There were deficiencies…. I am not saying that Yezhov was spotless, but he was a good party worker. There should have been more supervision…. There was some, but not enough.—V. M. Molotov, 1991
It is interesting that before the events of the thirties, we lived all the time with oppositionists, with oppositionist groups. All around—one against another, what good is that?—V. M. Molotov, 1991
THE WILD AND VICIOUS terror of 1937 is sometimes known as the Yezhovshchina: the “time of Yezhov.” This is a misnomer for several reasons. First, it puts excessive emphasis on N. I. Yezhov, who, although he was the head of the secret police that carried out much of the terror, was only one of the important political actors and forces involved. While he had a certain amount of freedom in identifying and arresting various “enemies,” he almost certainly took his orders from Stalin and the Politburo.
Second, Yezhovshchina is a misleading epithet because that terror consisted of a number of discrete movements, offensives, measures, and countermeasures. Party membership screenings were not the same as police arrests. Various groups and constituencies played changing roles; one group might be sponsoring the persecution of another at one moment, but a few months later their roles might be reversed. Regional party secretaries, midlevel officials of the party apparatus, rank-and-file party members, economic managers, former members of the left and right oppositions, Politburo and Central Committee members, and ordinary citizens interacted with and against one another in the 1930s in a bewildering series of combinations and alliances.
Because all of these groups were in some measure and at some time victims of the terror, it is tempting to see these events as part of a single event or grand plan directed by Stalin. Although there is no doubt that he was the author and organizer of much of what we call the Great Terror of the 1930s (another inexact shorthand for the disparate events of that decade), we have seen that on many occasions his policies were marked by contradiction and vacillation. More than once, documents reflecting his private remarks to the elite had to be altered or sanitized because they did not fit the prevailing policies or because they sharply contradicted subsequent events.
Under whatever name one chooses, terror continued throughout 1938. More than 638,000 people were arrested in that year (compared with over 936,000 in 1937), the vast majority being accused of “counterrevolutionary” crimes. At least 328,000 persons were executed in 1938, and the population of the GULAG labor camps increased that year by roughly the same number.1 Moreover, that year the Moscow show trials returned, this time for Bukharin, Rykov, Yagoda, and others, and arrests of Central Committee and Politburo members continued.
Huge as it was, the terror of 1938 hit some groups much harder than others. Indeed, unlike the wholesale slaughter of 1937, the victims and processes of 1938 lend themselves to more discrete analysis and breakdown. Unlike the maelstrom of the second half of 1937, the terror of 1938 can be broken down into identifiable targets, beneficiaries, and initiatives. Consistent with the emphasis of our study, we focus first on the party apparatus.
In January 1938 a Central Committee plenum produced a published resolution that criticized mass expulsions from the party based on “false” or excessive “vigilance.” This document has been interpreted in two radically different ways, both of them wrong. On the one hand, it has been seen as an early signal that the terror was to be slowed down or stopped, or at least that some in the leadership were pushing such a relaxation.2Others have seen the document as another bit of Stalinist misdirection, as some kind of attempt by Stalin and his circle to pose cynically as the saviors of people from a terrible and unjust phenomenon.3
Actually, this document had little to do directly with the mass terror that was sweeping the country. It was rather part of the continuing renegotiation of the status of the nomenklatura party secretaries that had been ongoing since at least 1934. Rather than some kind of signal (false or otherwise) about terror in general, it was really part of the multifaceted and multigroup politics that we have seen in the Stalinist 1930s.
The ongoing conflict between the Politburo and the regional nomenklatura had been discursively played out in terms of a constantly shifting attempt by various groups to identify scapegoats and to divert attention from real problems and real culprits: “Who was the enemy?” or “Who was to blame?” Since 1934 Stalin had criticized the regional secretaries as “feudal princes” who had tried to make their territorial machines independent of Moscow. When the enemy had been redefined in August 1936 as has-beens of the Zinoviev and Trotskyist oppositions, the nomenklatura jumped onto this bandwagon—not a peep was raised in defense of their old revolutionary comrades—as long as suspicion did not fall on members of their own regional machines with suspicious pasts. But in the fall of 1936 that identification was made and party machines were combed for disloyal officials, now labeled with the flexible Trotskyist epithet. The situation took yet another course at the February–March plenum, when the focus returned to former oppositionist leaders like Bukharin and Rykov. Although Stalin’s and Zhdanov’s remarks at the plenum served notice that the nomenklatura must be obedient, the focus was elsewhere, and the secretaries quickly backed the new line on enemies.
This dynamic between the Politburo and the nomenklatura was not a simple one. In a larger sense, the tension had to do with the very foundations of the regime. For Stalin to attack the nomenklatura head-on risked discrediting the entire regime: the nomenklatura was the Bolshevik Party, and to smash it—as he did in mid-1937—risked smashing the legitimacy of Bolshevik rule. On the other hand, unconditional Politburo support of the nomenklatura also risked discrediting the regime by endorsing elite pretensions and thereby alienating the rank-and-file party membership and ordinary citizens who were the targets of the secretaries’ control and arbitrary rule. This dilemma helps to explain the Aesopian language of official proclamations, the need to manufacture different Central Committee texts for different audiences, the abstract Kabuki plays with images of traitorous Trotskyists, and the high-level waffling on the fates of Yenukidze, Bukharin, and Yagoda. Each of these contradictory maneuvers, texts, and pronouncements carried strong symbolic content.
This long-standing game came to an abrupt end in the second half of 1937 when Stalin used the power of the NKVD to destroy the regional secretaries both politically and physically. Stalin tried to have it both ways: to destroy the independent-minded officials without casting doubt on the institutions they represented. Stalin’s annihilation of the regional secretaries was publicly characterized as the removal of “Trotskyist-Bukharinist” traitors, not as a calling to account of misbehaving or high-handed officials. Although the administrative “mistakes” of the arrested secretaries were discussed (now as treason), the main public lesson their fall was supposed to teach had to do with traitors and conspirators. Thus the central leadership sought to destroy office holders without weakening the institution of office holding.
This, however, was a difficult job. The mass removal of regional secretaries and their leadership machines could not but weaken the authority of leadership in general. Memoir accounts testify to the breakdown of authority in factories and other institutions in this period, as bosses were afraid to issue any orders that might later be interpreted as sabotage, and as their underlings took advantage of the situation to disobey, threaten, and denounce their chiefs. Factory workers defied managers amid a general breakdown of authority.4
The terror was destroying the party. The documents of the January 1938 plenum show that the Politburo, and Stalin personally, were now concerned about the decomposition and discrediting of the party. The party now had to be “rehabilitated”: mass expulsions had to be stopped; admissions of new members and readmissions of those expelled had to be speeded up. The Moscow leadership realized that it could not govern without a nomenklatura. The party, now cured of disease, had to be given a clean bill of health. The trick was finding a symbolic formula by which to do that.
In the course of the terror and in the wake of the removal of party officials, newly promoted secretaries were being installed. The new nomenklatura now needed protection, reassurance, and authority. The new, younger party officials had to have the authority to govern. At the same time, however, Stalin and his circle wanted to continue weeding out officials they considered disloyal. They wanted to continue the “mass operations” against kulaks, criminals, and others in the general population. It was necessary to find a formulation that would consolidate and restore the party: to stop the terror in the party without weakening it elsewhere. Formulating such a text meant taking a position against certain kinds of “excesses” and not others.
The Central Committee resolution of January 1938 provided such a formulation. It attacked the “false vigilance” of “certain careerist Communists who are striving to … insure themselves against possible charges of inadequate vigilance through the indiscriminate repression of party members.” Such a leader “indiscriminately spreads panic about enemies of the people” and “is willing to expel dozens of members from the party on false grounds just to appear vigilant himself. It is time to understand that Bolshevik vigilance consists essentially in the ability to unmask an enemy regardless of how clever and artful he may be, regardless of how he decks himself out, and not in indiscriminate or ‘on the off-chance’ expulsions, by the tens and thousands, of everyone who comes within reach.”5
Thus the mass depredations in the party were to be blamed (not without some justification) on former party secretaries who for the most part had already been removed. The “serious mistakes” and “false vigilance” of certain party leaders, however, were not to be taken as signs of a slackening of the hunt for enemies. “On the contrary,” the resolution said, vigilance against enemies was not to weaken. (The upcoming show trial of Bukharin and Rykov proved that.) In fact, the NKVD was in no way criticized in the January 1938 resolution. The police—the archetypal agents of vigilance—were given credit for righting various wrongs: “Large numbers of communists have been expelled from the party on the grounds that they are enemies of the people. But the organs of the NKVD found no grounds for arrest.” Indeed, the January resolution called on the party to increase vigilance against enemies of the people.
Although the resolution had identified certain party secretaries as the cause for the crisis in the party, it nevertheless had the effect of stabilizing the party nomenklatura as a group. By locating the problem with “certain” secretaries of the previous period, it implicitly gave approval to the actions of the new regional officials in general. In its final passages, the resolution gave responsibility for rectifying the situation (admitting more members, dramatically speeding up appeals and readmissions, and halting the expulsion of rank-and-file members) to party secretaries themselves. That is, the sins of the past were relegated to certain discredited members of the secretarial nomenklatura, but the group as a whole was not only exonerated but given the job of rebuilding the party, with the authority and credibility to do it. In the months that followed, mass expulsions from the party ceased, large numbers of expelled members were readmitted, and recruitment of new members began for the first time since 1933.6
As usual, the means chosen was the symbolism of a Central Committee plenum and resulting carefully worded texts. Symbolic policy messages were conveyed to the party and public through examples, case studies, or scapegoats on the agenda for discussion; it was government by metaphor. Everyone in the Bolshevik leadership implicitly understood this practice, and party discipline required all to cooperate, even the person singled out as the symbol of the negative. The nomenklatura became especially furious when one of its members refused to play the role assigned to him for the corporate good, Bukharin’s recalcitrance at the February–March 1937 plenum being the case in point.
In early 1938 the needs of the Politburo and the nomenklatura in general also included a scapegoat upon whom the sins of the preceding period could be heaped and whose admission of mistakes and downfall could thus put a final punctuation mark on the preceding period. Postyshev was to play this role.
Pavel Postyshev had long been known as a territorial party secretary who favored mass expulsions of party members. Since 1935 there had been numerous complaints against him from those expelled. At the June 1936 plenum he had been criticized for a “light-minded” attitude toward the rank and file. On those occasions, however, even though he had been called on the carpet, the matter was hushed up from the party generally. In January 1937 Postyshev had been fired from his position in Kiev and transferred to Kuibyshev; even then the Politburo had gone out of its way to shield Postyshev from any serious attacks. Postyshev was tough. He had recently requested the arrest of his own regional NKVD chief for expressing even oblique private doubts about the terror.7 Even though he had his detractors, he enjoyed high-level protection through 1937.
But given the need for a new party discourse at the beginning of 1938, Postyshev became the requisite negative symbol. Sometime around the beginning of the year, the Politburo member A. A. Andreev was assigned the task of gathering compromising material on Postyshev’s party expulsions in Kuibyshev.8 These documents, which became the basis for the January 1938 plenum attack on Postyshev and the resolution of the plenum, included documentation of mass party expulsions from the Kuibyshev soviet, from the ranks of party raikom secretaries, and from other organizations.9 One report from Bazarno-Syzgansky district noted that large numbers had been expelled as enemies by order of Postyshev’s men, although the NKVD subsequently found reason to arrest very few of them.10 Entire district party organizations had been disbanded because everyone had been expelled on Postyshev’s order.
As with other events we have studied, the decision to call a plenum of the Central Committee (and to make an example of Postyshev) showed few signs of long-range planning. Indeed, the documents Andreev compiled were procured only days before the plenum opened, and the Politburo decided only on 7 January to call a plenum for the eleventh. By 9 January, G. M. Malenkov had drafted a resolution for approval at the plenum, as well as a “Secret Letter” to all party organizations explaining the new line. But no Secret Letter was sent. Upon Stalin’s suggestion at the subsequent plenum, it was redrafted into a published resolution.11
In Malenkov’s original draft resolution, which he wrote after discussing the matter with Stalin, he recommended only a censure for Postyshev. Nevertheless, at the last minute Stalin changed the recommendation to include removing Postyshev from his Kuibyshev post and the Politburo and placing him “at the disposal of the Central Committee.”12 Although Postyshev was to be sacked from his Kuibyshev position, he was not expelled from the party, nor was he denounced as an enemy.
At the plenum the traditional forms were followed. Evidence was presented showing how Postyshev had ignored “signals.” A decision was taken in the form of a draft resolution to be formally adopted in a plenum ritual, at which the accused leader was to confirm the charges as “completely correct” and pay his symbolic taxes by confession. Thus lessons would be taught, new signals sent, and unanimity affirmed. As had been the case with Smirnov, Yenukidze, Bukharin, and Rykov, speaker after speaker rose to attack the accused in a self-affirming ceremony of the nomenklatura elite. They closed ranks against one of their own who was now to become a symbol. They also implicitly served as vehicles for corporate construction of elite identity through the prescribed ceremony.
In this case, however, Postyshev seems at first either not to have understood or to have rejected what was required of him. From his point of view, he had not done anything wrong. He evidently forgot that right and wrong, correct and incorrect policy were what the party defined them to be at a given moment. So important was the ritual that Kaganovich and other speakers even prompted Postyshev when he misspoke. Postyshev protested that he was speaking sincerely, but Kaganovich captured the essence of the matter by replying, “Not every act of sincerity is correct.” Only in the course of the meeting itself did Postyshev come to understand that correct and useful performative ritual speech, not constative right and wrong, was required.
A) Session of 14 January (day session).
POSTYSHEV: And now concerning the disbanding of the 30 Party district committees. I must say something, Comrades, concerning my mistake. My situation at the time was also a very grave one. In what sense? The Soviet and Party leaderships were in enemy hands, from the Provincial leadership at the top to the district leadership at the bottom.
MIKOYAN: All of it? From top to bottom?
POSTYSHEV: The entire district leadership. What’s so amazing about it? …
KAGANOVICH: But there were errors. Why do you keep talking only about objective conditions?
POSTYSHEV: I shall talk about my personal mistakes.
KAGANOVICH: You shouldn’t justify yourself by saying that they were all scoundrels.
POSTYSHEV: I never said all of them, I’m not so completely insane as to call everyone an enemy of the people. I never said that, I spoke only of the leadership of many of the district committees….
MOLOTOV: Not a single honest man remained in the leadership?
POSTYSHEV: Viacheslav Mikhailovich [Molotov], I’ll be glad to enumerate them to you [formal vy].
MOLOTOV: I’m [only] asking a question. I have doubts about what you are saying….
POSTYSHEV: What do you want?
YEZHOV: And so it turns out that you [informal ty] only committed a formal mistake. But you know that the CC has characterized your mistake as not a formal one, but as a major political error in substance. So, are you trying to say that the decision of the CC is not correct?
POSTYSHEV: Why should I reduce the whole affair to a formal mistake? Please permit me to finish and explain this whole business to the best of my ability.
KAGANOVICH: You are not very good at explaining it—that’s the whole point.
POSTYSHEV: Whether I explain it well or poorly, I am speaking sincerely, my thoughts are sincere.
KAGANOVICH: Not every act of sincerity is correct. Postyshev: In any case, I’m speaking sincerely. Molotov: And we too are criticizing you sincerely.
KAGANOVICH: You are speaking mistakenly. If you got confused at first, then at least, correct your mistake by the time you finish your speech.
Session of 14 January (evening session).
ANDREEV (CHAIRMAN): Comrade Postyshev, take your seat. This is no place for strolling about….
POSTYSHEV: I ask you to please give me the floor.
ANDREEV: I will put your name down on the list of speakers.
POSTYSHEV: Comrade Stalin, I ask that I be given the floor.
STALIN: Why should we let you speak out of turn?
POSTYSHEV: I ask you just once to make an exception for me.
ANDREEV: We’ll ask the Plenum right now.
POSTYSHEV: Comrades, please permit me to speak right now. Otherwise, I’ll forget everything.
VOICE: Wait your turn.
POSTYSHEV: Please, please let me speak.
ANDREEV: Take your seat! I’ll put it to a vote right now. Who is for letting Comrade Postyshev speak out of turn? One, two. Who is opposed to putting Comrade Postyshev’s name on the list ahead of everybody else? A majority— …
ANDREEV: There is a motion on the floor to give Comrade Postyshev the floor so that he can make a statement, after which discussions will cease and Comrade Malenkov will deliver the concluding speech. Any objections to this motion? None. Comrade Postyshev has the floor and he will make a statement.
POSTYSHEV: I can only say one thing, Comrades, and that is that I recognize the speech which I gave earlier to be fully and totally incorrect and incompatible with the Party spirit. I do not understand myself how I could have made that speech. I ask the CC Plenum to forgive me. Not only have I never associated with enemies, but I have always fought against them. I have always fought on the side of the Party against enemies of the people with all my Bolshevik soul, and I shall fight the enemies of the people with all my Bolshevik soul.
I have made many mistakes. I did not understand them. Perhaps even now I have not fully understood them. I shall say only one thing, that is, that the speech I gave was incorrect and un-Party in spirit and I ask the Plenum of the CC to forgive me for making this speech….
POSTYSHEV: I consider the decision of the Central Committee concerning me to be correct. I simply underestimated the situation. Do you really think I did this deliberately? Malenkov: You did not say this when you were speaking from the podium, when you were given the right to make a statement, and the shorthand record contains only your purely formal statement. Postyshev: I shall correct my speech and shall record in it admission of my mistake.13
At the January 1938 plenum, Stalin had proposed removing Postyshev from the Politburo but leaving him on the Central Committee. But a month later this decision was changed and Postyshev was now charged not only with party malfeasance and “principled mistakes” but with knowing of the enemy’s machinations. His refusal to carry out the apology ritual immediately hurt him in the end.
The documents show that Postyshev had enemies and critics in the party for years. Shkiriatov had assembled a file on him in early 1936. Postyshev had been attacked at the June 1936 plenum and fired from his Kiev job in January 1937. But it is hard to avoid the impression that Stalin was not among Postyshev’s longtime enemies. Shkiriatov’s 1936 complaint file was never pursued. The sharp personal attack on Postyshev at the June 1936 plenum was edited out of the final version of the minutes, and when Postyshev was removed from Kiev, he was given a new job running the Kuibyshev party organization. Stalin condemned Postyshev’s critics in 1937. Even in January 1938 Stalin proposed that Postyshev remain in the party and even on the Central Committee. Yet it seems that Postyshev’s enemies finally won the war when a month later Postyshev was expelled from the party on the basis of charges not mentioned at the previous plenum. (The dates of this and the preceding document are curious. Postyshev was expelled from the party by vote between 17 and 20 February, but the date on the document sending his case to the Control Commission [for expulsion] is 23 February.)14 The final chapter of Postyshev’s career saw the renaming of Postyshevsky District in Donetsk region.15
The sacking of Postyshev was accompanied by a large-scale reshuffling of the NKVD in Ukraine. It is possible to see in this a sorting-out of Ukrainian NKVD chiefs according to their membership in Postyshev’s circle. Some of the replaced officials were probably removed (“put at the disposal of the NKVD”) for their closeness to Postyshev. Others, however, were given equivalent or higher-ranking posts; perhaps they had helped gather evidence against him.16
Even as the overvigilant Postyshev was being sacrificed for the sake of ending mass expulsions in the party, the terror continued unabated on other fronts. The same week that Postyshev was expelled from the party for his excess zeal, the Politburo formally extended the time period for work of the murderous troikas; they were supposed to have finished their “mass operations” by the end of 1937. At the same time, the Politburo raised the execution and exile limits established in the original order. The Politburo considered and approved higher limits for various provinces on a weekly basis, and sometimes more often. These decisions would eventually prolong troika operations until nearly the end of 1938. In Ukraine alone, an additional forty-eight thousand people (“First Category”) were to be shot.17
The mass operations were particularly violent in the Far East, where they were influenced by the regime’s paranoia about sealing the country’s borders.18 The limit for the Far Eastern Territory would be increased again to twenty thousand (fifteen thousand to be shot; five thousand to the camps) in July 1938.19
As these large-scale repressions continued, so did the removals and arrests of high-level officials whom Stalin and the Politburo decided they could not trust. In 1938 there was a second purge of the military high command, as those officers who had sat on Marshal M. N. Tukhachevsky’s court-martial were themselves purged. In the fall, the Far Eastern Red Army was purged. Its commander, Marshal Bliukher, was arrested and beaten to death without confessing.20
At the same time, there are signs in the first half of 1938 that the terror was getting out of the control of the center. In February and March 1938 a series of decrees sought to reestablish centralized direction of the violence. As with the January 1938 plenum, the goal seems to have been to restore order and centralized control of parts of the terror without sending signals that might restrain “vigilance” altogether. Lower-level party secretaries and procurators had to be restrained from excessive purging. Whether from conviction or from a self-defensive desire to display their vigilance, their zeal had exceeded their authority. Texts were produced restricting repression of Red Army officers and restraining zealous local prosecutors. Once again, the emphasis was on limiting uncontrolled repression “from below.”21
On 2 March 1938 the third and last of the Moscow show trials opened. In the dock were Bukharin, Rykov, Yagoda, four former USSR commissars, and several other former officials, sixteen in all. The first show trial (of Zinoviev, Kamenev, and others in 1936) had centered around accusations of political assassination. The second (of Piatakov, Sokolnikov, Radek, and others in January 1937) had incorporated the former Trotskyist opposition and broadened the accusations to include industrial sabotage. This third spectacle tied together the previous sets of charges and associated the former Right Opposition with what was now called a Right-Trotskyist Bloc. According to this final scenario, the Trotskyists and rightists had since 1932 organized a series of underground cells for the purpose of assassinating Soviet leaders, sabotaging the economy, and carrying out espionage at the behest of German, Japan, and Poland. They were accused of conspiring with foreign powers to cede to them parts of the USSR, should they come to power. The inclusion of Yagoda allegedly showed that the secret police had been implicated in the plot, thereby explaining why it had taken so long to uncover the plot.
As with the two previous trials, the event received wide publicity in the press, and a lengthy transcript of the proceedings was published in a large press run.22 The intended lessons of the event are clear: that all oppositionists are traitors, that one must be constantly on guard against all kinds of sabotage, and that foreign enemies were everywhere.
Although one defendant (Krestinsky) initially balked, eventually all the accused pleaded guilty. Given that Radek and Sokolnikov in the previous trial had not received the death sentence, it is possible that the accused in the third trial may have believed that cooperation could save their lives. On the other hand, Molotov later said that they could have entertained no such hopes: “What, do you think they were fools?” he said in reply to a question on the subject.23
This “Bukharin Trial” has been analyzed numerous times in scholarly studies.24 One of the more interesting aspects of it was Bukharin’s testimony: he again contested the ritual. While pleading guilty and admitting to the overall validity of the fantastic charges made against him, he nevertheless refused to confirm specific details of the supposed conspiracy and argued with prosecutor Vyshinsky over numerous details. Although Bukharin’s real purpose will probably never be known, he may have been trying to fulfill his party duty (and perhaps preventing retaliation against his family) by confessing, while at the same time defending his personal honor.25 As he had done at the December 1936 and February 1937 Central Committee plenums, he refused to sully his reputation by admitting to monstrous accusations. In this way, he may have been trying to send a rhetorical “Aesopian message” to the party, the country, and the world: the accusations behind these trials are completely false, and we are being made to confess.
On the other hand, he may have been making a subtle attempt to save his life. It was possible for him to believe that his tactics could have led to a commutation of the inevitable death sentence. Bukharin in the weeks preceding the trial seems to have believed in the possibility in a letter he wrote to Stalin. Had he confessed fully and without reservation, there would have been no grounds to spare him; he would have been a confessed spy. But by suggesting that at least some of the charges were not true, he may have believed that he was leaving the door open for Stalin to spare his life. Whatever his thinking, the decision had already been made, and Bukharin, Rykov, and the others were executed the day after the trial.
Bukharin’s trial demonstrates what Karl Radek had once called the “algebra” of confession.26 According to Stalin’s formula, criticism was the same as opposition; opposition inevitably implied conspiracy; conspiracy meant treason. Algebraically, therefore, the slightest opposition to the regime or failure to report such opposition was tantamount to terrorism. This was the a priori formula behind the show trials, one of whose purposes was to fill in the facts—to assign values to the equation’s variables— with the desired concrete testimony. Although Bukharin refused to provide the details, he admitted to the logic and truth of the algebra.
His attitude received an odd resonance years later in Molotov’s reminiscences. When asked about Bukharin’s guilt in 1973, he said, “I do not admit that Rykov agreed, that Bukharin agreed, even that Trotsky agreed— to give away the Far East, Ukraine, the Caucasus—I do not exclude that some conversations about that took place and then the [NKVD] investigators simplified it.” But just a few pages later, in response to a question about the lack of any concrete evidence except the testimony of the accused, Molotov retorted, “What more proof do you need of their guilt, when we knew that they were guilty, that they were enemies!” When asked, “Then there can be no question that they were guilty?” Molotov replied, “Absolutely.”27 The algebra was compelling.
Our knowledge of events in 1938 is limited. We know that arrests and executions continued, although perhaps not at the hysterical pace of 1937. In 1938, according to NKVD archives, 593,326 people were arrested for “counterrevolutionary crimes,” compared with 779,056 in 1937. In 1938, 205,509 people were sentenced to labor camps, compared with 429,311 in 1937. Although the numbers executed in the “mass operations” in 1938 were roughly comparable to those in 1937 (353,074 in 1937; 328,618 in 1938), many of those shot in 1938 had doubtless been arrested the year before.28
There are signs that by the middle of 1938 the winds were shifting in the high leadership. In April, Yezhov was named commissar of water transport while retaining his leadership of the NKVD and the Party Control Commission. On the face of it, the appointment seems to have been a promotion; he now headed three important agencies: NKVD, the Commissariat of Water Transport, and the Party Control Commission. Moreover, the appointment to Water Transport was not an illogical post for a chief of the secret police. The NKVD (and OGPU before it) had always been heavily involved in purging transport agencies and building canals with forced labor, and Yezhov brought a number of NKVD officials with him to Water Transport.29 Still, it could not have escaped notice that Yezhov’s predecessor Yagoda had been eased out of his police position via the same post.
It is possible that several members of the Politburo (the names A. A. Zhdanov, A. A. Andreev, and K. E. Voroshilov are sometimes mentioned) began to complain that the arrests were weakening the state by promoting too many new and inexperienced leaders into high positions. It seems also that some officials of the NKVD complained to party officials about Yezhov’s administration of the police. These complaints are said to relate to misuse of government funds and Yezhov-authorized executions of some officials without investigation or trial.30
In the summer of 1938 several signals pointed to a decline in Yezhov’s status. In August, G. Liushkov, NKVD chief in the Far East Territory, fled across the Manchurian border and defected to Japan. A close Yezhov intimate and assistant, Liushkov had participated in key police investigations from the Kirov assassination through the purge trials. His defection represented not only a serious security breach but a black mark against his chief. Second, at the end of August, L. P. Beria was brought from Georgia to be Yezhov’s deputy at NKVD. Like Yezhov’s handpicked assistants, Beria was a career police official, but he was an outsider to the central NKVD circles. His appointment gave Stalin his own man inside Yezhov’s administration.31 Third, in the summer of 1938 Yezhov had had a violent disagreement with V. M. Molotov in a cabinet meeting, apparently threatening him with arrest. Stalin had forced Yezhov to apologize.32 However, Yezhov’s fall and the ending of the terror were gradual processes. Even as Yezhov’s personal prestige was falling, executions and arrests continued under his direction. In July a large number of arrested officials, including Yan Rudzutak, were shot. In the summer, the Politburo candidate members Kosior, Chubar, and Eikhe were arrested.
By the fall of 1938, however, the Politburo was changing course. A Politburo resolution of 8 October formed a special commission to study arrest procedures and the apparent lack of judicial supervision over police activities.33 Although Yezhov chaired the commission, it is significant that the other members were from outside his circle: Beria was an outside appointment as Yezhov’s deputy; Rychkov was from the office of the state procurator, and Malenkov was from the Central Committee personnel department. To have a committee looking into arrest procedures was bad enough for Yezhov, but to have it staffed by high-ranking people from other agencies was a real danger to him.
The threat was real. The following month, the Politburo approved and distributed a decree on arrest procedures and judicial supervision. On 15 November the Politburo suspended “until further notice” the work on the murderous NKVD troikas.34 Two days later the Politburo issued a more comprehensive decree, sharply criticizing the work on the NKVD and completely “liquidating” the troikas. The 17 November decree was characteristic of Stalinist shifts in the 1930s. Discursive rules in the party forbade any admission that previous policy had been in error, so one blamed the executors, not the policy makers, and praised the preceding policy while abolishing it. As we have seen, there is clear documentary evidence that the sins now attributed to the NKVD were encouraged, if not ordered, by Stalin himself. The “mass operations,” slipshod procuratorial controls, forced confessions, and the rest were part of high policy that did not originate with the NKVD.
More than cynical scapegoating was at work here, although there was plenty of that in the new discourse. As had been the case in the decrees of May 1933, June 1935, and March 1937 (which the 17 November decree referenced explicitly), the present order went out of its way to applaud repression while apparently seeming to limit it. As with those earlier decrees, the point was to centralize administration in fewer hands. The 1933 and 1935 decrees had not ended arrests; they had simply limited the number of people and agencies authorized to carry them out. The clear meaning of this decree, without saying so openly, was that the NKVD (and Yezhov) were responsible for disorderly repression. Yezhov and the NKVD were not blamed for terror; they were blamed for disorder. Procuratorial sanction before arrest, which had fallen into disuse over the past two years, was reasserted. The new texts thus offered a political transcript to the readers of the decree that enemies were still dangerous, but they were to be destroyed carefully and selectively.35
Two days later the Politburo again discussed the work of the NKVD, based on a report from Ivanovo NKVD chief Zhuravlev.36 Clearly, the Zhuravlev report was instigated by Beria in an attempt to finally discredit Yezhov.37 That report was an attack on Yezhov and several of his lieutenants, and Yezhov was blamed for not “unmasking” them himself. One expert on Yezhov long ago concluded that “Yezhov’s primary crime, however, consisted in the fact that he had not informed Stalin of his actions.”38Just as Yezhov had done to discredit his predecessor Yagoda, Beria now claimed that Yezhov had been hiding investigations from Stalin.39 When Stalin demanded an explanation, Yezhov sent him a list of more than one hundred pending investigations, all of which Yezhov had reported on to him.40 But it was too late for Yezhov. Four days later, after a four-hour meeting with Stalin, Molotov, and Voroshilov, Yezhov sent a letter to Stalin resigning from his post at NKVD.41 His text was formulaic, recognizing and taking the blame for the “mistakes” of the NKVD in espionage and investigatory work. In this and another explanatory letter, Yezhov cited overwork in trying to excuse his sloppy work and excessive drinking. The Politburo accepted his resignation the same day, then two days later named Beria to head the NKVD. The removal of Yezhov’s deputies proceeded quickly. Yezhov last appeared in public on 21 January 1939, atop Lenin’s mausoleum with the rest of the Politburo.42 His name was in good repute at least until April 1939, when Sverdlovsk Obkom “requested” that one of their districts be renamed from Yezhovsk to Molotovsk.43
Stalin removed Yezhov for a variety of reasons. The terror had to be ended, and the only text that could do that without discrediting Bolshevism, the party, or Stalin was one in which there had been “excesses” in the work of its executors. Postyshev became the evil “other” in party affairs; now it was Yezhov’s turn to take the blame for the police. Another reason for Yezhov’s fall was his chronic drinking. We know from various testimonies that during his tenure at NKVD he drank huge quantities every day. He himself mentioned this problem in one of his letters to Stalin and at his own trial. He was drunk at work, at home, at his dacha, and he presided over group drinking bouts with cronies. Stalin and others complained that Yezhov was often absent from his various jobs and was falling behind in his work. Moreover, given the political secrets in Yezhov’s head, it was inconceivable for Stalin to allow him to babble them to all and sundry at drunken parties. As far as we can tell, before and after Yezhov’s own arrest, all of his professional, personal, and family drinking buddies were rounded up and shot.
It may seem odd that Yezhov’s letters to Stalin did not refer to the “excesses” or failures in control over the NKVD that had been spelled out in the 17 November resolution. First, at his trial more than a year later, Yezhov did not admit to any excesses, seeming, rather, genuinely to believe that his only sin was in not purging his own apparatus. Second, to discuss this was to defend himself by pointing out that he was only following Stalin’s orders. This was impossible. Clearly, the new text was to blame those who had carried out the repression, thereby protecting the reputations of those—especially Stalin—who had ordered it. To deviate from that text would mean separating oneself from or “taking up arms” against the party, precisely the crimes Bukharin, Zinoviev, and Trotsky had been accused of. As always, protecting Stalin and the party was the main thing.
The texts of November 1938 not only ordered an end to the wild terror. They signaled a return to politics as usual. In this particular regard, it is less important to ascertain the degree to which procuratorial sanction would be rigorously enforced (although it seems to have been) than to recognize the return of a system of discursive politics aimed at achieving central control over politics and society. November 1938 thus represented not only an end to the terror but a return to the attempt to control events with “legal” hegemony and a systemic/systematic governing narrative.
Yezhov’s removal thus again presented the question of variant “transcripts” for different audiences. The question then, as it always was with the Bolsheviks, was finding the best way to use the decision (and information about it) in the service of the party leadership. The question of truth was always subservient to rhetorical control; more precisely, discourse control became truth. The real reasons for policy changes never had anything to do with the subsequent utility of information that was released to various audiences.
The general public could read only the terse Pravda announcement that Yezhov had resigned for reasons of health and had been replaced by Beria. It was not useful to tell them more: the mass operations had been secret (or at any rate never mentioned publicly), and in the interests of maintaining the facade of party/state unity, it was inexpedient to discuss “mistakes” of the NKVD and failures of the judicial system, or to hint at conflicts in the leadership. Any criticism of excessive vigilance or NKVD mistakes would have cast doubt on the entire vigilance campaign, as well as on party control over the police, and could lead (as it would in 1956) to questions about whether victims had been unjustly condemned.
The broader party and bureaucratic audience (down to the level of district party secretary) was told something different; according to their transcript, the problem was one of “excesses” in the terror. Whatever the real reasons for Yezhov’s fall, it was at that moment useful to the leadership to reassure the new nomenklatura that although things had been out of control, all was now in hand and there would be no more excesses. For example, in December 1938, Stalin piously replied to a letter from Orel First Secretary Boitsov: “I received your letter about false testimony from the six arrested. There are analogous communications from various places and also complaints against former Commissar Yezhov that, as a rule, he ignored such things. These complaints served as one reason for Yezhov’s removal.”44
For this broader elite, it was necessary for the Politburo to portray the problem as a rogue NKVD that had somehow escaped party supervision. In this version of reality, it would have been inconvenient to discuss the specific crimes of Yezhov’s deputies, since to the nomenklatura audience it would have been clear that those deputies had in fact been vetted by the Politburo in the first place.
The more restricted privileged circle of the top leadership, however, read from yet another script, that of Zhuravlev’s report and Yezhov’s resulting letter. Here the reasons for the replacement were different. Yezhov’s letter mentioned only in passing the defects in the work of the NKVD (underestimating intelligence work) and never referred to the problems of excesses, lack of judicial supervision, forced confession, faked interrogation protocols, and the like. For Yezhov, his mistakes consisted only in the fact that his deputies turned out to be enemies. It might seem ironic that this least believable version was the most secret. On the other hand, as we have seen, the information power of those on high consisted largely in knowing the transcripts of those below them, not in a particular version prepared for them. They knew the real story, or at least the possible stories about Yezhov’s removal. They understood the dangers of telling the masses too much and the necessity of reassuring the new nomenklatura about the previous excesses.
In 1973 Molotov was vague about Yezhov’s relations with Stalin and his fall from power. In some places, Molotov seemed to be trying to disassociate Stalin from Yezhov:
MOLOTOV: Yezhov was accused because he began to name quantities [of arrests] by provinces, and in the provinces numbers by district. In some provinces they had to liquidate not less than two thousand, in some district not less than 50 people…. That is what he was shot for. There was no monitoring over it….
QUESTION: Was it the Politburo’s mistake that they trusted the organs [of the NKVD] too much?
MOLOTOV: No. There were deficiencies…. I am not saying that Yezhov was spotless, but he was a good party worker. There should have been more supervision…. There was some, but not enough.45
Of course, Yezhov was not shot for establishing limits by province. We have seen that these limits were in each case approved by Stalin and the Politburo.46 The question of Stalin’s “supervision” is more ambiguous. Above, Molotov suggested that there had not been enough supervision over Yezhov. In other places Molotov gave a different evaluation, and his attempts to shield Stalin from criminal responsibility were hopelessly contradictory. Finally, he admitted that whatever Stalin’s role, in the final analysis the terror had been justified:
QUESTION: If Stalin knew everything, if he did not rely on stupid advice, then it means that he bears direct responsibility for the repression of innocent people.
MOLOTOV: Not quite. It is one thing to put forward an idea, and another thing to carry it out. It was necessary to beat the rights, necessary to beat the Trotskyists, to give the order to punish them decisively. For this [repression of innocent people], Yezhov was shot.47
QUESTION: Did even Stalin have doubts in 1937 that things had gone too far? Molotov: Of course, not only doubts. Yezhov, the chief of security, was shot.
QUESTION: But didn’t Stalin make him a scapegoat in order to blame everything on him?
MOLOTOV: It is an oversimplification. Those who think so don’t understand the situation in the country at that time. Of course, demands came from Stalin, of course things went too far, but I think that everything was permitted thanks to one thing: only to hold on to power!48
Yezhov’s fall meant an end to the mass operations and executions, but not to the terror or its effects. Mass arrests hit the Komsomol at the end of 1938 as that organization’s leadership was purged. In early 1939 several leading officials who had been arrested in 1938 were shot, including Kosior, Chubar, and Postyshev. Moreover, the new NKVD chief Beria purged the NKVD and arrested all of Yezhov’s deputies and department heads. In the short period September–December 1938, 140 NKVD officials from the central apparatus and 192 from the provinces were arrested, including 18 NKVD chiefs of union republics. Thus was the Yezhov patronage group removed.49
The 17 November 1938 resolution about judicial controls and procuratorial supervision over the NKVD represented a victory for USSR Procurator Vyshinsky. For a long time he had favored maintaining procedural norms: procurators had to agree to arrests, specific charges had to be leveled, and some semblance of procedural regularity was to be followed. This approach did not necessarily mean less terror; it meant rather that the bureaucratic t’s were to be crossed and the i’s dotted. After Yezhov’s fall, Vyshinsky became more visible in Politburo documents. While never criticizing the terror administration, he did make suggestions aimed at controlling, regularizing, and even limiting it. Vyshinsky wrote to Stalin,
Recently a great number of cases have been heard by the Special Board attached to the People’s Commissar for Internal Affairs of the USSR. 200 to 300 cases have been reviewed at each session of the Special Board. Under such circumstances it is not to be ruled out that wrong decisions may have been made. For this reason I presented my observations on the matter to Comrade Beria with a proposal to establish procedures for the work of the Special Board which would allow its sessions to be held more often, making it possible for fewer cases to be heard at each session.50
From the end of 1938, numerous NKVD cases were reopened. We have little evidence on the scale of these reconsiderations; rumor places the number of people released in the tens of thousands. Anecdotal evidence suggests that if one had not signed a confession, one’s chances to be freed were increased in the post-Yezhov period. Still, the numbers exonerated were small compared with the numbers repressed, executed, and sent to camps in the preceding two years. Releasing large numbers of the falsely accused would have raised inconvenient questions about the honesty and competence of the party and police, Stalin’s role, and the need for repression in the first place. And many in the leadership really believed that huge numbers were guilty. In response to a question in 1971 about why many innocent people had not been freed, Molotov answered, “But many were correctly arrested. They checked it out, some were freed.”51
Although some were freed and “mistakes” were admitted, at least within the party circles, the fall of Yezhov did not mean a significant relaxation in state repression. In the long run, the numbers of camp victims continued to increase (although with ups and downs) until Stalin’s death.52 In the short run, a series of memorandums in 1939 shows that the mechanism of repression was still in good repair and that the leadership had no intention of relaxing it. The camp regime was not to be modified, releases on probation were prohibited, and legal barriers to rehabilitation were strength-ened.53
After the fall of Yezhov and the turn toward legality, Chairman of the USSR Supreme Court I. T. Goliakov took the lead in the legal rectification of the “mistakes” of the terror. Joined by Procurator General Pankrat’ev and Commissar of Justice Rychkov, Goliakov attempted to streamline the procedure whereby procuratorial protests could result in successful appeals of those wrongly convicted. On 3 December 1939 Goliakov wrote to Stalin and Molotov proposing such an expedited procedure. Stalin referred the matter to Beria, who disagreed. Molotov concurred, and the idea was dropped.54 We conclude with another remarkable passage from Molotov in 1982 about the salutary effects of the terror on the subsequent period:
It is interesting that before the events of the 30s, we lived all the time with oppositionists, with oppositionist groups. After the war, there were no opposition groups, it was such a relief that made it easier to give a correct, better direction, but if a majority of these people had remained alive, I don’t know if we would be standing solidly on our feet. Here Stalin took upon himself chiefly all this difficult business, but we helped properly. Correctly. And without such a person as Stalin, it would have been very difficult. Very. Especially in the period of the war. All around—one against another, what good is that?55