The arguments against this notion are very powerful. Krestinsky’s phrasing during the first day sounds, on the face of it, genuine, and some of the points he makes seem to be both valid and extremely embarrassing to the prosecution. Vyshinsky’s attitude is sinister in a way which appears more compatible with a genuine threat to Krestinsky than the appeal to his reason and conscience, which would be supposed to have produced the change in his evidence on the second day.
There is another and most suggestive piece of evidence from the trial itself. At the beginning, Vyshinsky announced the order in which he proposed to question the twenty-one prisoners. This was evidently a predetermined list, since it is repeated in the order in which they made their final pleas, except for one transposition among the minor characters. But the actual order of questioning was different. The first day went according to schedule—Bessonov, Grinko, Chernov. But on the second, Krestinsky should have had his examination-in-chief immediately after Ivanov. Instead, as we saw, the agricultural official Zubarev took the stand on the morning of 3 March, and in the afternoon Krestinsky was called upon not for a full examination, but only for a brief recantation. This, too, was preceded by the short and unscheduled interrogation of Rakovsky, undermining Krestinsky’s point about the letter disavowing Trotskyism.
On the evening of 3 March, this was all that Krestinsky’s examination amounted to. His full examination was postponed until the following afternoon, when the examinations of Rosengolts and Rakovsky were rescheduled, the former to precede and the latter to follow Krestinsky’s. Rosengolts, as Krestinsky’s alleged closest Trotskyite collaborator, established both Krestinsky’s connections with Trotsky and the joint plotting activity of himself, Krestinsky, and the Tukhachevsky group in the days after the Pyatakov Trial and Bukharin’s arrest. Krestinsky confirmed and elaborated all this, and Rakovsky rubbed it in afterward. All this seems to show an emergency procedure.
Nor is it difficult to see that the circumstances of his original interrogation were such as to make retraction possible. Krestinsky had been arrested at the end of May 1937. He confessed “after the lapse of a week … at the end of the first interrogation.”51 This was by torture in its most intensive form, and he was not, therefore, broken in the long-drawn-out fashion described on here. It was just at that time that Bukharin was making his first confession, and it may be that it was momentarily intended to produce the next big trial at very short notice—and about as soon after the Pyatakov Trial as that had been after Zinoviev’ s .
If so, this perhaps came to nothing when Bukharin started to retract some of his confession (see here), and the whole business of interrogating him had to begin over again. Meanwhile, with the spread of the Purge right through the Party, other useful additions to the case kept emerging, and it was nine months later that the trial took place.
However that may be, it is certainly the case that the NKVD seems to have had in its hands a leading prisoner who had not been brought into the right state for a trial by the most tried and successful procedure. Torture and the conveyor could produce confessions, but, as we have seen, the victim, once rested, recovered to the degree that he could retract them. He was not reduced to the degree of submission obtained by the longer method. Yet Krestinsky had confessed, and was being cooperative. There was no overt resistance to break.
Krestinsky’s withdrawal of his confession was not unprecedented. During the Shakhty Trial, one of the accused (Skorutto) had refused to confess, been kept out of the court for a day on the grounds of illness, and then come back and confessed, only to withdraw the confession again, and the following day to reaffirm it. Again, in the Metro-Vic Trial, MacDonald withdrew, then reaffirmed, his original confession. It has never been thought that these withdrawals added anything to the credibility of the confession when finally produced.
Thus the argument that Stalin had planned the whole Krestinsky episode is a weak one. What seems more probable is that some such story was put round
In any case, Krestinsky’s withdrawal, so dramatic and so cogent in its cornments, barely affected the reception given to the trial by the foreign public. Stalin had won again.
