M. Maksudov, Losses Suffered by the Population of the USSR 1918-1958. 1977
The Tsar’s orders are that one should pray for those who fell in disgrace and that the mass for the dead should be said in their name. As for those who are not specifically mentioned in this memorial, or who are mentioned only by surname or by number, 10, 20, 50, it is proper to pray for them and to say: ‘You, God, you know their names.’ [Memorial for those who fell in Disgrace under Ivan the Terrible]
People are the main riches of our country. [Population of the USSR. Moscow, 1974.]
How many fell in this abyss?
How many? The question retains all its relevance; and is raised with pain and anger in the pages of the Archipelago. But even so ours is a generation that can live undisturbed, except when we remember the names of individual innocent victims.
We will not, however, become a people, a nation, until we have come to terms with this and with history and moved from a position of disinterest to respect for human beings and ourselves.
The aim of our study is to assess the minimum number of losses suffered by the Soviet population’. By losses we mean those people who disappeared prematurely, before their natural death. They include the victims of war as well as of camps, hunger epidemics, insufficient medical care, etc … The term ‘losses’ in this sense does not simply mean a decrease in population, or the number of victims of repression: it corresponds to the increase in mortality, worked out as the difference between the decrease in population and a theoretical ‘natural mortality’. The level of ‘natural mortality’ is obtained arithmetically. It is an easy calculation if the figures for population and mortality rate per age group are known.’ For instance, the 17 December 1926 census listed 933,000 men aged 30, and the mortality rates for men aged between 28 and 32 were 0.62, 0.63, 0.64, 0.66, and 0.68 per cent. This means, given a mortality level corresponding to these figures, that there must have been, at the end of 1927, 933 – (933 x 0.64 per cent) = 927,000 31-year-old-men. The number of men aged 32 in 1928 would apparently be 927 – (927 x 0.66 per cent) = 921,000. In 1929, the number of 33year-old-men would be 921 – (921 x 0.68 per cent) = 915,000. It is just as simple to count the men aged 29 in 1925: 933 + (933 x 0.63 per cent) = 940,000, and the men aged 28 in 1924: 940 + (940 x 0.62 per cent) = 947,000, etc …
In this way we follow the evolution of the number of men and women of each generation and thus, the evolution of the population of the whole country. The resulting table reflects the natural population trends (without taking birth-rate into account) and in cases where the actual population figure provided by the census is obviously lower than that assessed, we can then speak of human losses.
The method used for this study is not new, and has frequently been employed for population surveys . Iu. A. Korchak-Chepurlovskii has shown how important such forms are in historical research:
Bearing in mind the complex level of the method and technique which have been used, the estimates of M. V. Ptukhi, of S. A. Novosel’skii and V. V. Paevskii, are still held to be of great interest for the demographic historian attempting to analyze the demographic consequences of events which have interfered with the normal population development in our country.’
Calculations were made for three periods: 1926-38, 1926-1918 (by the retrospective method), and 1939-58; the pivot years were fixed by census dates. We have retained the population figures obtained in the 1926 census, taking into account the correction made by Novosel’skii and Paevskii. They are the authors who established the mortality rate tables which were used in our assessment. The rates from these tables have been used without modification for the 1927-38 period, but for the other periods studied, they have been either up- or down-graded: multiplied by 1.5 for the 1897-1922 period, by 1.2 for the 1923-6 period, by 0.9 for the 1939-49 period. The multipliers used were calculated on the basis of the data available for the mortality rate of those years (see table I, and table III). In conjunction with our proposed aim- i.e., the search for a minimum-we adopted figures higher than the population mortality rates published by the TsSU (table 11). It is useful here to draw attention to the constantly decreasing mortality in this country (table II), which must inevitably lead to an overestimation of the population losses, in relation to the data available. Such a phenomenon can be observed just as easily during the relatively peaceful periods of our history.
For the 1897-1913 period, the population decrease obtained by calculation was 1.4 million more than the actual decrease; 0.45 million more for the 1923-31 period, and 2.9 million more for the 1950-8 period (table V; table IX-XI).
Given that this tendency was characteristic of peacetime, the enormous discrepancy between the actual population figure and the estimates for 1914-17, 1918-22, 1932-8, 1939-45 (of 1.7, 14.3, 7.9 and 27.4 millions respectively) can without a doubt be considered to stem from human losses. The alternation of periods of peace and periods of catastrophe is also important here, since on principle we have used the same mortality rates for all the estimates. It is significant too that the younger generations, little affected by the (political) ‘measures’, follow the same pattern in the years of catastrophe as they do in the years of peace (see table X and table XI).
Description of the Material Used
We believe that the CSU must provide objective material, free from any preconceived idea, since any attempt to conceal figures for whatever reason constitutes a criminal act. [J. Stalin. Fourteenth Congress.]
Tradition demands that the author, at this point, should evoke the work of his predecessors with some contempt; and it is unlikely that a more appropriate occasion has presented itself to anyone. Amongst demographic publications of these last few years, which can be numbered in thousands, only the works of B. C. Urlanis are devoted to the question studied here. But what can be expected of him when he finally admits that his results had been predetermined by a series of party dictates? It is appropriate, of course, to take note of the remarkable progression of B. C. Urlanis’s theories, He was writing in 1960: ‘If we try to compare the losses suffered during the civil wars and the respective population figures for this kind of conflict in recent times, we must first consider the Spanish Civil War, between 1936 and 1939 (1.8 per cent); next the civil war in the United States (1.6 per cent), the war in the Vendee (I per cent), and lastly the civil war in USSR (0.5 per cent). Nine years later, he already remarks: ‘Therefore, in the space of three years, from 1918 to 1920, because of the number of deaths being greater than that of births, the population figure has fallen by approximately 4 per cent, and bearing in mind the losses due to emigration, like Volkov, we take to be in the order of 3.5 million. There therefore remain (see table 11), 10.3 million which strictly speaking can be considered as human losses. As was noted above, this estimate can be overrated by 0.4 million, through arithmetic errors, or underrated by one to four million through inaccurate calculation of the population figures and the number of emigrants.
Human Losses in the Years 1926-1938 Collectivization. Repressive Measures
Each particular system of historical production effectively possesses its own demographic laws, which have historical consequences. [K. Marx, Capital]
After the 1926 census, the material relating to birth-rate, mortality, population counts, was published every year until 1936. After that, there was a most revealing silence which, as Orwell rather pertinently observed, only underlines the importance of historical events in certain countries. For six years, i.e., 1932-8, the CSU and CUNHU publications repeated one and only one figure: 165.7 million inhabitants on I January 1933. The results of the 1939 census – 170.4 million inhabitants once and forever replaced the familiar preceding figure. So minimal was their difference that the two figures were never placed alongside one another. During these last few years, material covering the years 1937-8 has appeared, delineating precisely a new period of a rather murky nature: 1931-6. These years are like the thinnest cows of the pharaoh, who devoured the previous, relatively prosperous years of the NEP. Though the aims were the same, the pharaoh’s methods were more sophisticated, more efficient and even, one could say, more humane. Moreover, it is possible that responsibility should rest neither with Stalin nor with the pharaoh, but simply with the person who was then in charge of the Agriculture Department of the Central Committee. Then Stalin would neither have allowed the mass slaughter of cattle, nor would have deported the dissatisfied, nor would have tried to starve the insubordinate. But history does not repeat itself and, as everyone knows, ‘Soviet communism is a new civilization.’ This is why in the kolkhozes where the fields have deliberately not been ploughed or sown, there was no help available when they found themselves without food – in order not to encourage this sort of resistance except in the most serious cases, where entire villages were saved from famine because they were dragged just in time from their native soil …
It is because the Soviet government did not have workhouses and there was no time to build them. The only solution was to force the starving to leave their villages, where their presence had a demoralizing influence, and to go to faraway places … In other cases, the peasants inconspicuously hollowed out the ripe wheat, i.e., they crushed the grain or even cut the ears of corn and hoarded a personal reserve from what had been so unscrupulously stolen from the collectively-owned fields.
This really is a ‘new civilization’? It is hard not to agree with the Webbs, from whose book we have taken these quotes. A new civilization not only as far as the peasantry were concerned, but also when compared to more ancient times. It is the first time in world history that a famine was artificially caused, on such a large scale; informers, as in the Roman Empire or at the time of the Inquisition, received a part of the deportee’s property; the kulaks and the bourgeoisie (6.8 million according to the CUNHU publication: The Budding of Socialism in the USSR) were liquidated ‘as a class’ and one hundred and eleven million individual farms were collectivized.
As far as the geographical distribution of population is concerned the repercussions of both collectivization and deportation led to a decline in some areas, and an increase in others. (See table VII.) The increase in the population figure for the Ukraine, of 2 million in the years 1927-32, is followed by a significant reduction of one million between 1933 and 1938. The number of Ukrainians fell from 31.2 million in 1926 to 28.1 million in 1939. Some demographers attempt to explain this decline by ‘a different concept of national identity’ at census time. This is evidently not the only, nor even the principal reason. For instance, the Belorussians have not, during these same years, altered their ‘concept of national identity’, and their number increased by nearly 30 per cent (1.6 million; see table VI). What is important is that at the time of the 1926 and 1959 censuses, an identical percentage of Ukrainians (87 per cent) remained faithful to their mother tongue. This is significant because it was always on language that ‘Russification’ had an immediate effect.
In Kazakhstan, one of the main deportation areas, the population varies greatly; sometimes it goes up (by almost 500,000 inhabitants in six months), sometimes it goes down abruptly (see table VII). In all probability, it is not an accident that Kazakhstan registered the highest mortality rate for the whole country, even in 1940. It seems to have affected equally the newly-arrived and the local nomadic population which was adapting uneasily to the ‘new civilization’. Nobody even attempts to explain by ‘a difference in concept’ the fact that the Kazakh population fell by 860,000 between 1926 and 1939 (table VI). Other national minorities from that part of the country shared the Kazakhs’ fate. The number of Uigars fell from 108,000 in 1926 to 95,000 in 1959, that of the Altai people from 50,000 to 45,000, the Yakuts from 241,000 to 237,000, the Tungus from 39,000 to 25,000, and the whole of the Northern people from 140,000 in 1926 to 129,000 in 1959.
In 1933-4 there was a general fall in population. Conquest quotes the material from the OGPU, which was communicated to Stalin, in which the number of famine victims was estimated to be 3.3-3.5 million. Foreign correspondents in Moscow estimated it to be about 5 million4l. UrlaniS48 assumes that the population was 158 million at the end of 1933; in other words, that the decline in one year was over 7.5 million (see table 11). Even the population of manual workers who were best stocked in food supplies at the time, fell in 1933 (table II).
The 1933-4 famine resulted in the disappearance of an enormous number of children, particularly new-born babies. While the figure for the persons born in 1929-31 appears, in the 1970 census, as 12.4 million approximately, the figure for births between 1932 and 1934 is only 8.4 million. This fall in population cannot in any case be interpreted as an example of conscious birth control on the part of the population. Although 1929-31 were the most intense years of collectivization, the fall in the birth-rate was relatively low in relation to the previous three years (table II). The 1933 famine, was too much of an unexpected event, and furthermore, there was no knowledge of birth control in the Russian villages of the time. It is likely that at least 3 million children, born between 1932 and 1934, died of starvation.
A characteristic feature of that period is the complete prosperity presented to outsiders. The terrifying drop in population was ignored. Planners,50 foreign visitors, and party leaders never ceased to praise the rhythms of population growth. In January 1934, at the Seventeenth Congress, Stalin mentioned: ‘the growth of population in the USSR, which rose from 160 million at the end of 1930 to 168 million at the end of 1933’, thus forcing the figure from 8 to 10 million. His famous ‘We live better…’ is also dedicated to a powerful demographic increase. ‘We live better, we are happier.’ This is what was said to the kolkhozniks who had just lost their relatives, who have been deprived of their land and their cattle. ‘ … and this has the effect of the population reproducing much more rapidly than before. Mortality rate has gone down, birth rate has gone up, which brings about a clearly very sharp increase … at the moment our net population growth is around 3 million people.’ (Speech made at a meeting of male and female combine-harvester drivers, with some party members and some Government members in 1935.)
Is it ignorance, or deliberate falsification?
The December 1934 elections were, in themselves, a very revealing index of population losses. There were 91 million registered voters. ‘2.5 per cent of the whole country’s adult population are deprived of the right to vote, which represents a little more than 2 million people.’ (Molotov at the Seventh All-Russian Congress of the Soviets. ‘Main results of the Government’s activities’, Minutes from the Central Committee).
According to the 1926 census, the number of people aged ten and over was 109.4 million and in 1931, 103 million. If the mortality rate of the over-eighteen had remained the same as for the years 192731, at the end of 1934 there would have been 99 million electors, which in fact means a drop of 6 million people.12 Although this figure is fairly rough, it nevertheless shows that the losses suffered between 1926 and 1938 occurred mainly during the first half of that period.
In the years 1935-38, the Archipelago expanded rapidly. ‘You know that the economic problems in the Far East are now our top priority,’ declared Molotov at the Eighteenth Congress. ‘Consequently the problems of organization relating to the displacement of the Far East populations have become of prior importance.’ These problems were resolved. We note, in Siberia and in the Far East, that there was a 25 per cent population increase between the 1926 and the 1939 censuses. The records were withheld by the oblast’ in Magadan where the population figure was multiplied by 8.4.11 The newly-arrived people settled mainly in the rural zone (120,000 out of 150,000) and distinguished themselves by a surprisingly low birth-rate (10.9 per cent) and by an even lower mortality rate (8 per cent). It seems nevertheless that the mortality rate here has to do with some peculiar calculations. According to widespread opinion, the 1939 census counted in the Magadan oblast’ not the prisoners but only their jailers and the members of their families.
The Archipelago led at one moment to population expansions in the towns where the work camps were and then to the decimation of the population of these same towns (see table VIII). The drop in population in some old towns in the Urals and Siberia might perhaps be explained by the fact that the influx of population did not compensate for the numbers who were victims of repressive actions (table VIII). We can draw up an indicative estimate of the Archipelago’s inhabitants from the results of the 1937 elections. The number of electors, 94 million, was 3.5 million lower than the number of people aged eighteen and over (as calculated in the 1939 census). It is possible that a portion of those who did not vote – they were 3 million – might not have abstained from their own volition.
The official survey does not mention any losses in human lives during that period. It provides almost no clue. The increased mortality rate in 1937, when compared with 1935 and 1938-9 (see table II) gives a tiny indication. It is just possible that the very high mortality rates in various remote areas have borne some relation to the activity of the Archipelago. For instance, in the autonomous Republic of the Komi, the mortality rates for 1940, 1950, 1960 reached respectively 37.1, 14.4 and 6.0, when for the whole of the RSFSR the figures were 20.6, 10.1 and 7.4 .55
An evaluation of the population trends has been compared to the data from the 1931 and 1939 censuses (see tables IX and X). While in 1931 the official population figure was 900,000 more than the estimates, on the other hand in 1939 6.9 million people aged fifteen and over were not accounted for. Consequently, for this section of the population, taking into account the 0.645 million correction made from the 1931 data (table IX) the figure for the losses reaches 7.5 million (i.e., 5 million men, 2.5 million women). This figure does not include the millions of children who died of starvation (cf. above). According to the already formulated method, to underestimate the losses by between I and 3 million is possible (given the inaccurate population count in the 1926 and 1939 censuses), but in the way of overestimation the error would probably not be more than 300,000. The 1934 elections lead one to believe that 4/5 of these losses were between 1932 and 1934 and 1/5 (1.5 million) during the later period.
Source: Roy Medvedev, ed., Samizdat Register (New York: Norton, 1977), Vol. II, 220-223, 232-237.