1. A great deal of the information contained in the three volumes of Gulag Archipelago has been known, but not always believed, since the 1930s. See Aleksandr, Solzhenitsyn, Arkhipelag GULag 1918-1956: Opyt khudozhestvennogo issledovaniia, I-VII, 3 vols. (Paris: YMCA-Press, 1973-75)Google Scholar and Solzhenitsyn, , Gulag Archipelago 1918-1956: An Experiment in Artistic Investigation, I-VII, 3 vols., trans. Whitney, Thomas P. (vols. 1-2) and Willetts, Harry (vol. 3) (New York: Harper & Row, 1973-78)Google Scholar (English translation hereafter cited as Gulag 1, Gulag 2, and Gulag 3). Solzhenitsyn's description of concentration camp life is entirely consistent with earlier characterizations found in the writings of David Dallin, Robert Conquest, and S. Swianiewicz to mention only a few. A thorough survey and appraisal of this literature is provided in Steven Rosefielde, “An Assessment of the Sources and Uses of Gulag Forced Labor 1929-1956,” Soviet Studies, 33, no. 1 (January 1981). Solzhenitsyn's testimony nonetheless remains unique. In his thoroughness, in the uncompromising way he exposes all the rationalizations and lies used to conceal and mitigate the significance of Soviet forced labor, Solzhenitsyn conveys a sense of authenticity that cannot be gainsaid even by those who find fault with his work on other grounds.

2. These data are reported in Rosefielde, “An Assessment.”

3. Alexander, Erlich, The Soviet Industrialisation Debate 1924-1928 (Cambridge, Mass.: Harvard University Press, 1960)Google Scholar; Alec, Nove, An Economic History of the USSR (Baltimore: Penguin’ Books, 1969)Google Scholar; Alec, Nove, Economic Rationality and Soviet Politics, or Was Stalin Really Necessary? (New York: Praeger, 1964.Google Scholar

4. The following citations capture the essence of Nove's position: “However it now became urgent to find material and financial means to expand the industrial base. This at once brought the peasant problem to the fore… . Off farm surplus must grow rapidly to sustain industrialization… . As the ‘left’ opposition vigorously pointed out, the peasant … had somehow to be made to contribute produce and money, to provide the bulk of ‘primitive Socialist accumulation'… . The moderate wing, led by Bukharin, believed that it was possible to advance slowly towards industrialization … riding towards socialism on a peasant nag… . The case against the Bukharin line was … [that] free trade with the peasants could only provide adequate surpluses if the better off peasants … were allowed to expand… . A strong group of independent rich peasants was the Bolshevik's nightmare. … It was widely felt … that by 1927 the regime had reached a cul-de-sac… . P. A. Gavri … describes its dilemma quite clearly… . To justify its existence … a rapid move forward was urgent; but such a move forward would hardly be consistent with the 'alliance with the peasants'… . Stalin at this point swung over towards the left, and his policy of all-out industrialization and collectivization was a means of breaking out of the cul-de-sac, of mobilizing the Party to smash peasant resistance, to make possible the acquisition of farm surpluses without having to pay the price” (Nove, Economic Rationality, pp. 22-23; cf. Alec Nove in Millar, James R. and Nove, Alec, “A Debate on Collectivization: Was Stalin Really Necessary?” Problems of Communism, 25, no. 4 [July-August 1976]: 56–57).Google Scholar

5. Millar and Nove, “A Debate on Collectivization,” p. 56.

6. “The grain crisis of the winter 1927-28 was due much more to price relationships” (ibid., p. 57).

7. Moshe, Lewin, Russian Peasants and Soviet Power (Evanston, 111.: Northwestern University Press, 1968), p. 508.Google Scholar

8. “It was then (1931) that he (Stalin) made the justly famous prophecy: ‘We are fifty or a hundred years behind the advanced countries. We must make good this distance in ten years. Either we do so, or we shall go under.’ 1941 was ten years away” (Nove, Economic History of the USSR, p. 188). Cf. Nove, Economic Rationality, p. 31: “The urgent need to prepare for war has often been advanced as an excuse for Stalin's industrial 'tempos’ and the terror. This can hardly be accepted. In the worst years of social coercion and overambitious plans, i.e. 1929-33, Hitler was only just climbing to power. “

9. See Nove, Economic Rationality, pp. 22-23, where he attributes his views to two Menshevik pamphleteers, P. A. Gavri and N. Valentinov.

10. Millar and Nove, “A Debate on Collectivization,” pp. 49-62; James R., Millar, “Soviet Rapid Development and the Agricultural Surplus Hypothesis,” Soviet Studies, 22, no. 1 (July 1970): 77–93 Google Scholar; and James R., Millar, “Mass Collectivization and the Contribution of Soviet Agriculture to the First Five-Year Plan,” Slavic Review, 33, no. 4 (December 1974): 750–66 Google Scholar. Also see Barsov, A. A., Balans stoimostnykh obmenov meshdu gorodom i derevnei (Moscow: Nauka, 1969 Google Scholar; A. A., Barsov, “Sel'skoe khoziastvo i istochniki sotsialisticheskogo nakoleniia v gody pervoi piatiletki (1928-1933),” Istoriia SSSR, 1968, no. 3 Google Scholar; Michael, Ellman, “Did the Agricultural Surplus Provide the Resources for the Increase in Investment During the First Five Year Plan?” Economic Journal, 85, no. 340 (December 1975): 844–63 Google Scholar; Barbara, Katz, “Purges and Production: Soviet Economic Growth, 1928— 1940,” Journal of Economic History, 35, no. 3 (September 1975): 567–90.Google Scholar

11. Cf. Vasilii, Ivanchenko, “Leninskie idei o planovom upravlenii ekonomikoi v deistvii,” Voprosy ekonomiki, 1979, no. 4, p. 7.Google Scholar Cf. Nove, Economic History of the USSR, p. 191.

12. Population was growing about 2.6 percent per annum. TakifTg 1928 as a base, population in 1932 would therefore be 111, which is slightly less than the reported increase in consumption of 113. This implies that per capita consumption rose by 0.5 percent per annum. According to Warren Eason, population actually increased 1.7 percent from 1928 to 1932, suggesting that per capita consumption rose at 1.4 percent per annum (see Warren, Eason, “Labor Force,” in Bergson, Abram and Kuznets, Simon, eds., Economic Trends in the Soviet Union [Cambridge, Mass.: Harvard University Press, 1963], p. 72Google Scholar; also see Nove, Economic History of the USSR, p. 180).

13. The term “Great Leap Forward” is used by Solzhenitsyn to describe Stalin's industrialization drive (see Solzhenitsyn, Gulag 2, p. 91). On the grimness of living conditions during the 1930s, see, for example, N adezhda, Mandelstam, Hope Abandoned (New York, 1974).Google Scholar

14. See Steven, Rosefielde, “Knowledge and Socialism,” in Rosefielde, Steven, ed., Economic Welfare and the Economics of Soviet Socialism (New York: Cambridge University Press, 1981 Google Scholar, note 21.

15. Cf. G. Warren Nutter's estimate of 8.3 percent used by Holland Hunter in Hunter, , “The Overambitious First Soviet Five-Year Plan,” Slavic Review, 32, no. 2 (June 1973): 245.CrossRefGoogle Scholar

16. G. Warren Nutter takes a more critical attitude toward the reliability of Soviet statistics than most specialists, dwelling at length on the question of poor quality, expedient inflation of output statistics, and the problem of international comparison of what might be described as inferior Soviet goods with superior Western products (see Nutter, G. Warren, The Growth of Industrial Production in the Soviet Union [Princeton, N.J.: Princeton University Press, 1962], pp. 11–84 Google Scholar). His position, however, is decidedly moderate in comparison with Solzhenitsyn who perceives tukhta to be more pervasive and more quantitatively significant than Nutter. Moreover, and perhaps in the final analysis more importantly, Nutter does not incorporate his perceptions in his computations. A summary statistic will have to suffice to make this point clear. According to Nutter, the average annual rate of industrial growth during 1928-37 was 12.1 percent (ibid., p. 287). Raymond Powell's estimate on the same average annual basis is 11.5 percent (see Raymond Powell, “Industrial Production,” in Economic Trends in the Soviet Union, p. 160, table 4.2).

17. In retrospect this line of defense is less satisfactory than initially supposed. One does not have to argue for free invention to suspect deliberate misstatement, nor is it necessary to contend that the Soviets keep two sets of books to suggest that reported statistics have been significantly inflated. There is no reason to believe that because the system directors required accurate information they were able to acquire it, especially during the early 1930s when the bearer of bad news was often treated as a counterrevolutionary. For an authoritative statement on the reliability of Soviet statistics, see Abram, Bergson, Soviet National Income and Product In 1937 (New York: Columbia University Press, 1953, pp. 7–9 Google Scholar. See also Naum, Jasny, “Intricacies of Russian National Income Indexes,” Journal of Political Economy, 55, no. 4 (August 1947): 299–322 Google Scholar; Naum, Jasny, “Soviet Statistics,” Review of Economics and Statistics, 32, no. 1 (February 1950): 92–99 Google Scholar; Alexander, Gerschenkron, “The Soviet Indices of Industrial Production,” Review of Economics and Statistics, 29, no. 4 (November 1947): 217–25 Google Scholar; Treml, Vladimir G. and Hardt, John P., eds., Soviet Economic Statistics (Durham, N.C.: Duke University Press, 1972)Google Scholar; and Norman, Kaplan, “Arithmancy, Theomancy, and the Soviet Economy,” Journal of Political Economy, 61, no. 2 (April 1953): 93–116.Google Scholar

18. A growing number of specialists have begun arguing that contemporary inspection practices are lax, especially with regard to prices. Since the current institutional arrangements are superior to those of the past, this new position supports Solzhenitsyn's contention. However, I am not persuaded by the insurgent theories of hidden inflation. In my view, Soviet price discipline has improved substantially since the 1930s. See, for example, Morris Bornstein, “The Administration of the Soviet Price System,” Soviet Studies, 30, no. 4 (October 1978): 466-90.

19. Solzhenitsyn, Gulag 2, p. 69. Cf. Roy, Medvedev, Let History Judge (New York: Alfred A. Knopf, 1972), p. 105Google Scholar. On tukhta, see Lev, Kopelev, To Be Preserved Forever (Philadelphia: ’ J. B. Lippincott Co., 1977), p. 147Google Scholar; Victor, Herman, Coming Out of the Ice (New York: Harcourt Brace Jovanovich, 1979, pp. 261–63 Google Scholar; and Evgeniia, Ginzburg, Into the Whirlwind (New York: Harcourt Brace and World, 1967), p. 410 Google Scholar. The other side of the tukhta coin is the phenomenon of hidden, new product price inflation. Tukhta in this guise is widely recognized by specialists. Abram Bergson, for example, writes, “Thus, as everyone agrees, a major cause of upward bias in the official series before the war was the notoriously defective method used to value new products in 1926/27 rubles” (see Bergson, “Soviet National Income Statistics: Summary and Assessment,” in Soviet Economic Statistics, p. 151). Bergson deals with the new product problem for the most part by making his calculations in 1927/28 and 1937 prices and appears to conclude that the bias introduced by new products is relatively small, that is, closer to Dobb's view on this matter than to his own previous assessment (see Abram, Bergson, The Real National Income of Soviet Russia since 1928 [Cambridge, Mass.: Harvard University Press, 1961], pp. 185–86 Google Scholar). Note that the technique of valuing new products in prices of alternative years does not effectively resolve the problem of spurious innovation, tukhta type 3.

20. Social bads also differ from overvalued goods on another score. It can be assumed for convenience that overvalued goods are desired by society and the system directors. Overvaluation from either perspective is regarded as fraud. In the case of social bads, however, value judgments may be sharply divergent. Concentration camps, for example, may have been deemed good by Stalin and bad by society. No simple solution or rule can resolve contradictions of this sort. The reader himself must define his own social welfare function and draw his inferences accordingly.

21. In the calculations which follow, adjustments for tukhta and social bads are global and do not depend precisely on the distortion introduced by any individual project. It should be noted in particular that the adjustment factors employed do not depend importantly on the controversial category of social bads. If the reader should desire to make his own subjective allowance for bads, this adjustment should be performed in two steps. First, the money cost of the project should be subtracted from official output statistics, and then a further deduction should be made for its deleterious consequences. A nuclear reactor which pollutes the environment and must be shut down, for example, represents a wasted expenditure of billions of dollars in addition to imposing substantial cleanup cost on the economy.

22. Solzhenitsyn, Gulag 2, p. 161.

23. Ibid.

24. Whether these losses were retrospectively deducted from the national product accounts is highly problematical. Solzhenitsyn argues that such corrections were against the interest of the Central Statistical Administration and were probably not carried out (ibid., pp. 162-64).

25. Ibid., p. 166.

26. Ibid., chapter 3, “The Archipelago Metastasizes,” pp. 71-120.

27. See map, ibid., p. 97.

28. The fact that the Belomor Canal was built with prison labor was never a secret. Forced labor was publicly described as the only true form of socialist redemption for crimes committed against the people (see Andrei, Vyshinskii, ed., Ot tiurem k vospitatel'nym uchrezhdeniiam [Moscow, 1934]Google Scholar). According to David Dallin, three hundred thousand zeks labored on the canal. He quotes a French engineer sentenced to forced labor in Russia who estimated that there were two hundred thousand zeks employed on the project, and that fifty thousand died during a period of one and a half years. Dallin also asserts that seventy-two thousand prisoners—all thieves—were amnestied for their exemplary work on the canal (see David’ Dallin, , The Real Soviet Russia [New Haven: Yale University Press, 1944], pp. 241–44 Google Scholar). Nove also recognizes that zeks were used to build Belomor: “Vast projects got under way. While some were the work of forced labour (the Volga-White Sea Canal, for instance) there is no doubt that there was enthusiasm too. There was, for instance, the story of Magnitogorsk” (see Nove, Economic History of the USSR, p. 190). From Solzhenitsyn we learn, however, that much of Magnitogorsk was built by zeks (see Solzhenitsyn, Gulag 2, p. 592). Antonov-Ovseenko, Anton, Portret Tirana (New York: Khronika Press, 1980, p. 105 Google Scholar, estimates that three hundred thousand zeks worked at Belomor.

29. Solzhenitsyn, Gulag 2, pp. 102 and 85. Antonov-Ovseenko indicates that more than two hundred thousand zeks died (see Antonov-Ovseenko, Portret Tirana, p. 105).

30. Solzhenitsyn, Gulag 2, pp. 91-92. Cf. conditions in 1948 (Solzhenitsyn, Gulag 3, p. 63).

31. Solzhenitsyn, Gulag 2, p. 102.

32. The assertion that the canal was almost useless does not mean that it was unused. Before the war it carried some limited traffic. It was damaged by the German air force during the war and was not reactivated until 1946 (see David Dallin and Boris Nicolaevsky, Forced Labor in Soviet Russia [New Haven: Yale University Press, 1947], pp. 213-14). Today it is sometimes used to transport military equipment, but these weapons must be disassembled and carried in sections on barges. Neither large surface combatants nor submarines can navigate the waterway.

33. Dallin, Real Soviet Russia, pp. 242-43, and Dallin and Nicolaevsky, Forced Labor, p. 213. Dallin estimates that there were between two hundred thousand and three hundred thousand forced laborers working on the Belomor Canal. These figures correspond broadly with Solzhenitsyn's. According to Eason's research the average number of paid workers employed in the construction sector during the twenty-one months in which the canal was built was 3.33 million (see Warren Eason, “Labor Force,” in Economic Trends in the Soviet Union, pp. 77 and 82, tables 2.10 and 2.13, respectively). This figure is used here for simplicity as a rough measure of the total construction labor force. The share of total construction workers employed at Belomor is computed with this base. If forced laborers engaged in construction were added to those earning wages and salaries, the base would be significantly increased. Note also that Dallin and Solzhenitsyn disagree slightly about the amount of time it took to build the canal. Dallin reports the construction period as November 1931- August 3, 1933; Solzhenitsyn as September 1931-April 1933.

34. “In our camps, for most of the years of the Archipelago, they either paid nothing for labor or just as much as was required for soap and tooth powder” (see Solzhenitsyn, Gulag 2, p. 204). Cf. Abram Bergson, Soviet National Income and Product, p. 122, where it is assumed that penal workers earned pay equal to one-half that of the free workers. See also Gulag 3, p. 65. The total cost of construction produced from November 1931 to August 1933 in current ruble prices was 38.45 billion rubles. See Moorsteen, Richard and Powell, Raymond, The Soviet Capital Stock 1928-1962 (Homewood, III.: Richard D. Irwin, 1966), p. 407 Google Scholar, table A-13. Belomor cost 0.095 billion rubles, or 0.25 percent of the total. Although prisoners did not receive disposable cash wages of any significance, the cost of their maintenance is included in the 95 million rubles expended to construct the canal. These subsistence payments in kind are often described as “wages” in official sources. See Dallin and Nicolaevsky, Forced Labor in Soviet Russia, p. 88.

35. The figure of two hundred thousand forced laborers is Dallin's lower estimate. See Dallin, Real Soviet Russia. The average annual industrial wage in 1928, used here as a surrogate for the wage in the construction sector, is 843 rubles. See Abram Bergson, Real National Income, p. 422, table H-l.

36. The wage shares of total construction valued in constant 1928 prices were: 1928 — 3 percent; 1931-32.4 percent; 1932-36.2 percent-; 1933-41.8 percent. The constant construction output series is derived from Moorsteen and Powell, Soviet Capital Stock, p. 407; the average annual wage from Bergson, Soviet National Income and Product, p. 422, table H-l. These figures imply that for every ruble spent on labor, 2.7 rubles were expended on materials during 1931-33. This relationship is used to derive the materials estimate in the text. An alternative procedure would have been to deduct wages in kind from total expenditures, roughly thirty-two million rubles, and deflate material costs (see Dallin and Nicolaevsky, Forced Labor in Soviet Russia, p. 88). This approach produces an implausibly low materials estimate of thirty-nine million rubles, because it omits the imputed value of all material inputs, such as timber, produced at the work site instead of being purchased from outside sources. It should also be noted that the issue under discussion casts some doubt on the validity of the material input approach used by Moorsteen and Powell to derive their index of real construction growth. Eason's data indicate that the wage and salary construction labor force alone grew 237 percent during 1928-32, while real construction grew only 93 percent during the same interval. Even allowing for the low wage share of construction cost and ignoring the growth of nonwage forced labor, total real construction could not have grown 80 percent as officially claimed if both labor and material grew faster than this in real terms (see Eason, “Labor Force,” and tables 1 and 2 on pp. 563 and 566 in this article).

37. Average annual construction expenditures for the period 1931-33 were 7.8 billion rubles valued in 1928 prices (see Moorsteen and Powell, Soviet Capital Stock, p. 407).

38. Construction in 1928, excluding repairs, was 3.52 billion rubles. Assuming two hundred thousand zeks participated annually in contructing the canal, we have estimated that the canal cost 0.419 billion rubles annually. If three hundred thousand seks were employed, the estimate would rise to 0.629 billion rubles. The latter figure explains 17.9 percent of Soviet construction growth during 1928-32. Throughout these calculations, construction has been defined excluding repairs (see ibid.).

39. See notes 20 and 21. One reviewer objected to the tacit value judgment inserted above, arguing that for purposes of national income accounting, planners’ preferences should rule, not humanist preferences. This position has some merit. Humanist preferences are not unique, and if national income statistics were adjusted for each individual's notion of social welfare, the result would be statistical bedlam. Advocates of disarmament, for example, might treat weapons as bads while proponents of defense might treat them as goods. Although I recognize this difficulty and am generally sympathetic with the viewpoint of my critic, concentration camps, it seems to me, are extraordinary phenomena and should be treated specially. I believe most scholars will agree that to include them in the GNP as a measure of economic success badly distorts the meaning of this indicator.

40. Solzhenitsyn, Gulag 1, chapter 2, “The History of Our Sewage Disposal System,” p. 24. A figure of twelve million is given on page 595 as the largest number in the camps at any one time, although Nicolaevsky and Dallin are cited as sources giving higher estimates. See also the note in Gulag 2, p. 205, and Gulag 3, p. 28.

41. See Rosefielde, “An Assessment.” A reasonable estimate of the number of seks in the GULag archipelago in 1933 is five million.

42. Solzhenitsyn, Gulag 2, pp. 101 and 592. Holland Hunter has pointed out that it would be ridiculous to suppose that most railroads during the 1930s were built primarily to service the camps. He has also suggested that the role of railroads in Soviet industrialization should not be exaggerated, since only limited resources were allocated to this activity (see Holland Hunter, “Soviet Railroads Since 1940,” Bulletins on Soviet Economic Development, September 1950, no. 4, pp. 10-35.

43. Solzhenitsyn, Gulag 1, p. 492.

44. Holland, Hunter, Soviet Transportation Policy (Cambridge, Mass.: Harvard University Press, 1957), p. 49.Google Scholar

45. Ibid., p. 183.

46. Ibid., p. 184.

47. Cf. Nove, “Life in the villages was so miserable, and the people were so frightened of being labeled kulaks and arrested, that perhaps” up to 10 million of them fled to the construction sites and factories that were being built” (Millar and Nove, “A Debate on Collectivization,” p. 59). Solzhenitsyn would have been more explicit about which construction sites they fled to and who arranged their transportation (see Solzhenitsyn, Gulag 2, chapter 2, “The Peasant Plague,” pp. 350-68, chapter 3, “The Ranks of Exile Thicken,” pp. 369- 84, and chapter 4, “Nations in Exile,” pp. 385-405).

48. Aleksandr Solzhenitsyn, The First Circle (New York: Bantam, 1968), p. 674. Cf. Solzhenitsyn, Gulag 3, pp. 79 and 532.

49. Note that if Solzhenitsyn is right and prisoners bulked large in “Passenger Travel,” they must necessarily have been valued in 1926/27 prices to generate a 22.7 percent rate of annual growth.

50. Solzhenitsyn, Gulag 2, p. 594.

51. Ibid., p. 102. The Moscow-Volga Canal was seven times bigger in scale than the Belomor (ibid., p. 579).

52. Ibid., p. 140.

53. Ibid., pp. 591-93. The description of Soviet industrialization portrayed in the text is supported by Antonov-Ovseenko, Portret Tirana, pp. 102-11.

54. Eason, “Labor Force,” pp. 77-82. N.B. The term “wages and salary,” panel A, line 9 of Eason's table 2.13 includes nonwage workers in the labor force. See also note 60 of this article.

55. Rosefielde, “An Assessment.”

56. Solzhenitsyn, Gulag 2, p. 166.

57. See Berliner, Joseph S., Factory and Manager in the USSR (Cambridge, Mass.: Harvard University Press, 1957).CrossRefGoogle Scholar

58. Nove, Economic History of the USSR, pp. 187-95.

59. Rosefielde, “An Assessment.”

60. According to Eason's data, 66.1 percent of the nonagricultural labor force earning wages and salary were in industry, construction, transportation and communications (see Eason, “Labor Force,” pp. 77 and 82). My own research indicates that a million or so of the zeks in 1933 may not have been paid wages or salaries and were not included in the labor force. Making allowance for this by adjusting the total reported nonagricultural labor force in industry, construction, transportation and communications upward reduces the sek share (4.5/18.2) to 25 percent (see Rosefielde, “An Assessment” ). The share calculated here differs from note 54 because that measure excludes transportation and communications.

61. Nutter, Growth of Industrial Production, p. 524, table D-2. Valued in 1928 prices, consumer goods are reported to have grown at 1.7 percent per annum during 1928-33; in 19SS prices, at 3.6 percent per annum. Nutter explains that the coverage of the two series is not identical but this reversal of normal expectations is still perplexing. Nutter's findings, nonetheless, are entirely consistent with Janet Chapman's data on real Soviet wages, which decline during 1928-37, even though 1937 was the year in which agricultural output (officially) surpassed the 1928 level (see Janet, Chapman, Real Wages In Soviet Russia Since 1928 [Cambridge, Mass.: Harvard University Press, 1963], p. 144Google Scholar).

62. See Naum, Jasny, The Socialised Agriculture of the USSR (Stanford, Calif.: Stanford University Press, 1949)Google Scholar, chapter 22, especially p. 540.

63. “Except for certain adjustments made in calculating net agricultural output, we have accepted throughout official estimates of output for individual commodities. When Khrushchev stated that the use of the biological yield concept by Malenkov, and presumably by Stalin and himself, was dishonest, we believed that output data made available subsequently would approximately reflect the output of farm products available for sale, home consumption, or further use in production. It is now quite clear that this assumption was unjustified: the so-called barn yields of such crops as grains or sunflowers do not represent the amount of useful product of appropriate moisture content” (D. Gale Johnson, “Agricultural Production,” in Economic Trends in the Soviet Union, p. 212). Biological yields were introduced in 1933 (Nove, Economic History of the USSR, p. 185).

64. Nove, Economic History of the USSR, pp. 179-80. Cf. Solzhenitsyn, Gulag 3, pp. 24 and 28. See also ibid., pp. 350-68. Cf. Medvedev, Let History Judge, pp. 94-95.

65. Eason, “Labor Force,” pp. 77 and 82. Including the unemployed, industrial workers as defined in the text grew by six hundred thousand per year. From 1928 to 1933, the average annual increase was nearly treble this rate, or 1.75 million per year. The labor force shares prevailing in 1928 were used to estimate the number of industrial workers in 1926 and 1927.

66. S., Swianiewicz, Forced Labour and Economic Development (London: Oxford University Press, 1965 Google Scholar, chapters 5-9.

67. See Nutter, Growth of Industrial Production, p. 172. The inferences drawn in the text might be reversed if Soviet statistics on the decline in the workweek for this period encompassed all workers. They apply, however, only to large-scale industry and ignore the length of the workday in the archipelago. For a more detailed discussion of Soviet employment data, see Barney Schwalberg, Industrial Employment in the USSR (1933, 1937, 1950, 1955), U.S. Bureau of the Census, Series P-95, no. 55 (Washington, D.C., 1960).

68. The capital stock grew by 12.7 percent, 10.4 percent, and 6.1 percent, computed in an official, adjusted Western and IS percent tukhta corrected basis. The official rate was calculated from data supplied in Moorsteen and Powell, Soviet Capital Stock, pp. 387-88, 409, 412. Gross investment in fixed capital stock, excluding capital repairs, during 1928-32 was cumulated and added to the net value of the capital stock as of January 1, 1928 (58 million rubles). This figure was then depreciated by 4 percent per annum, a rate suggested by Bergson (see Real National Income, p. 14). The same procedure was used to compute the adjusted Western estimate, with 1937 prices substituted for 1928 prices. The last estimate is 85 percent of the adjusted Western figure.

69. See Moorsteen and Powell, Soviet Capital Stock, p. 361

70. Besides calculating 1937 GNP in 1928 prevailing prices, Bergson employs 1928 ruble factor cost prices as well, which exclude turnover taxes and subsidies. GNP grows, on this measure, 11.9 percent during 1928-37. A similar computation using our data yields a much lower comparative rate of growth, 6.6 percent. The divergence here is attributable mostly to the slow growth of industrial consumer goods, and, to a lesser extent, industrial durables, valued in 1928 prices exhibited by Warren Nutter's series (1928-32). By comparing Nutter's estimates for 1937 with those for 1932, it is evident that the low values for 1928 are due to the underlying volumes, and not prices, because his series shows much higher rates of growth to 1937. I, therefore, infer that the discrepancy referred to above is not one which principally involves prices and/or index number relativity, but is determined by the low output volumes reported by Nutter for 1928. Cf. Medvedev, Let History Judge, p. 107.

71. Cf. Norman Kaplan, “Arithmancy, Theomancy, and the Soviet Economy,” pp. 93-116. Also consider Dobb's conclusion: “The view that all Soviet figures are naturally suspect, designed as propaganda-instruments to deceive the unwary, is no longer seriously held, and scarcely merits attention here” (Maurice Dobb, “Comment on Soviet Economic Statistics,” Soviet Studies, 1, no. 1 [June 1949]: 18-27).

72. Gregory Grossman has properly pointed out that managers may sometimes have an incentive to understate their results. While this argument should be valid even for the First Five-Year Plan, I doubt that it significantly offsets the incentives to inflate production statistics. See Gregory Grossman, “The ‘Second Economy’ of the USSR,” Problems of Communism, 26, no. 5 (September-October 1977): 30-32.

73. Antonov-Ovseenko, Portret Tirana.