Content uploaded by Stephen Wheatcroft
Author content
All content in this area was uploaded by Stephen Wheatcroft on Aug 05, 2018
Content may be subject to copyright.
The Turn Away from
Economic Explanations for
Soviet Famines
Roundtable on Soviet Famines
STEPHEN G. WHEATCROFT
Anne Appelbaum’s work is a very readable and accessible story about the famine. In
her own words, her objective was to tell ‘what actually happened. . . . What chain
of events, and what mentality, led to the famine? Who was responsible?’ (xv). Right
from the beginning she indicates that she thinks that the famine was the result of
someone’s mentality, and that her objective is to find who should be blamed for it.
Her’s is a very simple story. It conforms to an increasingly popular trend in Soviet
history to ignore or oversimplify complex economic explanations and to reduce
everything to moral judgements.
The food problems that were explained by Alec Nove, Moshe Lewin, E.H. Carr
and R.W. Davies, and which most specialists used to think were responsible for
creating the circumstances in which extreme policies were formulated from 1927
to 1933, are largely ignored or misunderstood by Appelbaum and by many of the
current generation of specialists, who see no role for economic history. But the
idea that it was someone’s mentality that caused the problem is not a new idea. It
became popular under Khrushchev in the form of the denunciation of the cult of
the individual, and it has been used ever since to provide a scapegoat and to avoid
looking at the complexity of the problem.
Appelbaum and those most firmly invested in the cultural turn have been more
interested in trying to understand the feelings and emotions of those people who
experienced the tragedy of the famine than in trying to understand the complex food
problems of the time or in providing a critical social scientific explanation of what
happened. When Applebaum writes that ‘such an extraordinary catastrophe required
an extraordinary justification’ (209), she is unconsciously echoing the feelings of the
victims of this tragedy. They wanted a justification that they could both understand
and identify with. It is easier to understand and accept a tragedy caused by an
identifiable villain than to understand a complex problem in which impersonal factors
are at play, and in which deaths are to some extent accidental or were the collateral
damage of another process altogether. Survivors of major catastrophes can certainly
provide the best accounts of what it felt like to experience the catastrophe, but they
Contemporary European History,27,3(2018), pp. 465–469.c
Cambridge University Press 2018
doi:10.1017/S0960777318000358
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0960777318000358
Downloaded from https://www.cambridge.org/core. IP address: 175.34.21.238, on 02 Aug 2018 at 01:29:56, subject to the Cambridge Core terms of use, available at
466 Contemporary European History
do not necessarily provide the best explanations of the causes and the consequences of
these catastrophes. Victimhood does not necessarily result in improved understanding.
At a time when the government denied that a famine existed, when there was no
access to archival materials and when only censored reports were available, eyewitness
accounts were of particular importance. But even in the 1930 s there were sufficient
uncensored accounts of the famine to indicate the complex nature of the agricultural
crisis. The reports of Andrew Cairns and Otto Schiller were particularly important.
In addition, the results of the 1939 census showed clear signs of a major demographic
crisis with the population in 1939 16 million lower than planned. In 194 9 the
demographer Frank Lorimer identified a population loss of 4to 6million that he
thought was caused by the famine. The official revision of Soviet grain production
data in the 1950s, immediately after Stalin’s death, provided another indication that
there had been a major agricultural crisis which had been concealed at the time.
Even in the Soviet Union it was officially admitted that contrary to the official claims
of a 30 per cent growth in grain production from the mid 1920s to the early 1950s
there had been no growth at all. All these materials were available to allow a relatively
reliable explanation of the overall economic causes of the famine and its scale to be
drawn up in the 1970s. (See Melgrosh www.melgrosh.unimelb.edu.au, especially the
sections demography and famine). Despite this, the works of Mace and Conquest in
the early 198 0s generally ignored or underestimated the importance of the crisis in
grain supplies of these years. Conquest mentioned the food crisis but claimed that it
could easily have been resolved.
With the opening up of the Soviet archives in the mid 1980 s, the number of sources
of data available on the grain crisis were totally transformed. Professor Danilov and
his former students Elena Tyurina and Viktor Kondrashin ensured that many of these
materials were published in the major archival series The Tragedy of the Soviet Village,
1927–39, in five volumes (1999–2006), The Soviet Village in the Eyes of VChK-OGPU-
NKVD, in four volumes (1996–2012)andFamine in the USSR 1929–34,inthree
volumes (2011–13). Reliable demographic data also became available in the archives,
including data from the 1937 and 1939 censuses. This enabled more detailed estimates
than those of Lorimer to be constructed. I have found that the most reasonable
estimates of mortality rates caused by the famine in Ukraine (based on these new data
sources analysed by myself, Davies, Vallin et al) place the figure around 3.5million.
Attempts to claim the largest genocide in the world with 7to 10 million victims are
hard to justify. Appelbaum’s claim that ‘the Ukrainian scholarly community is now
coalescing, with some exceptions, around [Wolowy’s] number just below 4million
deaths’ would be good, if it were true, but her qualification ‘it is still possible to hear
numbers as high as ten million deaths’ (360 ) seems to indicate that some diehards are
finding it difficult to move closer to a realistic assessment. Using a team of Ukrainian
demographers may make this move more palatable to Ukrainian nationalists, but I
see no reason why academe in general should move away from 3.5million.
As regards the grain problem, the archival data published in TSD,Golod v SSSR
1929–34 and in Kak Lomaly NEP all provide confirmation of the views of Lewin,
Carr, Davies and their colleagues of the central importance of the grain problem in
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0960777318000358
Downloaded from https://www.cambridge.org/core. IP address: 175.34.21.238, on 02 Aug 2018 at 01:29:56, subject to the Cambridge Core terms of use, available at
The Turn Away from Economic Explanations for Soviet Famines 467
the economic and political decisions of the time. Davies and I have (2004)produced
the most detailed account of the grain crisis in these years, showing the uncertainties
in the data and the mistakes carried out by a generally ill-informed, and excessively
ambitious, government. The state showed no signs of a conscious attempt to kill lots
of Ukrainians and belated attempts that sought to provide relief when it eventually
saw the tragedy unfolding were evident. The relief measures that were given were
of course too few and too late to make much difference and they were also given in
secret with most concern over covering up the catastrophe that had occurred.
Throughout the early 1930s there was great uncertainty over the level of grain
production, as there had been since the First World War. The level of grain production
that was officially accepted in the late 1920sandin1931 and 1932 was already greatly
exaggerated, when an attempt was made to objectify harvest evaluation by switching
to a system of sample measurements. This produced the so called ‘biological yield’ of
grain, which was measured ‘on the stalk’, prior to harvest losses. Harvest losses were
normally about 20 to 30 per cent of the crop, but in 1932 they were probably much
higher. These harvest losses had to be deducted from the ‘biological yield’ to produce
the ‘barn yield’, or the amount of grain available for use. In 1931 and 1932 the level
of grain actually available for use was dangerously low. The Soviet government at the
time tried to cover up its failure to increase grain production and refused to scale
down grain procurements, claiming that more grain was available than was the case.
From 1933 to 1954 the official evaluations presented the biological yield figures as
though they were barn yield, and consequently exaggerated grain production by 20
to 30 per cent. In 1954 Khrushchev removed the post 1933 biological yield distortions
but kept the pre-1933 subjective distortions.
Many historians who have examined the famines do not understand the level
of genuine uncertainty that there was regarding grain statistics. They also fail to
understand the complexity of the problem over the possible level of harvesting
losses and how these impacted on the food supply problem, which has led to a
misrepresentation of how the famine progressed.
Anne Appelbaum’s treatment of grain availability in Ukraine epitomises the
dangers of misunderstanding the data. She uses the official grain production figures
of the time (for 1930–2) as if they were reliable indicators of the scale of production.
She then (for the years after 1933) switches to the official Soviet post 1954 series
of data which were 20 to 30 per cent lower than those officially used at the time.
This provides her with the startling, but unjustifiable, conclusion that the level of
grain production in 1931 and 1932 wasaboutthesameasin1933 and that therefore
there was no grain shortage in these years. This is incorrect. All experts, including
Prokopovich, Jasny, Tauger, R.W. Davies and myself, agree that the official grain
harvest figures for the late 1920sto1932 need to be deflated, and that the levels in
1931 and 1932 were dangerously low.
But it is not just confusion over the scale of harvesting losses; most historians
who have studied the famines in recent years are unaware of what is involved in
the harvesting process and how harvesting losses might arise. Anne Appelbaum even
thinks that ‘Ukraine has two harvests a year with Winter wheat harvested in July and
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0960777318000358
Downloaded from https://www.cambridge.org/core. IP address: 175.34.21.238, on 02 Aug 2018 at 01:29:56, subject to the Cambridge Core terms of use, available at
468 Contemporary European History
August, and Spring grains harvested in October and November’ (4). This is factually
incorrect. In Ukraine and other parts of the Soviet Union the two different sowings
are basically for the same harvest period. The only difference is that autumn sown
grain lies dormant throughout winter but can then start to germinate immediately
when the appropriate spring sowing period begins. This normally gives it a few
extra days of growth before the onset of hot weather in the summer, which can be
damaging to the plant, especially in the flowering period. Winter grains therefore
normally have a slightly higher yield than spring grains. Harvesting of winter and
spring grains occurs at roughly the same time with winter grains just a few days
earlier than the spring grains.
The famine was associated with two years of harvest failure in 1931 and 1932.
1931 was a year of drought with demonstrably excessive temperatures and low rainfall
in the early summer injuring the flowering and filling out of the grain. 1932 was
a year in which the biological yield (prior to harvesting) was relatively normal, but
in which harvest losses were excessively high as a result of damp weather during
the harvest period, and a slow progression of the harvesting which greatly increased
harvest losses.
The timing of the different stages of the grain harvest varies from region to region,
and these regional differences are far more important than the differences between
winter sown and spring sown grains. Harvesting begins earliest in the South of
Ukraine, where the spring sowing conditions begin earlier and where the growth
process ends earlier, and it then progresses in the following weeks to the more
northerly regions, where sowings begin later and where the harvesting stage will be
reached later. The harvesting process is a complex one, which at the time normally
required three separate processes which all developed at separate rates. First the grain
was mowed (cut). Then it was bound and stook into sheaves in the field. This was an
insurance policy carried out to try to minimise harvest losses from rain and dampness
in the fields that could rot the cut grain, if it was just laying around in the fields. The
final stage was threshing, which removed the grain from the sheaves on the stalk and
allowed it to be bagged and moved into the barns.
Detailed reports on the progress of these harvesting stages are available every five
days throughout the harvesting period, and it is clear that in 1932 there was a major
delay in the harvesting process, with much grain left unstocked and rotting in the
fields, This occurred for a variety of reasons, including unusually damp weather in
Kiev Oblast and low levels of traction power. Davies and I provide more detailed
explanations of this in our 2004 Years of Hunger: Soviet Agriculture 1929–1931.
Appelbaum makes two references to the above text, although she preferred to
cite grain production figures from a certain Bashkin (whose work is not listed in the
bibliography) indicating that production only fell from 69.9million tons in 1932 –3
to 68.4million tons in 1933–4. Our detailed, critical data analysis, however, estimates
grain production in 1932–3to have been 55 to 60 million tons, and that this was 15
to 17 million tons less than the following year when we estimate it to have grown to
70 to 77 million tons. It is this failure to understand that there really was a shortage
of grain at this time that leads to the conclusion that there was an easy solution to
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0960777318000358
Downloaded from https://www.cambridge.org/core. IP address: 175.34.21.238, on 02 Aug 2018 at 01:29:56, subject to the Cambridge Core terms of use, available at
The Turn Away from Economic Explanations for Soviet Famines 469
the problem, and that if Stalin failed to implement this easy solution, there must have
been a political reason why he did. This is the reasoning for thinking that Stalin must
have wanted to kill Ukrainians.
Robert Conquest had similarly originally underestimated the extent of the crisis
and had earlier written that ‘Stalin could, at any time, have ordered the release of
grain, and held off until the late Spring’ (Harvest of Sorrow,326), but when confronted
with the evidence, he changed his mind. When Davies and myself provided him with
documented details about the scale of the crisis and the large number of secret relief
measures carried out by the Politburo, and when we argued that we disagreed with
Conquest’s published view that Stalin ‘wanted a famine’, and that ‘the Soviets did not
want the famine to be coped with successfully’, he responded by modifying his earlier
criticisms. He asked us to state publicly that it was not his (Conquest’s) opinion that
‘Stalin purposely inflicted the 1933 famine. No. What I argue is that with resulting
famine imminent, he could have prevented it, but put “Soviet interest” other than
feeding the starving first-thus consciously abetting it’ (Conquest letter to Wheatcroft,
September 2003). We complied with Conquest’s wishes and included that statement
in footnote 145 on page 441 of our book, which then received an approving blurb
from Conquest. (Unfortunately Conquest’s blurb was only reproduced in the first
edition). It is consequently wrong to cite the views of Conquest as a justification
for accepting that the famine was a genocide, caused on purpose to kill Ukrainians.
We all agreed that Stalin’s policy was brutal and ruthless and that its cover up was
criminal, but we do not believe that it was done on purpose to kill people and cannot
therefore be described as murder or genocide.
With regards to broadening the narrative to include other regions than Ukraine in
discussions of the famine, I would welcome such a move including a broadening of
the maps of district (raion) level mortality, below Oblast level. My map of 1933 district
level mortality in Ukraine and in neighbouring Russian oblasts was published in two
Ukrainian books in 2013 and are available on Melgrosh. They clearly demonstrate that
claims that mortality fell immediately the Ukrainian border was passed are incorrect.
It is unfortunate that Appelbaum uses the Harvard Ukrainian Research Institute
(HURI) map, which fails to include mortality patterns in neighbouring Russian
districts.
Discussions in the popular narrative of famine have changed over the years. During
Soviet times there was a contrast between ‘man-made’ famine and ‘denial of famine’.
‘Man-made’ at this time largely meant as a result of policy. Then there was a contrast
between ‘man-made on purpose’, and ‘man-made by accident’ with charges of
criminal neglect and cover up. This stage seemed to have ended in 2004 when
Robert Conquest agreed that the famine was not man-made on purpose. But in the
following ten years there has been a revival of the ‘man-made on purpose’ side. This
reflects both a reduced interest in understanding the economic history, and increased
attempts by the Ukrainian government to classify the ‘famine as a genocide’. It is
time to return to paying more attention to economic explanations.
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0960777318000358
Downloaded from https://www.cambridge.org/core. IP address: 175.34.21.238, on 02 Aug 2018 at 01:29:56, subject to the Cambridge Core terms of use, available at
