Stalin and the Soviet famine of 1932 – 33 Revisited

Abstract

This article contributes to the debate about the role of Stalin in the Soviet famine of 1932 – 33. It provides data on Stalin's statements and actions in 1932 – 33, judicial and extra-judicial repression, and the process by which the 1933 deportation targets were drastically reduced. It is suggested that starvation was a cheap substitute for the cancelled deportations. It is argued that in 1932 – 33 Stalin pursued a multi-pronged policy of state terror against the population of the USSR. Some general issues of interpretation are also considered, such as Bolshevik perceptions, the characterisation of Soviet industrialisation, and approaches to Soviet history. Extensive attention is given to the classification of Stalin's actions according to national and international criminal law. In particular, the question of whether or not in 1932 – 33 the Ukrainian people were victims of genocide, is analysed.

Attentively studying the author's text, not only do [specialists] not stint their compliments, but they also make some critical remarks. Because (is it necessary to prove the obvious?) any really good book invites discussion (Ivanov 2006, p. 120).

The Stalinist leadership was only able to retain power then [in 1932] by using the most savage repression (Khlevnyuk 1992, p. 11).

Notes

¹This was not a specifically Stalinist policy but reflected long-standing Soviet priorities. In 1922 – 23 grain was exported to raise funds for the revival of industry at a time when famine was still widespread and the American Relief Administration (ARA) was still providing relief supplies (Serbyn 1986, pp. 165 – 169).

²This speech is discussed in Davies and Wheatcroft (2004, pp. 203 – 204). However, they do not draw attention to its implications for repression. Indeed, they assert that Stalin's speech was ‘moderate in tone’. This speech was not an isolated incident. Four days earlier Stalin had made another speech at the same meeting. In this he called for the ‘intensification’ of ‘class struggle’ and ‘revolutionary vigilance’. In their absence there was a danger, according to Stalin, of a revival of the old ‘counterrevolutionary parties of the SRs [Socialist Revolutionaries], Mensheviks, and bourgeois nationalists at the centre and the border regions’ and also of a revival of the ‘counterrevolutionary elements from the Trotskyists and the right deviationists’ (Stalin 1951a, pp. 211 – 212). Khlevnyuk (1992, p. 20) sensibly referred to this speech as ‘extremely harsh’. Davies (1996, p. 321) correctly noted that this speech ‘provided a justification for … repression’.

³The ultimate source of this ‘quotation from Lenin’ is II Thessalonians, 3:10.

⁴The speech is discussed by Davies and Wheatcroft (2004, p. 208) but the passage cited above is not mentioned.

⁵Davies and Wheatcroft do not point out that this slogan had been launched by Stalin himself a few weeks earlier.

⁶The report is in the Ukrainian archives and available on internet at http://www.archives.gov.ua/Sections/Famine/Publicat/index.php?1933-03 March (berezen') 12, 1933, accessed 27 March 2007. The inclusion of ‘thieves’ in this passage reflects the official interpretation of the decree of 7 August 1932. According to this interpretation, much of the food shortage resulted from the actions of ‘thieves’ who stole state and kolkhoz property for speculative purposes.

⁷This terminology was introduced by Wheatcroft (see Wheatcroft 2004).

⁸At the same time, he informed Stalin of the resources (boats, grain, timber-cutting equipment, horses, money, timber, 800 Communists, OGPU troops with 300 bayonets, and six fast armed and armoured boats) that would be required to implement this operation.

⁹He wrote on the paper (Khaustov et al. 2003, p. 406), ‘The expenses (1,394 million rubles) are a gross exaggeration. It is necessary to use the deportees themselves for these expenses’.

¹⁰According to a Politburo decision of 17 April 1933 (Pokrovskii 2005, p. 599), the 1933 deportees were to comprise six groups. These were, ‘kulaks’ exiled from the collectivised regions; those exiled for ‘sabotage’ of the grain procurements; marginals expelled from Moscow and Leningrad in connection with passportisation; ‘kulaks’ who had found work in industry; people exiled from areas along the western frontier of the USSR; and people condemned by the OGPU and the courts for periods of between three and five years. Virtually all of these people will have been peasants (or members of their households) at the time of their detention or shortly before it.

¹¹On 21 March 1932 Yagoda sent Stalin a request for large quantities of seed to be issued to the ‘special settlers’ (Pokrovskii 2005, pp. 515 – 516). He concluded by stating that if adequate seed to enable the ‘special settlers’ to sow food crops was not issued to them, then ‘removing the special settlers from the state supply of food and vegetables will be impossible’. Yagoda's request was approved by the Politburo two days later (Pokrovskii 2005, p. 518).

¹²Stalin had originally proposed that ‘as a rule’ the sentence under this decree would be the death penalty. The fact that only a tiny minority of those sentenced were shot probably resulted from a general unwillingness by judicial and security personnel to implement as originally intended what was widely seen as an impractical and barbaric decree (Khlevnyuk 1992, pp. 22 – 24; Solomon 1996, pp. 116 – 117). As Davies and Wheatcroft correctly observe (2004, p. 167): ‘The decree of 7 August was not only savage but impracticably savage’. This unwillingness is an interesting example of passive resistance by bureaucrats preventing Stalin from killing as many people as he would have liked to kill. Already the Instruction of 16 September 1932 on implementing the decree, drawn up by the top judicial and security officials and approved by the Politburo, provided that kolkhozniki and individual peasants (if they were not involved in organised theft, and were not ‘kulaks’, former traders, or other ‘socially-alien elements’) caught for ‘stealing’ kolkhoz property, would ‘only’ receive 10 years (Danilov et al. 2001, pp. 477 – 479). Although the death penalty and 10 years were supposed to be the only punishments under this decree, actually many shorter sentences were given. According to Khlevnyuk (1992, pp. 10 – 31), this opposition in society to extreme repression was one of the reasons why the Stalinshchina was postponed for four years.

¹³The data in Tables 2 and 3 show that Table 1 of Ellman (2005) underestimates the number of peasants repressed in 1930 – 33 because it excludes judicial repression.

¹⁴This decision was quickly implemented. By 27 December 2,158 families (9,187 people) had been deported from Poltava stanitsa to the Urals (Pokrovskii 2005, p. 568).

¹⁵On 24 January 1933 the members of the Politburo rejected a proposal by the northern kraikom to deport 3,000 households to the Pechora. These seem to have been households already living in the northern region.

¹⁶For the information in this paragraph, see Pokrovskii (2005, pp. 565 – 589).

¹⁷For a number of other quotations suggesting that Soviet officials in 1932 – 33 saw the famine as a way of teaching the peasants a lesson, see Kul'chits'kii (2005, p. 212). An alleged statement by Khataevich (in 1932 – 37 a member of the Ukrainian Politburo and in 1933 – 37 first secretary of the Dnepropetrovsk obkom), which seems to relate to 1933, would if accurate corroborate the idea that many Bolsheviks in 1932 – 33 saw the famine as a way of socialising the peasants: ‘A ruthless struggle is going on between the peasantry and our regime. It is a struggle to the death. This year was a test of our strength and their endurance. It took a famine to show them who is master here. It has cost millions of lives, but the collective farm system is here to stay. We've won the war’. This passage has been quoted by Conquest, Scott, and Pavlova. The original source seems to be Kravchenko (1946, p. 130). If authentic, this quotation would be relevant evidence. However, this statement is part of a conversation written from memory 12 years after the event, in a book published in the United States, which was intended to blacken the Soviet regime, and is part of a self-congratulatory chapter which portrays the author as having acted in a very positive way. Hence, it may be more literary than historical. The historical Khataevich tried—unsuccessfully—to defend the peasants against excessive procurements (Ivnitskii 1995, pp. 37 – 38; Davies & Wheatcroft 2004, pp. 150 – 151).

²⁰Kuromiya (1995) has pointed out that during the 1930s the view that difficulties had been caused by ‘enemies’ remained constant, but these ‘enemies’ were increasingly seen not as ‘class enemies’ (such as ‘kulaks’) but non-class ‘enemies of the people’.

¹⁸Davies and Wheatcroft recognise that in determining events ‘ideology played its part’ (Davies & Wheatcroft 2004, p. 441). However, they devote most attention to other factors. This is understandable, since some of the facts they have unearthed were hitherto unknown, and previous writers have given much attention to ideology. However, a balanced approach has to take account of all relevant factors, not just the newly discovered and hitherto neglected ones.

¹⁹For a theoretical discussion, see Lukacs (1971). On p. 198 of the translation Lukacs states that ‘when judging whether an action is right or wrong it is essential to relate it to its function in the total process’. In a 1990 interview (Kumanev 2005, p. 103), L. M. Kaganovich explained that, considered from the class-historical point of view, the repression of the 1930s had been entirely justified: ‘Look, if you investigate everything in detail, and look at every single case, then of course it is possible to find flaws and mistakes, no doubt about it. But if we approach the issue historically, then it was necessary to cleanse the country. This is shown by the current situation. Are there not people today who are open enemies of socialism and of the October revolution? There are lots of them! Therefore, those who want to defend the October revolution have to beat the enemies of this revolution, beat the enemies of Soviet power and of the Soviet state. The present situation demonstrates that we were right’.

²¹In his recent biography of Lenin, Loginov (2005, pp. 119 – 131) strongly criticises the accuracy of Vodovozov's account [which was utilised in Ellman (2005)] of Lenin's attitude to the famine of 1891 – 92. However, Loginov's reliance on a 1987 Soviet thesis and Plekhanov as sources for the famine of 1891 – 92, and the policy of the government towards it, undermines the credibility of his account.

²²The idea that the famine of 1891 – 92 had played a positive role in the transition from feudalism to capitalism and thus helped the ultimate victory of socialism was commonplace in Bolshevik circles. The first edition of the Bol'shaya Sovetskaya Entsiklopediya (Moscow, 1930, vol. 17, p. 458) observed in its article about famine in Russia that: ‘In this connection, famine years played a definite role in differentiating the peasantry. Already in 1891 Engels noted that: “the famine accelerates the disintegration of the old rural commune, and the enrichment of the kulaks, transforming them into big landowners, and in general the transfer of land from the nobility and peasantry to the new bourgeoisie”. This “flourishing of the kulaks” in connection with the famine of 1891 – 92 was also noted by Plekhanov. In his articles on the famine, Lenin too repeatedly wrote about this’.

²³The published version was an edited version of the speech actually delivered. For the stenogram of the speech actually delivered, and some explanations, see Nevezhin (2003, pp. 66 – 70, 76 – 91).

²⁴The term ‘state terror’ to describe Stalinist repression seems to have been introduced by Popov (1992). It contrasts with the Stalinist concept of ‘kulak terror’.

²⁵That is, the three million from Table 1 of Ellman (2005) plus a rough estimate of a million for judicial repression in 1929 – 33 (Table 2 above).

²⁶The evidence for this is Stalin's 10 January 1939 telegram (Ellman 2005, footnote 22) and the discussion at the June 1957 CC Plenum (Kovaleva et al. 1998, p. 119). According to Okhotin and Roginskii (Danilov 2006, p. 571) torture was authorised in July 1937.

²⁷For an analysis which incorporates both elements, see Gregory and Harrison (2005).

²⁸According to the Canadian Criminal Code, ‘A person commits culpable homicide when he causes the death of a human being, … (b) by criminal negligence’ (Article 222, section 5). Furthermore, ‘Where a person, by an act or omission, does any thing that results in the death of a human being, he causes the death of that human being … ’ (Article 224—italics added).

²⁹However, Schabas (2000, p. 228) notes that: ‘the plain words of the statutes of the ad hoc tribunals and of the International Criminal Court, recognising the application of command responsibility to genocide, make it at least theoretically possible for a superior or commander to be found guilty of genocide where the mental element was only one of negligence’. He adds, however, that: ‘The limited case law on this point indicates that the courts remain rather uncomfortable with the concept’.

³⁰According to Kul'chits'kii (2005, p. 196), the number of famine deaths in Ukraine was 3.2 million. This is 56% of the Davies and Wheatcroft estimate of the total number of famine victims. Since the Kazakh famine began in 1931, whereas the overwhelming majority of the Ukrainian victims died in 1933, the Ukrainian proportion of 1933 excess deaths was higher than the Ukrainian proportion of total famine deaths. The 3.2 million estimate is an estimate of excess deaths in Ukraine, not of excess deaths of those of Ukrainian nationality. To obtain the latter one would have to add the excess deaths of Ukrainians outside Ukraine and subtract the excess death of non-Ukrainians in Ukraine.

³¹According to Marcus (2003, p. 252), ‘In one year, between five and eleven million Ukrainians died of hunger or famine-related maladies … ’. This is a huge exaggeration. It seems that the actual number of Ukrainian famine-related deaths was much less than the lower bound of Marcus's range.

³²International Criminal Tribunal for the Former Yugoslavia, Judgement, Prosecutor v. Radislav Krstic, Case no. IT-98-33-T, 2 August, 2001, p. 208, available at: http://www.un.org/icty, accessed 12 January 2007.

³³The interpretation of the Armenian massacres is still very controversial, primarily for political reasons. Whether or not their treatment qualifies as genocide has been debated up to the present. On the academic level, one reason why proof of genocide in this case has been difficult to establish is that many of the relevant documents were destroyed. Another is the fact that Ottoman documents were written in the Arabic alphabet and their vocabulary was influenced by Persian and Arabic, whereas modern Turkish is written in the Latin alphabet and many words of Persian and Arabic origin have been replaced by their Turkish equivalents, so that Ottoman documents are difficult to read for those only familiar with modern Turkish. Furthermore, the orders for mass killings were sometimes given orally. Nevertheless, Akçam (2006) makes a convincing and well-documented case that the Armenian massacres really do qualify as genocide even according to a strict definition of the UN Convention. For a discussion of the deportation and massacre of the Circassians by the Russian Empire in the 1860s and whether or not that constituted genocide, see Stephen Shenfield ‘The Circassians: A Forgotten Genocide?’ in Levene, M. & Roberts, P. (eds) The Massacre in History, 1999, available at: http://www.circassianworld.com/A_Forgotten_Genocide.pdf, accessed 15 February 2007.

³⁴This has been recognised by Kul'chits'kii (2005, p. 197) who wrote, ‘We will never prove to the grandchildren of the citizens of Ukraine who died of starvation, still less to world public opinion, that people died in the USSR in 1933 because of their nationality. That is, like the Armenians in the Ottoman empire in 1915 or the Jews in the European countries occupied by Hitler's reich’. (He added, however, ‘There is no need to prove this, because the mechanism of the Soviet genocide was different’.)

³⁵For the text of the decree, see Danilov et al. (2001, pp. 576 – 577). For a discussion of its significance, see Martin (2001, pp. 302 – 308).

³⁶For an alternative view see Serbyn (2006). The fact that many famines, despite being terrible human tragedies and wholly or partly avoidable, are not genocide as defined by the UN Convention, because of the requirement of specific intent and the type of group effected, has been pointed out by Marcus (2003, pp. 264 – 265). Accordingly, Marcus argued for a new international agreement criminalising famines.

³⁷The ‘national operations’ were directed against the following ethnic groups (in each case the number of victims arrested by 1 July 1938 is given in brackets after the name of the group): Poles (148,000), Germans (65,000), Latvians (24,000), Iranians (16,000), Greeks (16,000), Finns (11,000), Chinese – Koreans (9,000), Romanians (9,000), Estonians (9,000), English (3,000), Afghans (3,000), Bulgarians (3,000), and others (7,000) (see Danilov 2006, p. 157).

³⁸For information about POV, see Khaustov (1997, p. 11).

³⁹In many cases determining who was a ‘Pole’ in the USSR in 1937 – 38 was arbitrary in view of changing frontiers and the use of different criteria (such as citizenship, passport nationality, language, religion, nationality of parents) in determining ‘nationality’.

⁴⁰When the USA incorporated the UN Convention into its domestic law, it changed the definition from killing a group ‘in whole or in part’ to ‘in whole or substantial part’. The purpose of this seems to have been to exclude the lynching of individual, or small groups of, African-Americans from the definition of genocide. This raises the question of whether the Polish killings were numerous enough to count as ‘substantial’. Killing more than 100,000 people, and in total sentencing more than a fifth of the relevant group (either to death or the Gulag), would probably meet the US ‘substantial’ criterion. However, there does not seem to be any US jurisprudence about how many killings are necessary to make them ‘substantial’.

⁴¹For 1937 there is data on clergy arrests both by category of ‘crime’ and by social composition. According to the former the number of clergy (dukhoventsvo, sektanty) arrested was 37,331. According to the latter the number of clergy (sluzhiteli religioznogo kul'ta) arrested was 33,382. For 1938, only data on the former has been published up to now. The number is 13,438. Most of these clergy were probably priests, monks and believer-activists of the Russian Orthodox Church, but they probably included also clergy of other Christian denominations and also of non-Christian denominations. For the figures see Mozokhin (2006, pp. 337 – 341).

⁴²The notorious NKVD order 00447 of 30 July 1937 specified that one of its targets was (section I point 6) ‘sectarian activists, churchgoers (tserkovniki) and others’ who were already detained in prisons, there camps, colonies and special settlements but who continued ‘active anti-Soviet, subversive, work’.

⁴³If the French additional category of genocide victims is combined with the more relaxed understanding of intent explained in the following paragraph, then Soviet policy in 1917 – 53 certainly included several genocides.

⁴⁴Kul'chits'kii (2005, pp. 157 and 153) agrees that the excess deaths of Russians (but not of Kazakhs) also qualifies as genocide.

Additional information

Notes on contributors

Michael Ellman

I am grateful to Joke Bakker, Karel Berkhoff, Patricia Ellman, Ferrie Feldbrugge, Mark Harrison, Marc Jansen, John Keep, Vladimir Kontorovich, Hiroaki Kuromiya (who intends to make his own contribution to the debate), Peter Maggs, Erik van Ree, Roman Serbyn, Peter Solomon, Paul Stephan, an anonymous referee, and the editor, for comments, help and advice. The author alone is responsible for the interpretation and the remaining errors.