In response to:

Stalin's Two Famines from the March 26, 1987 issue

To the Editors:

It is splendid that my The Harvest of Sorrow so greatly pleased as acute, even acerbic, a critic as Peter Wiles. But, though he is indeed one of the world’s leading authorities on Soviet economics and demography, his mild criticism of me for inadequate tabulation seems misplaced. It did not seem appropriate when dealing with what had, in the nature of things, to be no more than rough, round estimates.

However, this may change. My figures are based on what I rightly describe as a conservative estimate of the Soviet population shortfall between 1930 and 1937 of circa 13.5 million. I now find that the leading Soviet scholar of collectivization, in an obscure context not available to me when I wrote, makes it fifteen to sixteen million (V.P. Danilov, in Arkheograficheskiy Ezhegodnik za 1968 god; Moscow, 1970, p. 249): and this should serve in a more rigorous approach.

Two more points: Professor Wiles has me estimating the losses in the peasant insurrections of 1918–1921 as circa six million, which would indeed have been an exaggeration: in fact I say circa two million (of the five million deficit of males in 1926 as compared with pre-World War I, two million died in that war, and less than a million in the Civil War proper, which leaves some two million for the Peasant War).

Then, he disputes my reporting a decline of the real wages of the urban worker in the early Thirties to 10 percent of its precollectivization level, on the grounds that if so they would have starved like the Ukrainian peasantry, while in fact state shops sold supplies cheaply. This is true: but these shops provided little more than a subsistence bread ration—though they did provide that. The real income of the peasantry in the starvation areas was virtually nil. (The precise real wage level in the towns is still, of course, a refractory calculation).

Robert Conquest

Stanford, California

Peter Wiles replies:

The New York Review of Books is not Population Studies. More power to it, for allowing me to present a demographic table, but it should not get dragged in too deep.

Bob Conquest has kindly shown me his rare source. Danilov only uses the statistical yearbook in his footnote, and arrives at the same result as anyone else must. I have checked his total shortfall, and it amounts to eight million excess deaths in 1930–1935, and about eight million “never-born.” Danilov has evidently not had access to the 1937 census, which his only source, the statistical yearbook, never mentions. This is disappointing, for that census, which was totally suppressed, is universally thought to have given a lower total for 1937 than that implied by its extremely suspect substitute, the 1939 census. This lower total would entail, of course, a great famine shortfall.

In this context, Conquest does not mention in his letter, and only once in his book, the never-born. But these were extremely numerous, and I believe my Kazakh estimate in your columns (NYR, March 26) underestimated them. They should be, I suggest, not a quarter but a half of the shortfall. This raises logical issues: they were not Stalin’s victims, since a non-existent person cannot be a victim, and some of them must have been replaced by a “post-famine baby boom.” But they were part of the shortfall, and they relieve us of the necessity in which Conquest finds himself of inventing six-and-a-half million non-Kazakh, non-Ukrainian victims who actually existed and then died. I don’t believe in these people: dekulakization, to which he attributes their deaths, i.e. shooting or freezing to death on the road to exile in 1929, did not claim anything like so many victims as I read the evidence. If we say 10 percent of kulaks and their families died, that’s only one million people.

No, Bob, one must always criticize inadequate tabulation. One makes do with the materials one has, of course, but errors of definition can still unnecessarily double or halve the final guesstimate. Pedantry is a duty, nay a pleasure, at all times. Re-read your pages 53–54, and admit the numbers (on Civil War deaths) might mean anything. I accept that it was meant to mean what you say, of course!

Ne sutor ultra crepidam said Apelles to the cobbler who, having told him he had carved the sandals incorrectly, went on to criticize the whole statue: you’re a cobbler, you may criticize my work but not above the sandal. Well, sir, your reviewer is indeed but a cobbler, a provider of the humble bases upon which great works ought to stand. But he dares to reply, ne sculptor infra crepidam: the great artist must tread carefully in his own statistical basement, or he will bring his house down.

This Issue

July 16, 1987