“
It turns out that in that terrible year Andrei Yanuaryevich (one longs to blurt out, “Jaguaryevich”) Vyshinsky, availing himself of the most flexible dialectics (of a sort nowadays not permitted either Soviet citizens or electronic calculators, since to them yes is yes and no is no), pointed out in a report which became famous in certain circles that it is never possible for mortal men to establish absolute truth, but relative truth only. He then proceeded to a further step, which jurists of the last two thousand years had not been willing to take: that the truth established by interrogation and trial could not be absolute, but only, so to speak, relative. Therefore, when we sign a sentence ordering someone to be shot we can never be absolutely certain, but only approximately, in view of certain hypotheses, and in a certain sense, that we are punishing a guilty person. Thence arose the most practical conclusion: that it was useless to seek absolute evidence—for evidence is always relative—or unchallengeable witnesses—for they can say different things at different times. The proofs of guilt were relative, approximate, and the interrogator could find them, even when there was no evidence and no witness, without leaving his office, “basing his conclusions not only on his own intellect but also on his Party sensitivity, his moral forces” (in other words, the superiority of someone who has slept well, has been well fed, and has not been beaten up) “and on his character” (i.e., his willingness to apply cruelty!). In only one respect did Vyshinsky fail to be consistent and retreat from dialectical logic: for some reason, the executioner’s bullet which he allowed was not relative but absolute. . . . Thus it was that the conclusions of advanced Soviet jurisprudence, proceeding in a spiral, returned to barbaric or medieval standards. Like medieval torturers, our interrogators, prosecutors, and judges agreed to accept the confession of the accused as the chief proof of guilt.
”
”