Implicit editorial policies and the integrity of psychology as an empirical science.

In practice, however, the question often becomes, as it does with Greenwald, What selected phenomena can support my theory? The touchstone here is no longer observable reality. Theory is no longer just the best tool for achieving convergence between phenomena and concepts, but is an end in itself; experiments are no longer just the best method for testing convergence, but become an effort to prove oneself right. Just what Greenwald says, one might think. But not quite. For being able to say it and to see its limitations makes it possible at least to consider escapes from the trap; there is, after all, a rational ego, too, even if it is less powerful than the totalitarian one, even if its emergence is prompted by irrational factors. One can look, for example, at Greenwald's impressive experimental evidence with a different eye and, reverting to the basic question of psychology, note that every single experiment contains some recalcitrant persons who refute the experimental hypothesis, Results are never unanimous; some 60% will often do for statistical significance, while 40% are excluded from the realm of psychological explanation as "random errors." But the latter, too, survive genetically and deserve explanation. The great advances in statistical techniques often disguise the not-hypothesis-confirming minority to the naked eye. It hardly needs saying that such techniques are invaluable, just as theory and experiments are—in their place. It would be quite easy to precede a sophisticated analysis of covariance or factor analysis by simple descriptive statistics that recognize persons before dissolving them into variables, thus drawing attention to those who remain unexplained and deserve further study. But this has become almost infra dig. George Orwell and Hannah Arendt, whom Greenwald quotes apparently so appropriately, were persons whose lives were devoted to unmasking self-aggrandizement, falsification of history, and conservatism. I did not know them well enough to judge the relative sizes of their totalitarian and their rational egos, but I am convinced that they had both and that their achievements were the result of experiencing conflict, not of denying it. If Greenwald's powerful exposition were to result in the production of many more experimental demonstrations of the genetic necessity for the totalitarian ego that disregard conflicting evidence, psychology, in its modest way, could contribute to submerging the rational ego. But his funny and disarming acknowledgments lead me to believe that he knows he has got hold of only one leg o f , the elephant, not of the whole beast.