Knowledge for What?

Human Behavior: An Inventory of Scientific Findings

by Bernard Berelson and Gary A. Steiner
Harcourt, Brace & World, 712 pp., $11

The behavioral sciences are often met with disinterest, contempt, and even hostility. To sneer at sociology and psychology, to attack at least certain methods or findings of these enterprises, is thought by many people to show hard-headedness about human affairs or a deep sensitivity to the human spirit. The radical right perpetuates a confusion of “social science” with “socialism,” and from the left come warnings that we are only dehumanized by such research, and made increasingly subject to manipulation. The serious student of human behavior is caught in the middle.

The fact is, of course, that behavioral scientists do not have either enough knowledge or enough power to do much of anything to our way of life. In any case, what someone proposes to do with his knowledge does not settle what it is that he really knows. A scientific finding is one thing, and its application is quite another. The first question is what the scientific findings are, and whether they are indeed scientific.

That the behavioral sciences are not really science is today widely agreed to, though most commonly, I think, not by scientists themselves but by those who are alienated from the spirit and method of science, whatever its subject-matter. If science is viewed, not as a body of precise and indubitable propositions, but as a systematic attempt to learn more than we know and to become more sure of what we think we know already, then the behavioral sciences surely deserve to be taken seriously. No reader of this book can pretend that it teaches him nothing worth knowing about human behavior, or deny that at least it gives him a better basis for the convictions that in his wisdom he has arrived at himself. The authors do themselves an injustice in inviting their readers to a “willing suspension of disbelief”—Coleridge’s formula for the enjoyment of fiction. Whatever its shortcomings, behavioral science even now is more truth than poetry.

Bernard Berelson and Gary Steiner report over a thousand numbered findings about human behavior, what they regard as “important statements of proper generality for which there is some good amount of scientific evidence.” These findings are not, of course, “all there is” to the behavioral sciences; they have been selected as representative of methods and results. The findings are accompanied by the actual data, which are fairly and carefully reported, and by full citations to the original studies. Such an inventory does not make for good reading, but it is fascinating to dip into, and rewarding as reference.

It could be argued whether the selection really is representative. On the one hand, the most developed fields, like linguistics and economics, are omitted as being too specialized and technical. On the other hand, fields like psychoanalysis are largely ignored as insufficiently “behavioral” and “scientific.” We are told, for instance, that dreaming is important (“subjects kept from dreaming by being awakened whenever eye movements signaled the onset of a dream became irritable and anxious”), but no findings are reported…

This article is available to online subscribers only.
Please choose from one of the options below to access this article:

Print Premium Subscription — $99.95

Purchase a print premium subscription (20 issues per year) and also receive online access to all content on

Online Subscription — $69.00

Purchase an Online Edition subscription and receive full access to all articles published by the Review since 1963.

One-Week Access — $4.99

Purchase a trial Online Edition subscription and receive unlimited access for one week to all the content on

If you already have one of these subscriptions, please be sure you are logged in to your account. If you subscribe to the print edition, you may also need to link your web site account to your print subscription. Click here to link your account services.