While these measures certainly had a restrictive effect, they were only first steps on a long path to the comprehensive, centralized censorship of later years. The next important measure was the establishment of an office of the censor, known as Glavlit, on 6 June 1922. (Its full name, the Chief Administration for Literature and the Press, went through many variations, but the short form was retained to the end.) The first list of prohibited topics was promulgated: anti-Soviet agitation and propaganda, military secrets, false information causing alarm, information inciting ethnic and religious hatred, and pornography.[28] But the system was not watertight because censorship remained decentralized and was implemented with many inconsistencies and loopholes.
By 1930 or so, the wrinkles were ironed out and the leaks were stopped.[29] From that time, the censors ensured that scarcely a word was printed or broadcast from one end of the Soviet Union to the other without prior scrutiny. They kept from circulation the least hint that could reflect badly on the ruling party or its leaders.
Comprehensive censorship, the second pillar of the Soviet regime of secrecy, was exceptionally effective. Throughout the history of the Soviet state there were few leaks. Great facts that were effectively concealed figure 1.1. Censorship worked: The frequency of the term “Glavlit” in 600,000 Russian books over the twentieth century (cases per 100,000 words per year)
Notes: Glavlit was the shortened name of the office of the Soviet censorship, founded in 1922. The table shows the frequency of cases of “Glavlit” per 100,000 words per year. This unit is chosen, based on the average length of a book in the Google Books Russian-language corpus of 2019 (described by Michel et al., “Quantitative Analysis of Culture”) over the period shown, which was 97,777 words. The search parameters were “Glavlit” in Cyrillic characters, switching off case-sensitivity and year-to-year smoothing, A figure of 0.1 on the vertical axis would mean that, if mentions of Glavlit were randomly distributed, the probability of finding a single mention of Glavlit in a randomly chosen book in a given year and of average size over the sample was one in ten. If mentions were clustered (for example, by passages of text in which Glavlit was mentioned repeatedly), the true probability would be much lower.
include the famines of 1933 and 1947, the mass killings of 1937 and 1938, Soviet responsibility for the Katyn forest massacre of 1940 (the event itself was uncovered only by the accident of Germany’s wartime occupation of the territory on which it took place), the scale of forced labor, the military and population losses of World War II, the burden of Cold War military outlays, and the exact geolocation of almost everything, including the existence of some cities.
A simple yet unambiguous measure of the effectiveness of the censorship is shown in Figure 1.1. How well did Glavlit suppress discussion of its own existence? The Google Books Russian-language corpus of 2019 gives access to the digitized texts of 600,000 books published in the twentieth century. The vertical axis shows the frequency of mentions of “Glavlit” per 100,000 words per year; this was close to the average length of a book in the corpus. The data start in 1901, when Glavlit did not exist. In each year before 1922, when Glavlit was established, the probability of finding a single mention of Glavlit in a published book was zero or, if not zero, so close to zero as to be explained, most likely, by a typographical error or an error of digitization. In the years from 1922 to 1930, the censorship was being built, and Glavlit was mentioned and discussed in public, so the probability of finding a mention of “Glavlit” in a book of average size soared, peaking at around one-quarter. Of course, a single mention is not a discussion, so the probability of finding a discussion of Glavlit might be one or two orders of magnitude less. But in 1931 the frequency fell back to zero, or close to zero. The hush that followed was disturbed only by an occasional murmur (in 1936, for example, the frequency briefly returned to one-tenth before falling back). After that, from the late 1930s to the late 1980s, that is, for a full half century, silence reigned.
Only after Gorbachev initiated the policy of “openness” (glasnost’) in 1987 did uncensored discussion of Glavlit resume. It never returned to the level of the 1920s, for public interest was now limited to the narrow circle of historical scholars. Anyway, the lesson is that Soviet censorship worked.
What was permitted by the Soviet censors varied over time, although generally within strict limits. Writing in the first half of 1953, the economist Abram Bergson offered a simple gauge of the tendency of the secret sphere to encroach on what was previously disclosed. He used the quantity of published documentation of successive Soviet five-year plans for the national economy.[30] The first five-year plan (1929) amounted to four published volumes, full of detailed economic and social statistics (although already with notable gaps and half truths). That quantity was halved for the second plan (two volumes in 1933) and halved again for the third (one volume in 1938). The third was the last plan to be published before war broke out. The postwar period saw further restriction. The fourth plan (1946) was published in six newsprint pages in Pravda, with the fifth plan (1950) covered in just three pages.
After Stalin’s death, the policy of increasing restriction was reversed. During the post-Stalin “Thaw” (discussed below) the censors allowed a return to regular publication of economic and social statistics. Although the data published were heavily selected and the gaps remained notable, the change in policy was large enough to have important effects on the atmosphere and substance of the Soviet Union’s public discourse. Thus, a policy was changed, while the system remained the same. The strongest evidence of continuity is that, as shown in Table 1.2, the existence of the office of censorship was strictly shielded from public exposure.
The Bolsheviks did not invent censorship, which was practiced in Imperial Russia before the Soviet period. If there was continuity within the Soviet period, was there continuity from pre-Soviet times? The two systems cannot be evaluated in exactly comparable terms. The work of the historian Benjamin Rigberg makes possible only a rough comparison. He finds that the prerevolutionary statute on censorship seemed to give the imperial government sweeping powers to control the press and prevent publications that might disseminate critical views or subvert public order. But these powers remained largely unused for lack of resources. In 1882, forty-four censors were employed across the Russian Empire. By 1917, after thirty-five years of growth of the print media and in the middle of a world war, the number had risen to just forty-six. While the censors’ efforts were not completely ineffective, they did little to intimidate a lively public sphere where great issues were debated in a spirit that was often hostile to authority.[31] Rigberg concludes:
It is now clear that the tsarist regime made improper provision for the attainment of its aims. Even a staff twenty or thirty times larger than that actually functioning could scarcely have supervised effectively the vast printing operations of tsarist times. It thus seems obvious that the government cannot itself have attached overriding importance to censorship operations; else why so paltry a budget for the Chief Administration? However one speculates the conclusion seems inescapable: Censorship was not a major arm of repression which the Old Regime relied upon.[32]
29
The centralization of the censorship is described by Goriaeva, Politich- eskaia tsenzura, 162-240. The relative liberalism of the 1920s allowed E. H. Carr to take the story of his multivolume History of Soviet Russia through the end of 1928. Beyond that point, Carr decided, the growing secrecy of Soviet politics and policy making became too great a barrier to the writing of history. Carr, Bolshevik Revolution, Book 1, 6, and Carr, “Russian Revolution and the West,” 27.
31
Rigberg, “Tsarist Press Law”; “Efficacy of Tsarist Censorship Operations,” and “Tsarist Censorship Performance.” In 1906, for example, the scope of budgetary secrecy was debated in print as Russia moved toward constitutional rule. Avinov, “Sekretnye kredity v nashem gosudarstvennom biudzhete.” Avinov was a prominent Cadet (Constitutional Democrat). I thank Vasilii Zatsepin for this reference.