"How do I know what I think 'till I see what I say," Alfred North Whitehead quoted someone as saying. One can imagine H. Stuart Hughes reading his new book--History as Art and Science--with the same wonderment. For in the space of 107 pages Hughes has collected five scholarly but extraordinarily unlikely essays on the changing nature of historical knowledge, where it came from and where it is going. In a book whose subtitle suggests serene contemplation of "twin vistas on the past," one finds assertions just as controversial as those in his last book, An Approach to Peace.

I do not mean "controversial" in the gossipy sense, but rather to convey the taste of shock which some of his suggestions produce. "I hope that in coming years," Hughes writes,

a significant minority of young historians, particularly those concerned with the psychological aspects of historical interpretation, will be going through personal analysis under the guidance of experienced clinicians....Nothing less, I believe, will be adequate to the needs of historical understanding in the second half of the twentieth century.

This appeal for the mass analysis of a segment of the coming generation of historians is contained in the summary of what is probably the crucial essay in the book, "History and Psychoanalyis: The Explanation of Motive." In this essay, Hughes proposes that perhaps "psychoanalysis can help history to cope with its supreme difficulty--the motivation of great historical actors of the past." This is at once the best and the most uneven of the chapters--uneven in the sense that it is brilliantly lucid and brilliantly erratic in close turn.

The problem Hughes faces, and is never quite able to cope with, is that no matter how many parallels exist between history and psychoanalysis, the dynamics of the psychoanalytic revelation and of the historical perception will never be the same. In the first case an individual, all of whose thoughts are in some way directly relevant, is actively engaged in trying to understand a self-contained unit, himself. In the latter case, a historian faces great masses of evidence, much of which is only indirectly relevant, in an effort to understand an historical individual or, worse yet, an epoch, which has no desire to tell him anything about why things happened.

And yet Hughes' case is impressive. Both history and analysis, he says,

seek out the reasons for which individuals and groups did what they did, and in each case the method of the search is itself part of the process of understanding. Both strive for a precise, detailed reconstruction of the circumstances surrounding an action: both operate on the assumption that the patience of the investigator will bring its appropriate reward...both distrust the ready explanation that springs first to mind. The connection between the two seems obvious, but it has only recently been explicitly recognized.

The problem is, as suggested above, that the ambiguities involved in these parallels seriously limit them, and could easily lead even a careful scholar toward a pseudo-analytic approach to history. Hughes recounts, for example, how one of his students, studying the life of an armaments expert with pacifist tendencies, could not understand why the expert developed a crippling block against completing the military research he was involved in. "The solution lay right before him," Hughes says. "For the student had quite unwittingly run up against a classic case of inner conflict. His protagonist's technical and military pride were locked in hidden combat with his leftist and pacifist leanings..."

Such psychological interpretation, one must suggest, is about as subtle as a tank. What is this if not "the ready explanation that springs first to mind," against which Hughes was militating a few paragraphs earlier? It may be right; it also may well be wrong.

But the dangers of trying what Hughes is attempting are necessarily great, and occasional loose pieces left over from puzzling together the relationship of psychoanalysis and history should not obscure the essay's value. Hughes is doing what he has done before--overstating so that some of the radical truths he sees will be conveyed to his readers. When he sees what he has said, I think he will find he has said it very well.