by personal communication from the Romanian Ministry of Agriculture.

53. I was unable to determine exactly how many court cases involving land were pending in the courts. The leader of the main opposition group in Romanian politics, Emil Constantinescu, used the figure of 400,000 in a May 1994 speech on television, but the paper Expres published a lower figure of 300, 000 (19–25 October 1994, 10).

54. The land of each village is divided into a number of named fields and subfields. As I was able to confirm from maps dating from the 1880s, many of these names are of at least a century's duration. For Vlaicu, most of them appear to be locative, rather than invoking personal names: thus, “At the pear trees,” “Behind the church,” “The Vinerea road,” “Where the parcels are long,” “Where the parcels are short,” etc. A few invoke persons: “At Barcsay [Băiceana, from Barscay, a noble Hungarian family living in the village in the nineteenth century]” or “At the ford by the priest's [Vadul popii].” For many of these names, no one I asked could explain the meaning.

55. Official measurement was usually taken in yokes until the communist regime came to power and began to use measurements in hectares. Villagers are genuinely confused about what measure applies, having become accustomed to “hectares” and not realizing that a different measure applied when their parents were property owners. Since a yoke is slightly more than half a hectare (.5757), the difference is considerable.

56. A few of these people spend the winters in town with relatives. Urban Romanian Germans also can claim land, and do.

57. See Verdery, op. cit., chap. 5.

58. In the early phase of property restitution, when many villages were forming associations to work the land received, it was common to give association members their land in a single block for ease and efficiency of cultivation. Vlaicu's Germans’ request for a single block thus conformed with practice up that point. One need not suspect, as Vlaiceni do, that the mayor was bribed to give them this large field.

59. Andrei Cornea, “Directocraţia remaniază guvernul,” 22 5 (16–22 March 1994): 7.

60. I know the party affiliations of only the second with certainty (he is in the Democratic Agrarian Party). I did not note those of the other two.

61. See Jadwiga Staniszkis, The Dynamics of the Breakthrough in Eastern Europe (Berkeley: University of California Press, 1991) for a discussion of how the inability of higherlevel party authorities to “deliver the goods” to those beneath them contributed to undermining party rule.

62. Gail Kligman (personal communication) points out that in this respect as in several others, the situation I report is not universal. In the northern Transylvanian region of Maramures, for instance, long-term labor migration by men gave women almost complete responsibility for running family farms. Kligman doubts that Maramures women would let themselves be pushed around as some of Vlaicu's widows do.

63. Another group likely to be cheated consists of people without social connections or people who are not themselves valuable as connections. Those least likely to be cheated are powerful persons who could cause trouble for a local commission that trod on their rights. Even this is no guarantee, however, for a judge I know told me that she couldn't seem to get four hectares belonging to her family.

64. My term “local officials” here includes officers of village associations, who are among the chief beneficiaries of the land obtained by urbanites. Not all such officers will handle the land unscrupulously but urbanites are the association members most easily cheated if one is bent on doing so. As is clear from my text, the most common form of abuse is to underpay the landowner for use of the land. Commune mayors are more likely to abuse the rights of urban-based proprietors by assigning the amount of their land to a state farm, which will pay them little or nothing, and working the actual land themselves.

65. The data concern three sorts of conflicts: between Vlaicu's Germans and the Romanian farmers impropriated on their land, among Romanians who are all locally born and gave land to the CAP, and between locals and the inmigrants who worked in the CAP but gave it no land. My sources are my discussions with people and their responses to questions about whose ownership claim ought to prevail, volunteered opinions on this topic and arguments that I overheard as I followed the land commission on its rounds.

66. Lampland, op. cit.

67. For a further discussion of this sense of entitlement, see my “From Parent-State to Family Patriarchs: Gender and Nation in Contemporary Eastern Europe.” East European Politics and Societies 8 (Spring 1994): 225–56.

68. Special thanks to Ashraf Ghani for the ideas in this paragraph.