1
Paper prepared for ARC Workshop, University of Melbourne, March 18,2009
This is a draft. Please do note cite or distribute without the author’s permission.
Soviet Famines and Food Problems in Historical Perspective
(and in comparison with the Chinese Great Leap Forward Famine)
S.G.Wheatcroft, University of Melbourne
1. Introduction
The objective of this study is to historicize the term famine and to consider some of the detailed specifics of the food problems and famines in the USSR and China in the 1920s, 30s and 50s, in order to provide an explanation for some of these detailed specifics. This is not an attempt to explain why famine occurred in general, but why a certain kind of famine, affecting a certain number of certain people, in certain regions, occurred at these specific times.
General explanations about unembedded problems, whether they be famine, repression or terror, are far easier to deal with than specific cases, but ultimately they seem to me to offer less satisfactory explanations. They allow so much slippage that almost any explanation is possible. Such explanations move away from testable (disprovable) hypotheses and hard quantifiable data towards narratives and subjectivities that are not amenable to strict analysis and testing. But this is not to say that we should accept uncritically any data. Quite the opposite: I think that anyone who uses quantitative data has an obligation to investigate that data, to understand how it was constructed and by whom, and to assess how reliable they are.
Let’s begin by defining the difference between a food problem and a famine. A food problem can be simply defined as a perceived shortage of food. This food problem becomes a famine when the extent of concern over the shortage reaches a certain level. Some definitions of the level at which a food problem becomes a famine are fairly mechanical, eg. when food-related excess mortality increases the normal mortality rate by 5 per thousand points per year, or when the crude annual death rate per thousand population (CDR) increases by 20%. But such mechanical definitions of famine are problematic for several reasons.
First, famine is partly a political and cultural term. What is recognised as a famine in a well-developed society with normally-low mortality levels, is probably very different to what passes as famine in an under-developed society, with normally-high levels of mortality.[1] It is to a certain extent a culturally and politically determined factor. A famine can be taken to be the level of food-related mortality, or even the level of threat of food-related mortality, which is considered to be intolerable in certain societies and which requires certain extra-ordinary famine relief measures to be taken. The declaration of a famine is the declaration that some extra-ordinary measures need to be taken. Different societies with different customs and cultures (and levels of normal mortality) will differ as to where they draw that line.
Second, famine is a social phenomenon. It occurs in societies or groups of people. Individuals starve, but groups suffer famine.[2] But how large a social group are we talking about? If we wish to quantify the phenomenon the scale of the social group that we are looking at is critical.
So in quantitative terms famine is a difficult term to use. We can quantify excess deaths in comparison with some established level of normal mortality for a particular size of population, and we can compare different levels of excess death for similar sizes of population. But even then we need to be a little wary before we begin talking about this level as if it was the level of famine. We ought to relate this level of mortality to the historical stage in the demographic transition of the given society. It makes a lot of difference if we are talking about a 10-20 per thousand increase in a high mortality society (with a normal CDR of 30 to 40 per thousand) or in a low mortality society (say 10-20 per thousand). But for the moment, and bearing in mind that this is our cultural construct and not necessarily that of the time, we will accept the fairly arbitrary figure of a 25% rise in annual mortality rates from a recent base level in a relatively high mortality society as a sure sign of famine.[3]
Once we accept this definition of famine we can identify certain patterns in the history of these food problems and their relationship to famine. The Soviet food problems of 1927-33 appear in their internal dynamic and their relationship to famine to be similar to the earlier food problems of 1916-22, and to the Chinese food problems of 1953-61. The food problems of all of these periods were complex and long drawn out and at some points they all resulted in major increases in mortality above the minimum levels that we would normally call famines.[4]
The 1916-22 food problem differed from the others because it generated two different types of famines: 1) There was an early and direct effect of these problems in the urban famines of 1918-20; but there was also a later secondary effect 2) as the government’s response to these problems led it to take drastic action against the rural population, which when weakened through other factors (in 1921 a drought) then contributed to the production of another and more terrible famine, amongst the rural population in the Volga, and parts of Urals and Southern Ukraine.
The distinction between an urban famine, (as a result of inadequate transfers from rural supply areas to urban food-deficit regions), and a rural famine (as a result of insufficient production and excessive taxation) is an important distinction to make. There are obvious connections between the two. Shortages in supply areas will normally lead to shortages in the receiving areas, and extreme attempts by the receiving regions to extract grain from the supply areas could affect future supply. We often think of this in terms of urban and rural areas, but for the USSR the distinction between deficit and surplus food areas is equally important, and is operationally easier to measure. (The same may apply for China.)
However, once we shift our analysis away from a simple urban/rural division, and into a deficit and surplus area divide, we are faced with the problem of what happens to the urban population in surplus areas. In normal circumstances there is no problem and they have access to small local level trade, but during a crisis in which the state is concentrating so much effort on central grain collections, the local market gets squeezed and either ceases to exist or goes underground.
The later Soviet and Chinese famines had a different but somewhat-related trajectory from the 1916-22 famines. The food problems were again initially felt in the urban areas, but this time urgent (extra-ordinary) measures were taken to provide protection to the main urban areas, which moved into rationing. An elaborate Central State Procurement system developed to protect the urban areas and to ensure that grain was exported.
In both cases the state was initially able to withstand this challenge to its urban population and unlike in 1918-20 the urban population receiving rations did not experience famine.[5] In fact, during these periods urban population greatly increased in size.[6] For a while when there was a temporary improvement in production, the state was also able to capture the extra surpluses and provide significant amounts of grain for export. It was the combination of a central state procurement system with drastic rural reorganization that enabled this growth in extra-rural marketings to occur. Although it was claimed that the object of rural reorganization was to increase agricultural efficiency and move towards socialism, the major result was to build an agricultural system that could be planned and that would market the large amounts of extra-rural grain that the authorities needed for their growing urbanization, industrialization, military and export plans.
Increasing tensions with the peasantry were another factor that favoured the construction of large units under central control where it would be more difficult to conceal peasant rebellion and opposition.
Statistical, planning and legal organizations were all mobilized in an attempt to increase grain marketings. A significant disjuncture arose in all these cases as to what was claimed to be the scale of agricultural production, and the level that was subsequently accepted.[7] Both countries claimed that leaps in production were taking place as their peasantries were forced to undergo these transformations.
In desperate circumstances peasants and some local leaders pleaded to have unrealistic procurement plans reduced, and claimed that grain production had not grown as sharply as had been claimed. But challenging these plans led to accusations of kulak sympathies and right deviationism (rightism in China). At the same time workers in the urban population and other local leaders begged the centre for more allocations of feed for the urban population as they also faced shortages. In some areas including Ukraine, the local leaders were faced with both problems, and the central leadership left them to resolve their own problems. A stock response to such problems was that underfulfillment in central collections in any region would need to be covered by reduced allocations from central collections in those regions.
Faced with these difficulties, and extremely harsh penalties for not implementing government policy, local leaders often preferred to ignore the contradictions in state policy and go along with the pretense that more grain was available than was demonstrably the case. This was a strategy that had been applied by Stalin and others in the Civil War, and it had probably contributed to their victory at that time. But in the Soviet Union in the 1930s and in China in the 1950s this military Stalinist expedient approach would be one of the major causes of the tragedy.
Given the prominence of balances in the central planning system, it was not all that easy to get away with exaggerating production, unless you could pass it on to some consumer who was prepared to accept this scam. There were very harsh penalties for exaggerating the level of grain collected and held in state stores. The data on central state procured grain and on centrally transported grain are relatively reliable. This is not the case for grain consumed in the collective farm, in the rural sector, or the grain collected and distributed through non-central procurement agents, where the data suffered greatly from what has been described as plan constructivist distortions.
The grain procurement officials, who after 1926 had a seat on the inter-departmental expert soviet that was ultimately responsible for harvest evaluations, had a vested interest in exaggerating the level of grain production, because it made it easier to collect grain and to justify the use of harsh extraordinary measures. This tendency for grain procurement officials to exaggerate the level of grain produced by peasants from whom they wished to collect grain was common to the 1918-21 period and 1927-32 periods in the USSR and the 1956-61 period in China. There were, however, some differences.
A crucial difference between the three cases was the attitude of the government to peasant resistance, and their willingness to come to a compromise solution over trade. In February 1921, when the Civil War was over, when peasant resistance was growing, but before it was known that there would be a drought in the spring and early summer of 1921, the Central Committee of the All Russian Communist Party debated what should be done. Lenin was worried that peasants would withdraw their labour from agriculture and reduce sowings. Osinsky, who at the time held the critical positions of Deputy People’s Commissar for Agriculture and Deputy People’s Commissar for Food Supply proposed the extension of planning to set grain sowing plans. Other radical politicians like Larin and Kritsman also advocated the expansion of the planning system and proposed the replacement of the technical State Committee for the Electrification of the Countryside (GOELRO) with a Universal Planning Commission – Gosplan. Lenin was horrified at what he thought was the irresponsibility of attempting overnight to issue a sowing plan to 20 million separate households. He argued that this was well beyond the organizational capabilities of the party at the time, which was undoubtedly correct. Lenin felt that it would be totally irresponsible to attempt to move into universal planning at this time, and publicly (in Pravda) let loose one of his most vitriolic attacks on the ‘Communist conceit’ of those party members who knew nothing about the real world and wanted to leap into a fantasy of universal planning. He was insistent that GOELRO remain as the only state planning body and that it be run by professional engineers like Krzhizhanovskii. He spoke out totally against the establishment of a Universal Planning Commission.[8]
The same day that Lenin’s article appeared the Central Committee came to a series of decisions on what to do in these difficult economic and political circumstances. The decisions appeared to be contradictory, and presumably indicated deep divisions within the party over this issue. The Central Committee decided to accept the proposal that Lenin had developed to move away from central planning and towards restoring the market by replacing the grain requisitioning system with a tax in kind. This proposal that inaugurated the New Economic Policy was then formally discussed and accepted at the Xth Party Congress in March, 1921, just prior to the Kronstadt Mutiny. But at the same Central Committee meeting an apparently contradictory decision was made to construct a universal planning agency (Gosplan), which ominously indicated that the change in policy was not totally accepted and that anti-market forces were still very influential within the party leadership. This would be very significant for the future, but for the moment in the history of the 1916-22 food problems, the decision had been made to restore the market.
In the 1927-33 period a similar problem emerged in 1929 after two years of extra-ordinary measures, which had effectively brought back a requisitioning system and destroyed the local grain market. Again the question arose as to how could the regime proceed with its harsh procurement policy without risking a massive peasant withdrawal of labour. This time the decision was made to press ahead. The party, security force and planning agencies felt that they could enforce sowing and procurement plans on the peasantry. A series of decisions were taken to press ahead with high procurements, to force 20 million independent households into about 10,000 collective farms and to expropriate and exile about a million of the most antagonistic richest peasant households who would be denounced as kulaks.
These plans were not fulfilled in exactly the way that they were initially intended. Up until the beginning of March 1930 the government and party leaders and activists clearly favoured very large scale collective farms with maximum levels of socialization. The three basic forms of collective farm were: 1) the least socialized Communities for Mutual Land Working (TOZ); 2) the intermediary Artel, which preserved private plots and private ownership of small (non-tractive) livestock; and 3) the maximally-socialized Commune where all land and all livestock were communally held. Of these three forms it was initially the most highly-socialized Commune that was the predominant form. Then for reasons that are still not totally clear, but which reflect Stalin’s fears that this extreme system was not going to work, Stalin called a halt and a rethink in his ‘Dizziness with success’ article in Pravda.[9]
This halt is often presented as if it was only a temporary halt while Stalin shifted the blame onto others, when he thought that things were progressing dangerously fast, and that he subsequently resumed what he had earlier been doing after a period of consolidation. But in fact there was a more significant change that was foreshadowed in Stalin’s speech. The very large giant Communes were broken up and it was decided that the Artel rather than the Commune would be the primary form of collective farm. These would be much smaller than the Communes and eventually there would be about 20,000 of them. This was a large number, but in terms of planning and control it was far more feasible than 20 million, but it did imply a degree of compromise with the peasantry and allowed for a limited amount of legal trade of surplus products on the collective farm market once central procurement targets had been satisfied.[10]
This seems to have been very different to what happened in China where collectivization developed rapidly from lower to higher stages and then unfolded in the movement to construct People’s Communes, which went further than the earlier Soviet Communes and incorporated elements of Khrushchev’s Sovnarkhoz reforms of the same time. The remarkable thing about the Chinese People’s Communes is that they left no space for the remnants of private trade, and offered no compromise at all with the peasantry. The Soviet system by contrast did allow for elements of free-trade once the central procurement plan had been completed.