Robert Conquest is one of those rare gifted beings who can combine in one book the results of research, documented and footnoted, with the haute vulgarisation thereof. So The Harvest of Sorrow is a very good book “in both kinds”—and let no mere academic say which is more honorable.

The story of Stalin and Soviet agriculture is an often told and deeply depressing one, and Conquest gets it straight. It starts with the liberation of the serfs in 1861, and the curious substitution, as local agents for the government, of the (preexisting) village councils for the feudal landlords. During the free peasant period, between 1906 and 1928, the peasant could dispose of his own land and sell his own crop in the market. There was, however, the horrible but brief interruption of “War Communism” (1918–1921), when the peasant was left with his private land but could not market his crop, which government posses simply “procured,” usually for nothing. In 1928 came collectivization, artificial famine, and the victory of the new system.

To my knowledge there is no great new truth in the story Conquest tells, until he deals with the famines: he merely tells the story better and gives many more quotations from the actors on the scene. He does not omit the scandalous behavior of the Western press, whose correspondents sat, and were allowed by their superiors to sit, in Moscow, reporting official denials while the Ukrainians starved to death. G. B. Shaw, incidentally, a “Western medium” all by himself, proved rumors of the starvation false from the banquets he had personally struggled to ingest in Moscow. But this story too has been told before, by Eugene Lyons and Malcolm Muggeridge.

Before coming to Conquest’s real contribution, his account of the “administration” of the famines, we must warn readers that he is not a natural statistician. He has taught himself the rules of statistics, but his heart is not in it, and when he’s away from the subject of the two famines in the Ukraine and Kazakhstan, he simply ignores them. For example, real wages were not, in 1933, one tenth of what they had been in 1926 and 1927. Otherwise the population of the cities would have died; while only Ukrainian and Kazakh peasants were dying. Conquest’s source has used only the free-market price index, leaving out the stable prices in the state shops, still a major source of supply. Again Conquest’s estimate of 6 million deaths from the various peasant uprisings between 1918 and 1920 is far too high. The figure is a residual from various data on deaths from other causes, and includes deaths from natural causes—about 6 million in themselves. People would not have been immortal if they had not been at war. Again Conquest provides no tables, perhaps in the absurd belief that prose is clearer.

What was “collectivization”? It arose from Engels’s late realization that the concentration of capital did not work in agriculture. The proletarian revolution would be achieved in the towns only, but would carry with it millions of neutral or hostile (and religious and ethnically minded, Stalin added) peasants into socialism. The peasants must, therefore, be persuaded to contribute their land, livestock, and tools to big collective farms (kolkhozes), where economies of scale would rapidly convince them that this was right. The collective farms would not be state farms, and they would be of unequal wealth, according to the average previous wealth of the members. The farms would pay no wages (not being state farms, and not employing workers) but a sort of dividend connected with labor inputs. The members would elect their own chairman, and in that way continue to manage their “own” property—and they would still be called peasants.

Nearly all the peasants were blind to the fine distinctions between state and collective farms, and so between confiscation and collectivization. Lenin used tax relief for the peasants as “persuasion,” but without force he failed to persuade. Yet something had eventually to be done: no respectable socialist state has private peasants; they generate capitalism from within, and the richer ones (kulaks) simply are capitalists.

For Stalin the issue was bound up with the “procurement” question of “War Communism.” He could not believe in a safe, voluntary market procurement system: the kulaks would not cooperate; they would get him. The failure of voluntary procurements—or what he thought was a failure—in 1927 sent him spinning down the road of price control, compulsory procurement, violence, and so collectivization for procurement’s sake, and much more violence. It was no longer an efficiency issue, and indeed hardly a Marxist or property issue. It had become a security issue: Who shall eat?

There are four large unsettled questions: Did the collectivization between 1928 and 1930 directly cause the Ukraine famine between 1932 and 1933? Did that also happen in Kazakhstan? Did Stalin intend these famines? If so, were these crimes unique?


Conquest does not ask the first question directly, but his book makes clear that the answer to it is no. There was no famine in Russia proper, Belorussia, or Transcaucasia—that is, most of the country. Nationwide, the harvest of 1931 had been adequate, and even in the Ukraine not catastrophic. Harvests in the Ukraine as elsewhere had been, of course, hit by collectivization ever since 1929, but again not catastrophically. This is a surprising fact, which Conquest certainly should have more fully recognized. But fact it is; so collectivization was not a catastrophe, only a very bad thing. It did not directly cause the Ukrainian famine. After all, efficiency was not much at issue, and Soviet agriculture was overpopulated nearly everywhere; deporting those defined as kulaks—perhaps 10 percent of the agricultural population—should not have, and did not, grossly diminish crop, as opposed to livestock, production.

But collectivization certainly did cause the Kazakh famine of 1930–1933, to which Conquest to his great credit devotes a whole chapter—one of the very few writers to do so in the West. The nomadic Kazakhs put up violent resistance to the twin, and simultaneous, processes of settlement and collectivization. Their way of life was totally transformed to an extent no Ukrainian could imagine. Furthermore, like all other peasants they slaughtered their livestock to prevent its “confiscation.” But since their only crop was grass, they had no source of food, and died.

How many died? An excess of deaths is a very difficult statistical concept. From all the data Conquest provides (albeit in very disorderly fashion), I infer the following (figures are in thousands and extremely approximate):


Conquest estimates only 1 million unexpected deaths, and he does not show how his calculations were made. Note how complicated such sums are, and how many assumptions and curious definitions they contain. Of these, there are even more assumptions than appear, but this is not the right place to explain them. Suppose, for instance, that some of those assumed to die natural deaths within the three-year bad period in fact died of starvation at the beginning of it. Kazakhs living outside Kazakhstan, and within that territory in state farms or towns, are included. The former must have suffered rather less than the peasants and nomads at home, the latter much less. Other ethnic groups living in Kazakhstan are excluded. This is a good statistical tactic: in a period of mass migration a territory may lose or gain huge populations despite a famine, but an ethnic group should be also reported as such, wherever it migrates.

One million two hundred fifty-seven thousand deaths is 29 percent of the population living when trouble began: about the average in Cambodia under Pol Pot. Apart from Conquest and Martha Olcutt, his main source, we all stand accused of unconscious racism: we have overlooked the bigger tragedy, no doubt in part because “Turks don’t matter.”

So in Kazakhstan, and elsewhere in Central Asia, collectivization itself was a complete failure and directly generated not only famine but civil war. Yet it was only natural: fanatical Russian Communists like E. I. Goloshchekin, the Party secretary of the Kazakh Republic, imposed a foreign and totally inefficient system suddenly on a non-Slavic people. It needs no conspiracy theory to explain this: Bolshevism and a tincture of racism will do. Indeed, Goloshchekin himself was dismissed in February 1933.

In the Ukraine, on the other hand, where there was merely “resistance” to collectivization—some rioting, the murder of a few Party functionaries—it was upon a (barely) functioning kolkhoz system that Stalin turned his fury in 1932. Why did he do so? Conquest rejects my previously favorite explanation: that Stalin was building a grain reserve for his Far Eastern Army because of the Japanese invasion threat, and so collected too much grain. This theory has a beautiful irony about it: it proves how efficient the new system was at collecting grain, as it was certainly meant to be, though not at producing it.

But irony does not prove a theory, and Conquest’s explanation is a better one, while it also takes account of the procurement issue. He describes at length how Stalin became convinced that Ukrainian nationalism was his greatest enemy. Nationalism was strong in the Ukraine, reaching from intellectuals, priests, and peasants, right into the Party itself (the postwar boss of the Ukrainian Republic, Petro Shelest, was even a nationalist Stalinist). Only the proletariat was seen as a healthy element—and during the famine it did not starve. The Ukraine Republic was, I might add, on an invasion path.

Stalin imposed vast grain procurement targets on all the new collective farms in July 1932 (perhaps this supports my Japanese story). But in the Ukraine the targets were higher and the harvest smaller. Did he have famine already in mind then? We do not know, but the grain deliveries were not made, and he proceeded to impose further enforcement measures which got the grain and killed 5 million people in the villages. Was this genocide? Nearly everything hangs on the evidence, not new but recently insisted upon by the Ukrainian community in North America, that he set up administrative and police barriers all along the Russo-Ukrainian border to prevent food from leaking through. That is, not only did he provide no state famine relief and suppress all mention of the famine in the press (par for the Communist course); he also took a great deal of trouble to prevent private famine relief. That, then, is genocide.


Craig Whitney, reviewing this book in The New York Times Book Review, complains that, on this last issue, “a few citations from The Black Deeds of the Kremlin and other exile sources do not make the case.” I count fifteen citations on administrative barriers on pages 379 and 391. Six are from The Black Deeds.1 I must confess that the title of that book has always put me off reading it, but it is not the least of Conquest’s merits to have ploughed ahead. He also cites four Ukrainian exiles, including Leonid Plyushch; the rest are contemporary sources, two of them official. Fifteen is not “few,” and to require only non–Ukrainian exile sources is to close the case. Here is one example from page 327:

One Ukrainian peasant who had earlier been recruited to work on the railway in the Moscow Province heard of the famine at home and left Moscow in April 1933 with seventy-nine pounds of bread. At Bakhmach on the Russo-Ukrainian border seventy pounds of it was confiscated. He was allowed to keep the rest as a registered Moscow worker, but two Ukrainian peasant women who were also trying to bring in bread had it all confiscated and were “detained.”

When we add Stalin’s extremely frank reactions attested to by well-placed protesters, we cannot doubt Conquest’s conclusion. Stalin said that the Ukrainian peasants are trying to starve “us”; but if anyone starves in this “battle” it shall not be “us.”

So Conquest (and James Mace before him)2 has adopted the Ukrainian exile view, and he has persuaded this reviewer. Indeed, a further point may be argued that he does not stress: the North Caucasus, an area adjacent to the Ukraine, was the third of the three great famine sufferers in 1932–1933, and consisted of at least half Ukrainians or Cossacks or—worse—Ukrainian Cossacks. So Stalin chased his least favorite ethnic group wherever he could find it.

It has fallen to me recently to supervise an excellent doctoral thesis in “catastrophology,” the study of the social and economic causes of and reactions to catastrophes, natural and man-made. A growing branch of the subject is the comparison of government reactions. All large national events are political, and many catastrophes are large events. There is always a government reaction, and it is usually a self-serving one, not in the victims’ best interest. Relief is steered this way or that, or even prevented altogether. Thus recently after the earthquake in San Salvador the government diverted relief teams from the slums to bourgeois quarters so that foreign TV crews would photograph only good, if collapsed, housing; and the Mexican government has consistently played down the number of deaths in its earthquake, because to admit to a higher number would have entailed military rule in the capital. So in order to obtain foreign disaster relief it overplayed the economic loss instead.

But these are relatively petty matters. In contrast, Czar Alexander II, in 1891, opened his ports to incoming grain but closed them to exports of it, major grain exporter though he had become. There was under Alexander little foreign relief, but much voluntary private relief at home. In 1921 Lenin, who had sneered at these volunteers (see Conquest, page 234), let in the American Relief Commission under Herbert Hoover and made no attempt to conceal a catastrophe of which he was partly the cause. In 1930, however, Stalin did not lift a finger to save the Kazakh peasants. After all, they were a hostile class and nationality. But he did not hound them to their deaths. He was “responsible,” but then so were all Soviet Communists in much the same ignorant way: they would all have insisted on pursuing policies that “objectively speaking” kill illiterate nomads. The ineffable Goloshchekin was but the individual executor of the collective fantasy.

The Ukraine was different. Without collectivization, of course, there would have been no resistance, and without resistance no revenge. But there was little fighting in 1931 and 1932, and no reason to expect any more. Yet the destruction of people was of genocidal proportions. Conquest implies that 5.5 million ethnic Ukrainians were destroyed, out of a population of 34.1 million at the beginning of the trouble, or 16 percent. These deaths, then, were almost unprovoked, and the sheer number matches that of the Holocaust. The percentage is mine, based on Conquest’s statistics for excess deaths in the Ukraine and the North Caucasus (seemingly correct, but I have not reworked them as I have for the Kazakhs), but on my own population figures. For it is better, as in the Kazakh case, to take account of the earlier and later census totals for ethnic groups, not for the republics, since so many Ukrainians migrated during the period between the two censuses.

Incidentally both Conquest and I have been careful to exclude the premature deaths of deported kulaks and others if they probably occurred in prison after January 1939. For it is fair to say that those deaths occurred as a consequence of the prison system. On the other hand, deaths that occurred during riots or from brutal treatment during deportation in 1930 are indistinguishable from those caused by the famine two years later. They were, however, much more evenly spread across the whole country.

In the much greater famine that began in 1932 food exports should have been stopped as they had been in 1891. They were not. Even Ukrainian food was exported. Offers of foreign aid should have been accepted. They were not. Indeed, no information on which foreign aid might be based was allowed out. Food should have been imported on credit or for gold. It was not. Again in 1936 the USSR, including the Ukraine, suffered near-famine conditions. Stalin exported no food this time, but he imported none either, and again concealed his problem.

Collectivization, and the anger of a communist government with a rebellious ethnic group, played their full part in Ethiopia under Mengistu, the Aid Diverter. But his predecessor Haile Selassie hardly distinguished himself in 1973 when he refused to appeal worldwide for help, claiming that “poverty is the will of God.” And the Sudanese government, which is not communist at all, diverts aid for its famine victims though it does not collectivize. The Irish famine of 1846 comes to mind also, when the British government set a distinguished precedent for Haile Selassie—it held that aid would “pauperize” the people.

Come to think of it, the author of The Great Terror has shown himself to be a professional “catastrophologist.” For he who would pursue that specialty in Soviet circumstances needs a very special nose for the new motivations and the new self-deceptions that govern that unhappy country. Let me change the tense, prayerfully, to “governed.”

This Issue

March 26, 1987