• 1 Very few researchers, however, have gained entry to the security services' archive (AFSBRF). The Ru (...)

5Meanwhile progressive press organs like the weeklies Argumenty ifakty and Ogonek featured readers' letters, memoirs and other material on the Stalinist « repressions » (a vague but convenient term). In several localities civic organizations sprang up. Especially significant were branches of the « Memorial » society. Its members sought to honour the dead by erecting monuments at the sites of former prison camps, holding documentary exhibitions, and generally collecting as much relevant information as possible. Excavations undertaken in several places, notably Kuropaty in Belarus, uncovered mass graves containing thousands of human remains. These activities have continued since the Soviet Union's collapse in 1991, despite adverse conditions. Funds are scarcer than ever, due to the economic malaise, and officials often less supportive than they should be. On the other hand archive access has been liberalized1.

• 2 Neue Zürcher Zeitung, 15 January 1997, p. 43.

6Public interest has likewise waned. Most citizens are preoccupied with their immediate needs. Many are nostalgic for a Soviet past that seemed to offer greater political stability and better material conditions. They are not in a mood to assume their moral responsibility by trying to come to terms with a past that now lies far distant. The few living Gulag survivors are in their eighties or nineties. Although practically every family in the land suffered to some degree from Stalin's terror, memories of it have been subsumed into those of World War II, which have consistently received far more media attention. People would like to forget all these associated horrors. As Sonia Margolina notes, « there is no social need for a differentiated analysis of history. Men and women living between deprivation and the mafia have no tomorrow to believe in. And whoever takes no thought for the morrow needs no yesterday »2.

• 3 Russian-language journals are abbreviated as follows : Ist. = Istochnik, (suplement to Rodina) ;NNI (...)
• 4 Yampolsky et al.(1995) ; Nekrasov (1995).
• 5 Sbornik zakonodatel'nykh i normativnykh aktov o repressiiakh i reabilitatsii zhertv politicheskoi r (...)
• 6 Po materialam sledstvennykh del i lagernykh otchetov Gulaga, Moscow 1993 ; Tak eto bylo : natsio-na (...)
• 7 Yakovlev et al.(1991, 1994) ; also his essay (1993) and especially his report on the treatment of r (...)
• 8 Bugai (1995), cf his articles in Voprosy istorii, 1990,7 and Otechestvennaia istoriia, 1992,4.

8Let us first construct a typology of the recent literature3. In the Russian Federation - books and articles appearing in Ukraine and other ex-Soviet republics cannot be treated here - some works are apparently the fruit of a partnership between security officials and « independent » scholars - a practice that, given Soviet antecedents, should come as no surprise. The first volume of a history of the « organs of state security » during World War II (1995) bears the FSB's imprimatur, whereas V. Nekrasov's history of the Interior ministry (NKVD/MVD) is apparently an unofficial study4. Neither of these concentrates on the Gulag, but a compendium of relevant legislation appeared in 1993, and the military historian O.F. Suvenirov has contributed a brief study of the para-judicial organ that settled the fate of nearly 50 000 victims of the Great Terror5. In 1993 there appeared a collection of essays written « on the basis of investigatory materials and Gulag camp reports » and a three-volume collection on « repressed » nationalities6. Yakovlev himself has followed up a book on the rehabilitation process (1991) with some general reflections under the title A Bitter Cup7. Another prolific writer, N.F. Bugai, heads an academic study group on the deported minorities ; one of his documentary collections is available in English translation8. In general the « special settlers » (exiles) have attracted more attention, notably from V.N. Zemskov, than the inmates of camps and colonies - a point we shall return to later.

• 9 Belenkin et al.(n.d.).
• 10 Danilov, Krasilnikov (1992), reviewed by I.Ye. Zelenin in Otechestvennaia istoriia, l996, 5, p. 195 (...)
• 11 Striuchenko et al. (1994) ; Ylantsev (1992,1995).

9Significantly, some of the works just noted appeared in provincial centres, and it is to such publications that we must turn for studies of particular regions. Those put out by « Memorial » branches are usually printed in small editions and rarely reach Western libraries9. The best work has been done on western Siberia, by the wellknown agrarian historian V.p. Danilov and an associate. Their three-volume study of special settlers in this region, published in 1992-4, runs to over 900 pages ; it reproduces the relevant orders (also those of local provenance !) and subordinates' reports on their fulfilment10. In the Far East, where the NKVD ran the notoriously homicidal Dalstroy camps, I.G. Striuchenko and his colleagues in Vladivostok are editing a multi-volume work to remind people of « forgotten names », while O.p. Yelantsev has written two articles on railway-building projects constructed by forced labourers in the region11.

• 12 Slavko (1995) ; Pervukhina (1994) ; there are also two useful articles by Plotnikov (1994, 1995).
• 13 Morozov, Rogachev (1995) ; Shashkov (1996) ; Kaneva (1991) ; on the topic Shashkov has written a di (...)
• 14 Sanukov (1996) ; Litvin (1994) and a paper for the ICCEES congress in Warsaw (1995), to be publishe (...)

10Moving back west to the Urals, the fate of the so-called « kulaks »(actually, peasant householders of all kinds and their families !) who were deported there in the early 1930s is analysed in a monograph by T.I. Slavko and a volume of essays edited (locally, in Yekaterinburg) by V.I. Pervukhina12. Further to the north-west lay the multiple camp complexes situated between Kotlas and Vorkuta (Pechora valley, Komi ASSR), about which much can be learned from two all too brief articles, one by N.A. Morozov and M.B. Rogachev and the other by V.Ya. Shashkov, as well as the memoirs of A.N. Kaneva13. For the « repression » of the non-Russian minorities of the middle Volga, the English-speaking reader now has an excellent study by K. Sanukov of the « purge »in the Mari republic, while the Kazan' historian A.L. Litvin deals expertly, in several articles, with the Tatars14.

• 15 Razgon (1991).

11The list could be extended, but enough has been said to show that local historians are producing rich fruit. Such works are often more explicit, and more humane in approach, than studies of the Gulag phenomenon in its entirety. The same applies of course to memoirs, such as those by the writer L.E. Razgon, Nepridumannoe (1987/8), now available in major Western languages15. This is of necessity a genre in decline, but the well-known accounts published earlier, notably those by E. Ginzburg or V. Shalamov, have not lost their relevance. So far as is known, none of the thousands of officials, guards and so on who ran the penal system has given us his side of the story - for reasons that are obvious.

• 16 Stettner (1996).
• 17 Getty, Rittersporn, Zemskov (1993), and ensuing correspondence.
• 18 Bacon (1994), which develops earlier studies in Soviet Studies,1992, 44, p. 1069-1086 and Revue d'é (...)
• 19 Rosefielde (1996) ; Wheatcroft (1996).
• 20 Jakobson (1993).
• 21 Hochschild (1994).

12In the West historiography of the Gulag has recently entered a qualitatively new phase with the appearance of R. Stettner's comprehensive study16. This synthesizes the vast amount of material previously published in the West as well as much of the post-Soviet literature. Its author has not, however, had access to the archival sources on which the latter studies are based and omits the special settlers. He also foregoes sociological analysis of the convict population, but offers, as the title indicates, a thorough account of their conditions and the way the system worked. In an appendix Stettner reviews the « numbers controversy » which continues to rage among Anglo-Saxon historians, but without offering a definitive answer of his own. For this the time is not yet ripe, since we lack the detailed information (still in FSB hands !) that could show how the figures on population, mortality etc., which Zemskov and two Western « revisionist » partners have put forward with such confidence, were obtained17. Also sceptical in this regard is E. Bacon, who did use some archival sources for his book on the Gulag in World War II18 ; predictably this has been warmly welcomed by some historians, such as R. Conquest, but has encountered a cool reception from others. In 1996 Stephen Wheatcroft, for the revisionists, and Steven Rosefielde, for the traditionalists, dealt yet another round in their long drawn-out battle of statistical computations19. M. Jakobson has written a scholarly account of the way Stalin transformed the Soviet penal system, suppressing its early reformatory features20, while the journalist Adam Hochschild, a sensitive and thoughtful observer, has recorded impressions of a visit to the USSR in 1991 when he was able to inspect several Gulag sites and talk to former prisoners, Memorial activists and others21.

• 22 Karner (1995) ; Bonwetsch (1993).
• 23 For a preliminary account : Dahlmann, Linden, Scheide (1995).

13Anglo-American writers sometimes neglect scholarship in German, which is making a significant contribution. In addition to Stettner's book just mentioned, there is an important study by S. Karner of the fate of German prisoners of war22. They came under a different part of the security establishment, GUPVI, which was independent of GULag. Also relevant are the proceedings of a conference held at Mülheim (Ruhr) in 1995, due to appear shortly, at which the participants boldly sought to compare the Stalinist terror with that of the Nazis23. The field is growing exponentially as new documentation becomes available.

• 24 Razgon (1991, p. 117).

14We may now attempt to summarize what has been learned so far. Perhaps the most significant finding relates to the categorization of prisoners - although we have to remember that in a system that was inherently arbitrary it would be misleading to expect legalistic precision, so that there are exceptions to every generalization one makes. What might be called the « slave class » consisted of two major groups : a) inmates of forced-labour camps (ITL) and colonies (ITК) and b) exiles, i.e. settlers or « special settlers ». The camps were of three degrees of ascending severity ;the colonies were designed for short-term offenders (in principle, up to three years) and theoretically life there was more bearable, although in practice the distinction might not matter much. As Razgon remarks, « l'essentiel n'est pas le règlement mais le geôlier »24, so that in bad periods colony inmates, too, suffered high mortality due to starvation and overwork. There were several other categories of camp : those for forced labour (katorga) from 1943 onward, and the « special camps » set up in 1948 ; furthermore the « filtration camps » for repatriates with their subordinate « points ». Such « points » really correspond to the English term « camp » ; these were grouped in larger entities best referred to as « complexes ». Several sources refer to the existence of execution camps, but these units were apparently as a rule concealed within a particular sub-system and did not comprise a category of their own. Apart from all this, there were people in jail or en route from one installation to another.

• 25 Zemskov (1995a).
• 26 Razgon (1991, p. 121). Note especially the case of Colonel Vypolzov at Ustvymlag (Ukhta), who used (...)

18Possessed of absolute power, Gulag administrators at all levels had every incentive to exploit the convicts' labour to the full in order to meet the production targets they were allocated, but precious little incentive to care for their charges' well-being. Guards were forbidden to socialize with the zeks (as prisoners called themselves). They were trained to look on them contemptuously as human flotsam, as anonymous « labour power » (rabsila). This attitude alone requires one to be cautious about accepting these officials' own estimates of the size of the convict population. Zemskov has recently claimed that they had no reason either to inflate or minimize the numbers25. This is a dubious assertion ; for if in the command economy industrial managers regularly falsified data on stocks, output and the like, as is now generally acknowledged, why should not their NKVD/MVD counterparts - fearful, ambitious and lacking administrative skills - have behaved likewise ? Many of them were, after all, so-called pridurki : prisoners, in the main common-law offenders with a criminal background, who had been co-opted to assist with the office work -a job that enabled them to escape physically demanding and dangerous labour « outside » on construction projects. Razgon describes eloquently the cheating that went on over the allocation of jobs, rations and the like26. The more conscientious bureaucrats were better at recording the movement of vehicles or materials than those of human beings. Typically, when reporting the transport of people eastward, they would pedantically note the number of railway waggons used, and so conceal the loss of prisoners en route.

• 27 Zemskov (1989).
• 28 Wheatcroft (1996, p. 1331, 1336).

19Curiously, there was no central agency to co-ordinate statistics on the number of prisoners reported by subordinates. This is another reason why doubts persist about the veracity of the totals supplied in 1953/4 by the MVD's First Special Department to the post-Stalin Party leadership when the latter requested it after Beria's elimination. The original draft (December 1953) included 282 926 persons convicted for anti-state offences under articles in the code other than Article 58, who were not mentioned in the better-known February 1954 document signed by the security chief, Kruglov, and two other senior officials. This stated that, over the entire period since 1921, the security organs had sentenced 642,980 individuals to death, 2 369 220 to camps (and colonies ?) or jails, and 765 180 to (ordinary) exile27. This document was not made public until the Gorbachev era, and the figures are clearly under-stated. Wheatcroft proceeds from an assumption of about 1 million put to death in the late 1930s, but reproduces without criticism the figures for those despatched to the Gulag28.

• 29 Popov (1992).
• 30 Ibid., 23. A Ministry of Justice report for civil court sentences in 1952-1953 gives other figures, (...)
• 31 Bacon (1994, p. 27, 36).

20In 1992 Vp. Popov came up with other figures for sentences by the security authorities alone (from 7929) that were higher for those sent to camps or jails (2 530 654) but lower for ordinary exiles (362 041)29. He also gave figures for civil court sentences, but only for a few years, 1940 and 1945-5330. These show that, during these ten years, a further 5 949 370 persons lost their liberty. This gives a total of 8 842 065, to which one would need to add an estimate of those given judicial sentences over the remaining 25 peace-time and 3 war-time years. One may agree with Bacon's educated guess that between 1934 and 1947 the net intake into the camps was 8.1 million, and if colonies are included 12 1/4 million. For the whole period to Stalin's death he suggests a figure of over 18 million31. This estimate is by no means excessive. We know from Popov's list that in 1948-53 the civil courts and the security authorities respectively passed custodial sentences on 3 275 336 plus 294 583 individuals. These figures apparently do not include that proportion of repatriates and repressed ethnic minorities who, as actual or potential security threats, landed up in camps as distinct from settlements.

• 32 Wheatcroft (1996, p. 1328-1329).
• 33 Zemskov (1995a, p. 110)

23The figures for the Gulag population on 1 January of successive years, though of interest in themselves, do not tell us how many entered or left the system during any given year. Wheatcroft helps us by adding figures for « mid-year » (1 July ?) as well32. From these data (whose accuracy, to repeat, is far from assured) it is clear that the initial influx into the camps, mainly peasant « kulaks » and ex- « bourgeois », during the First Five-year plan era was followed by another in the late 1930s, the years of the Terror, and another after 1944/5. There was a slight decline (2,3 per cent) between January 1936 and 1937, but then expansion continued more rapidly, so that when the Germans invaded the camps alone held 1 458 000 persons, an increase of 77,6 per cent on January 1937 (820 881). This is of course no guide to the actual number sent to camps during the « purges », since many died there - and some were released. These new entrants probably numbered about 4 million, although Zemskov acknowledges less than 1 million (1936-40), over 200 000 of whom entered in 193733. The third great influx, after war-time losses, brought the figure up from around 1,5 million in January 1945 - roughly half in camps and half in colonies - to the already cited 2,47 million (1,72 + 0,74 million) in January 1953, shortly before Stalin's death forestalled yet another increase at the expense of suspect « cosmopolitans » (mainly Jews). This is still a dark era in the history of the Gulag, as of other aspects of Soviet life, although the murk is slowly clearing in regard to, say, the post-war famine.

• 34 Stettner (1996, p. 283).
• 35 Zemskov (1995a, p. 119).
• 36 The ratio of colony to camp inmates 1935-1938, which he oddly calls 'approximately twice as few', i (...)
• 37 Getty etal. (1993).

24Much more work needs to be done on mortality rates in camps and colonies before we can arrive at plausible figures. One cannot really expect precise recordkeeping of deaths by officials who, as Stettner reminds us, systematically mutilated the corpses of deceased prisoners in order to prevent inmates escaping by feigning death34. By keeping « dead souls » on strength an administrator could hope to receive more rations (but also higher work norms). Zemskov claims « absolutely precise information » from an archival source that only 1 083 829ITL inmates died over the twenty-year period beginning 1 January 193435. No composite figure was found for ITK inmates but, extrapolating from their average ratio to camp inmates, he puts their mortality at « not more than 0,5 million », and adds in a figure of 86 582 for deaths in prison (1939-44, 1946-51) to arrive at a total estimate of 1,6 -1,7 million custodial deaths for 1935-53. The arithmetic of this computation is not beyond reproach36. In his joint article with two Western revisionist historians the estimate is raised to « a little more than 2.3 million », including settlers, but to 1940 only, when the worst was of course yet to come. The authors take refuge in evasion : « we have some reason to believe that the new numbers... are likely to bring us within a much narrower range of error than the estimates proposed by the majority of authors... »37. Perhaps : but we know that losses were nevertheless catastrophically high, especially during and just after the war.

• 38 Bacon (1994, p. 110).
• 39 Rosefielde multiplies the number of sentences to camp and prison by the number of years to be serve (...)
• 40 Litvin(1994).

26These are clearly understated. They probably exclude prisoners in transit, inmates of filtration camps and other categories as yet unascertained. From these figures the mortality rate can be reckoned as in column 2 (they are a little different if one takes the annual average of inmates, but this increases the margin of error). A.N. Dugin has put the death rate in 1942 and 1943 at 22.7 and 22.8 per cent respectively, which is probably nearer the truth ; Bacon has shown that in 1942, in camps working for the forestry department (ULLP), 34 per cent of those « lost » died, whereas in that for railway construction (GULZhDS) the rate was below 15 per cent ; in the Dalstroy camps it was 19 per cent (round figures)38. Wheatcroft offers a maximum and minimum range (col. 3 in Table 1), including in the former those who « disappeared in transit » or escaped without being recaptured. Rosefielde claims to have shown that the figures advanced by Zemskov and associates are internally inconsistent, although it is hard to accept his methodology39. He apparently adheres to an earlier estimate of an average camp mortality rate of 10 per cent per annum, and holds that 2.8 million died in detention in 1937-9 alone. The debate could be resolved only if we had new information from the papers of subordinate Gulag organs. The Tatarstan local archives contain data on the number sent to camps in certain years, but not about the losses within them40. It is clear that some regions, like Kolyma, were worse than others, and that political offenders were treated more harshly than common-law offenders.

• 41 This point is well made by Graziosi (1992) where she puts the proportion of 'politicals' in 1946 as (...)

27How large was the proportion in each category ? Scattered information suggests that roughly one-quarter to one-third were « politicals », but the ratio varied over the years, increasing spectacularly after 1944 and then falling again. We need to be careful about taking the official categories at face value. Some of those not indicted under Article 58 committed offences that were indirectly the result of the regime's repressive policies and would not normally be considered criminal, as when peasant women stole stalks of grain from the collective fields to feed their starving children41.

• 42 Getty et al.(1993, p. 1034).

28One could, for example, be sentenced for committing an offence described by the rubbery formula « socially dangerous (harmful) act », which meant more or less whatever non-conformist behaviour the local authorities were keen to suppress at the time. Zemskov and associates, who seek to minimize the number of political as against common-law offenders, claim that over the twenty years 1933-53 (omitting 1934) only about 8.4 per cent of sentences, whether by the courts or by extra-judicial bodies, were rendered for alleged political offences42. They even claim that in 1937-8, « when many believe the terror was mounting » (sic !), the share of « politicals » in the camps declined - but this decline, if indeed it occurred, can be explained by the fact that so many of those convicted were simply shot. In short, the nature of the charge in an accused's record is not a reliable guide as to the reason why he or she was repressed, and should not be taken too seriously by researchers. They should rather proceed from the assumption, difficult as it may be for outsiders to grasp, that the documentation in an individual case is fraudulent unless proved otherwise. Stalin's USSR was a society in which the criminals at the top determined who should be deemed guilty of what, and the job of court and police officials was to identify them, so filling their quotas. (Some, to their credit, resisted - and paid the penalty).

• 43 Plotnikov (1995).

29The newly accessible archives make it possible to attempt a sociological analysis of the Gulag population, but we cannot take this line of investigation here. Of all the criteria employed (age, gender, class and so on) that for ethnic allegiance is of greatest interest to the non-specialist. In what follows we shall be concerned mainly with the exiles, otherwise known as « special (or labour) settlers ». In the early years these were predominantly Russians or other eastern Slavs - and by social origin peasants. The recent studies of the Urals and north Russia provide graphic evidence of the way these « kulaks » were treated. Nation-wide the first wave (1930-1) is thought to have numbered at least 1.8 million, family members included. It is now clear that, since local officials were given little advance warning of the influx, even the most benign of them could do little to alleviate the newcomers' lot. Deported from their native regions en masse and in great haste, without proper provision for the journey, many died en route. In 1930-1 alone 571 000 such deportees arrived in the Urals. OGPU officials in Sverdlovsk reported (March 1931) that they were short of food and protective clothing essential to survive the harsh climate ; those with ready money found nothing to buy ; they had to do unfamiliar work in logging and were set impossibly high norms ; local Party cadres were subjecting alleged laggards to cruel punishment (e.g., making them dig what they were told would be their graves). All this prompted a revolt, hitherto unknown to historians, which was brutally suppressed43.

• 44 Shashkov (1996).
• 45 Fitzpatrick (1994).

30Shashkov, a scholar in Murmansk, tells us that by January 1933 the northern region had been required to accommodate 301 000 deportees (two-thirds of them by September 1930). He expresses sympathy for the victims - « humiliated, barefoot, hungry and cold » - but seems more concerned to hail their subsequent achievemerits in boosting output : sowing potatoes, building up herds of cattle and so on44. This is a characteristic line for a post-Soviet historian to take. So, too, is a reluctance even to estimate what the mortality rate may have been. According to Zemskov and others, a remarkably large number of exiled peasants managed to escape, evade recapture, and reach home, where they were often made unwelcome by fellow villagers who had meanwhile taken over their property45.

• 46 Ivanov (1996, p. 164), offers a figure of 110 000, which seems rather high.
• 47 Gelb (1993) ; Sanukov (1996).
• 48 Bugai (1992, 1994). A « White Book » on this affair was published in 1992 by V.A. Li and E.E. Kim.

31In 1933 the peasant settlers were joined by common-law offenders, after Stalin ordered the country's overcrowded jails to be cleared, and in 1935-6 the process began of shifting to the interior masses of non-Russians whose presence in the border regions made them seem a liability should war break out. The first such group appears to have been Poles living on the western frontier, who were moved to Kazakhstan46. Another comprised Karelian Finns (and Finns generally). The regime took the threat of Fenno-Ugrian nationalism seriously (or pretended to do so) and this served as the rationale for the severe repression meted out inter alia to the Mari (Cheremis) people on the Volga47. At the other end of the country, in the Soviet Far East, no less than 172 000 Koreans were rounded up and deported to Central Asia. Already en route they suffered much hardship ; as in the case of the « kulaks », no proper arrangements had been made to receive them ; and, although they had done nothing wrong, their status subsequently deteriorated until they came to be treated as « enemies of the people »48. At the outset of this operation, which served as a dress rehearsal for others better known in the West, 3000 Koreans were arrested as political suspects and presumably sent to the « inner Gulag », along with 8000 Chinese. This proportion was relatively modest by comparison with the fate of Poles and others in the territories annexed in September 1939, in accordance with the terms of the Nazi-Soviet pact.

• 49 Parsadanova (1989).
• 50 Wheatcroft (1996, p. 1344).
• 51 This figure is made up as follows : 4 421 in Katyn forest, Smolensk province : 3 820 in Starobelsk (...)

32VS. Parsadanova, the first Russian scholar to study this tragic episode, stated that 1.2 million people were deported, in several waves, and that 82 per cent of them became « special settlers », of whom 16 per cent died (presumably within the first two years)49. Wheatcroft states that 1.5 million were deported, of whom 0.6 million were Poles50. Several thousand officers and other former members of elite groups (« class enemies ») were taken to three internment camps. The « Katyn affair » is no longer the state secret that it was for so many years. In 1992 investigators came across one of Stalin's « special folders » revealing that the massacre was carried out on an express order by the Politburo dated 5 March 1940. The total number put to death by the NKVD was a greater than previously supposed : 2 1 85751.

• 52 Izvestiia TsKKPSS (1990,10, p. 129-139).
• 53 Misiunas, Taagepera (1983, p. 41).
• 54 Zemskov (1993).

33Many Jews resident in the annexed territories made their way eastward, voluntarily or otherwise, in 1939-41, which enabled some of them to escape the Nazi Holocaust. Whether this was Stalin's intention is unclear. Ukrainian and Belarusian deportees also seem to have fared better than Poles, but their lot was far from pleasant. They were distrusted and a proportion of them, as yet undetermined, ended their days in the Gulag. The same was true of inhabitants of northern Bukovina and Moldova (Bessarabia), seized from Romania in 1940. When the Baltic states fell into Stalin's grasp, they, too, were « sovietized » and in June 1941, on the eve of the German invasion, at least 16 000 Lithuanians trod the path of exile52. Emigre sources put the minimum number at 34 000, plus 10 000 Estonians and 15 000 Latvians53. According to Zemskov, a total of 25 586 Baits were exiled to settlements, where, as he puts it, « their numbers significantly diminished due mainly to an excess of deaths over births ». How many were shot or sent to camps he does not say. The figures for Baltic camp inmates on 1 January 1941 was 8896, and four years later 8129, but much happened in the interim and the post-war influx had yet to begin54.

• 55 J.T. Gross in Sword (1991, p. 50 ff.).
• 56 Beria to Stalin, 1 May 1944, in Bugai (1991, p. 153).
• 57 This figure is given as 119 855 by Taranova (1993), as cited by Wheatcroft (1996, p. 1345).

36Returning to 1941, when the Germans invaded, some inmates of camps in the western regions were shot while others had to trek east on a forced march that cost the lives of an estimated 10 per cent55. Stalin now had to change his policy toward Poles in the USSR. An amnesty was granted (12 August 1941), at which time, according to official data, there were 389 382 Polish citizens in the country, of whom 120 962 (31, 1 per cent) were in prison, camp or exile and 243 106 (62.4 per cent) in « special settlements » ; a further 25 314 were held as prisoners of war56. Of these 341 were excluded from the amnesty ; 76 100 combatants57, with 43 755 dependants, were allowed to leave via Iran to fight on in the Anders Army, while 36 510 were enrolled in the Polish forces set up on Soviet soil. Beria's figure for deaths in 1941-3 is only 11516 and not credible.

• 58 Kokurin(1993,p. 102).

37As the fortunes of war changed after Stalingrad, the settlements, which like the rest of the Gulag had suffered heavy population losses, began to fill up again as entire ethnic minorities, arbitrarily categorized as disloyal, were punished by deportation eastward. The Soviet Germans, from the dissolved Volga German ASSR and other parts of the country, had been sent there already in 1941. According to Chebrikov (1988), they numbered 815 000, although other sources suggest that there were more than 1 million of them. The next big batch comprised the north Caucasian peoples : Chechens, Ingush, Karachai and Balkars, arrested en masse in the winter of 1944. According to a 1990 study the original target figure was in the region of 320 000, but was then raised to 460 000 - and then over-fulfilled ! We now have day-to-day figures for the numbers of Chechens and Ingush (only) loaded into railway waggons, which add up to 478 479 ; by September 1944 this had risen to 496 000, perhaps because Red Army soldiers of these nationalities, who were arrested separately, were included58.

• 59 Bugai (1991, p. 152-153). Sachka (in Osteuropa, 1988, 11, p. A512-516) put the figure at 191,088 an (...)

38They were followed in May 1944 by the Crimean Tatars, who numbered 191 044 ; at least this is the figure reported (to Beria, presumably) by the two NKVD generals responsible for the operation. They stated smugly, » no excesses of any kind took place » - as if the action had not itself been one great excess ! Nearly 6000 Tatars were arrested as « anti-Soviet elements » and presumably shot or sent to camps. One wonders whether the proportion was comparable among the other Oriental groups. We also hear of a 5000-strong « special contingent » (presumably not the same as that just mentioned), and the gross figure of those sent east is given in the source as 180 014 - which may still be too low59.

• 61 Zemskov (1993, p. 5-9).

41Zemskov and Popov have estimated the gross post-war increase of exile population at over 600 000. The chief groups were those just mentioned from the western territories, for there was considerable resistance here to the reimposition of Soviet rule. By 1 January 1951 there were 75 646 Baits in camps and 20 507 in colonies. To this figure must be added about two-thirds as many in settlements, for the latter are known to have held 172 362 Baits two years later. They were located mainly in Siberia. About 28 per cent of these Baltic settlers were children, most of them born in captivity61.

• 62 Zemskov, in Argumenty ifakty,1989, 39, p. 8.

42The MGB was interested in the adult settlers' ethnic make-up and worked out data for 1 January 1953 which show that, taken together, the 139 957 Baits comprised 6,9 per cent of the total. Germans (exclusive of prisoners of war) were the largest group and numbered 1 224 931 (44,5 %) ; north Caucasians stood at 498 452 (18,1 %), and Tatars, Greeks and others from the Crimea 204 698 (7.4 %). Other significant groups comprised Kalmyks (3,0 %), Meskhetian Turks (1,8 %), Moldavians (1,3 %) and Armenians (0,6 %). Curiously, there were still 36 000 Poles from the first batch deported in 1936, or their descendants. Far more numerous were Ukrainian anti-Communists, classified as « OUNovtsy » or as « Banderites » or « Melnikovites »(after two guerilla leaders), at 175 063 (6,4 %). The mainly Russian « Vlasovites» - i.e. supporters of the Russian Liberation Army (ROA) - were only a third as large (56 746, 2,1 %) ; most of these men will have been shot or sent to camps62.

• 63 Zemskov (1995b, 5, p. 9-10).
• 64 Yakovlev et al. (1996).

44In 1990 Zemskov put the number of those repatriated at 4,2 million. More recently he has offered a higher figure : 5 352 963 (to January 1946 only)63. Of these 1 153 405 were « internally displaced persons » who, not having left Soviet territory, were technically not » repatriates ». Of the latter, 1,5 million were armed forces personnel and the rest civilians ; there was an almost exactly equal number of Russians and Ukrainians. 1996 saw publication of an official report which a commission under Yakovlev had submitted to President Yeltsin64. This gave figures compatible with Zemskov's : by December 1946, 5,4 million people had been repatriated (among them the « internally displaced » ?), including 1,8 million prisoners of war. As regards their subsequent fate, Yakovlev's data are preferable to Zemskov's. Some 1 230 000 men were returned to the armed forces, of whom most were placed in a reserve under the Defence ministry and were employed on mine clearance or construction work - later some were sent on to the Gulag (a new detail !). Also new is a figure for those who were convicted : 994 270, of whom 157 593 were shot - by military tribunals alone. Those whose lives were spared became forced labourers. They included a 600 000-man « special contingent » which toiled under much the same conditions as other zeks.

• 65 Ibid.,(p. 108).

45As for those - men in the ranks and civilians - who, after passing through the network of filtration points and camps, were allowed to go home, they remained under police surveillance, suffered discrimination over jobs, housing and the like, and were especially prone to be denounced as politically unreliable by vindictive neighbours. Yakovlev shows that it was often the local authorities who pressed for harsher measures to be taken against the unfortunate repatriates, whose situation deteriorated after 1948. The military, to their credit, tried to intervene on their behalf with the MVD/MGB but to little avail so long as Stalin lived. Not until 1955 were some men amnestied or had their sentences reduced. The next year a commission under Marshal Zhukov discovered that since 1945 « no less than 1.5 million repatriated prisoners of war, the overwhelming majority of them innocent, had gone through a machine of terror that yielded nothing to the Gulag »65.

• 66 Dahlmann et al.(1995, p. 212).
• 67 M. Yerin, review of V.B. Konasov (1996) in Otechestvennia istoriia,1996,6, p. 194.
• 68 Wheatcroft (1996, p. 1345-1346). InCamp 517 in Karelia the death rate in April-August 1945 was 35 p (...)

46Perhaps the oddest feature of this mind-boggling injustice was that the Soviet authorities treated their own people worse than foreign prisoners of war. This was presumably because the latter, if they survived, were expected to return one day to their homelands. As S. Karner remarks, « a comparative study is urgently required of the Gulag, GUPVI and the German internment camps for prisoners of war »66. This task cannot be tackled here ; suffice it to say that on both sides of the line there was crass neglect and maltreatment, especially in the early years, but that later conditions improved - more so on the Soviet side, which at least did not apply genocidal policies in their regard. Rations in Soviet camps were on a par with those of civilians outside the wire. V.B. Konasov, in the most thorough inquiry yet made into this subject, leaves open the question how many prisoners died - German sources suggest about 1 out of 3 million - and points out that the figures were manipulated by the NKVD/MVD67. Wheatcroft has shown that the so-called « mobilized Germans », who were picked up in eastern Germany in 1945-6 and sent to work in the USSR, suffered an exceptionally severe fate68.

• 69 Vilensky (1990).

47Post-Soviet writings contain some evocative memoirs, notably in S.S. Vilensky's collection of material by and about women prisoners69. How far this modifies the picture we already have of physical conditions in the Gulag or victims' survival strategies cannot be considered here. Another subject that deserves extended treatment elsewhere is the contribution which forced labour made to the USSR's economic growth in the 1930s and to post-war reconstruction. It was clearly greater in some sectors (e.g. gold-mining) than others. There has been some discussion as to the relative importance of economic motivations in establishing and maintaining the Gulag, as distinct from the aim of bolstering the regime's security by ridding Soviet society of elements deemed undesirable. To my mind this is not a very propitious avenue of inquiry. Certainly, the location of camps and settlements was dictated by economic considerations, and the officials who ran them were, as pointed out above, primarily concerned with boosting output. Yet the reasons why so many people were « put inside » were non-economic. They had to do rather with the totalitarian essence of a regime ideologically committed to building a social Utopia, one that scorned as « bourgeois » considerations of legality or morality, and considered it« scientifically correct »to repress anyone who dissented or was deemed likely to do so. The security police and judiciary were cogs in what Zelenin calls « a state machine built especially for organizing forced labour and destroying people on a mass scale ». This does not mean that the machine functioned efficiently, and we should not forget the few brave, honest officials who tried to mitigate the victims' lot or even refused to carry out inhuman orders.

• 70 Graziosi (1992) ; Kokurin (1994).

48Resistance within the Gulag is another rewarding theme. It gathered force after 1948, notably among certain non-Russian groups as well as hardened criminals. This was a consequence partly of improved conditions (the end of the « Great Hunger »), and ranged from petty insubordination to the great strikes of 1953-195470.

• 71 Combe (1991, p. 214 ff).

50In Russia itself legal consciousness is poorly developed. There is neither any judicial mechanism capable of dealing with such matters nor a widespread public demand for retribution. On the contrary, even the victims' descendants display a remarkable readiness to « forgive and forget » which may owe something to the country's cultural and religious traditions as well as to the practical calculation that retribution would provoke further violence, and that it is therefore better to let sleeping dogs lie. Some citizens adhere to a concept of collective responsibility so broad as to rule out any accounting with the past : everyone had to compromise himself or herself under totalitarian conditions and so « we are all guilty up to a point ». At the other extreme we have the brash ex-camp commandant of Ozerlag, Yevstigneev, who claims that all his actions were legal and for the good of his captives71.

• 72 Zemskov (1995a, p. 107).

51Neither sentiment should obstruct the historians' efforts to identify the individuals who bore responsibility for crucial decisions and to evaluate their acts according to conventional criteria. Such mundane labours are an essential preliminary to any eventual stock-taking on a higher moral or judicial plane. This may take a generation or so to accomplish : one thinks of the taboo that for twenty years or so kept scholars from studying the Holocaust. The work done in recent years by our post-Soviet colleagues is commendable in many respects. It has the virtues (and defects) of being written close to the sources. For understandable reasons these scholars tend to work up a particular body of archival information to which they have managed to gain access, without paying much attention to the context or drawing broad conclusions. Rarely do they refer to parallel work being done by colleagues, and even more rarely to the work of foreign specialists. Their isolation is partly due to the sheer practical difficulty of obtaining expensive books and journals published in the West, a problem that ought not to be insuperable. Partly, though, it is a matter of intellectual barriers. Many seem unduly sensitive to possible charges of injuring their country's international reputation and so exercise restraint when passing judgement. As noted, their studies concentrate on the settlers rather than ITL/ITK inmates, a somewhat less painful theme. Writers are not always sufficiently attentive to the problem of falsification of official material. Zemskov has yet to reveal the circumstances in which he began to publish hitherto top secret information from NKVD/MVD files - although he has told us that in 1989 he was appointed a member of a commission of historians investigating population losses72. He is openly hostile towards foreign critics of the Communist record and even of Russian « democrats » (his inverted commas !).

• 73 Werth (1993).

52Western « revisionists »take a similar line. Both Rittersporn and Wheatcroft, the former more vehemently than the latter, see it as their duty to attack the position taken by anti-Communist stalwarts such as Conquest, allegedly in the cause of scholarly objectivity. The risk of sliding towards an apologetic stance, though less marked than it was, has not yet been wholly eliminated. This is perhaps more apparent in American than European (especially German) scholarship. However, lapses can occur anywhere. The respected N. Werth was perhaps a little unwise to claim, on the basis of Zemskov's revelations, that we at last have « les vrais chiffres » on the population of the Gulag73. Yes, the official data are clearly better than earlier outside estimates, but are they complete ! They need critical scrutiny. We do not yet know the answers to many important questions, because the accounting system was chaotic and the figures lent themselves to manipulation. Bureaucratic as well as political motives led to the separate registration of various categories of prisoner. In judging the Stalinist system one has to beware of institutionalized mendacity and avoid leaping to conclusions. Scholars in this sensitive field need to be humble about the extent of current knowledge but ambitious in setting future goals.