CRIME AND HUMAN NATURE. By James Q. Wilson and Richard J. Herrnstein. Simon and Schuster. 639 pp. $22.95.

IN A RECENT "debate" on a local New York news show, two commentators were discussing the merits of a massive new study of crime by Harvard University social scientists James Q. Wilson and Richard J. Herrnstein. One commentator, who called himself a conservative, took the authors to task for allegedly failing to realize that "criminals are born, not made."

The other, who called himself a liberal, praised the authors for their sensitivity to the fact that individual, possibly inherited, personality traits may play as important a role as environment does in the development of criminal behavior.

These conflicting assessments illustrate both the main strength and weakness of Wilson's and Herrnstein's magnum opus on the causes of crime. The book is so thorough and so meticulous in its acknowledgement of contradictory evidence that it is possible for two people to read totally different messages in its pages.

This scholarly fair-mindedness is certainly preferable to the political ax-grinding that usually characterizes dissertations on crime, but it makes for soporific reading. The authors' point of view is buried amid so many tedious studies -- many of which could just as easily be summed up in aphorisms like "the apple doesn't fall far from the tree" -- that it is difficult to discern their real opinions on the causes and prevention of crime.

The authors do, of course, have a point of view -- albeit one that is carefully qualified and obscured by social science jargon. They believe their colleagues in the social sciences (and a great many politicians and criminologists) have underestimated the importance of "constitutional factors" in criminal behavior.

These constitutional factors, the authors argue, may be the result of either heredity or environment -- but they appear at or soon after birth. Low intelligence, for instance, may be the product of genetic inheritance, poor prenatal nutrition, a brain-damaging accident at birth -- or all three factors.

The authors do not dwell on the issue that preoccupies right-wing researchers like Arthur Jensen -- the belief that heredity, not environment, is the main determinant of intelligence. Instead, they simply point out that extremely low intelligence -- and a subsequent inability to succeed in conventional terms -- is one of the factors contributing to violent criminal behavior.

"It is a mark of decent and compassionate instinct," they argue, "for people to fear any explanation of crime that could be interpreted as suggesting that some individuals are destined, beyond hope, to experience those penalties (occasioned by law-breaking)."

"We offer no such theory of predestination . . . Nevertheless, the evidence . . .leaves little doubt, we think, that constitutional factors are implicated, to an unknown but not trivial degree, in the prevalence of high-rate offending. The problem is to acquire a better understanding of how those factors interact with familial and other social experiences."

Wilson and Herrnstein devote considerable attention to what they regard as the most important "constitutional factor" of all -- the gender of violent criminals. In every society -- regardless of how high or how low the crime rate is -- the overwhelming majority of violent offenders are young men.

The fact that most violent crimes are committed by men is, of course, known to every criminologist, police officer and ordinary citizen who flinches at the appearance of a well- muscled young man on a dark street at night.

THE CORRELATION between crime and gender has, however, been underemphasized by nearly everyone except feminist scholars. One night, after backing out of a subway station because the only people on the platform were two young men, I found myself shaken by the realization that I wouldn't have to be afraid if only women rode the trains. On both a personal and social level, this is an unnerving thought: I suspect that most students of criminal justice prefer to keep it way down in the unconscious.

As Wilson and Herrnstein point out, there is a much stronger association between crime and gender than between crime and race, class, economic status or any other factors that receive so much public attention. The studies they quote show there is little evidence to support the idea that violent crime among women has risen as a result of recent advances in women's rights. Interestingly, international crime statistics also show little relationship between the status of women and female crime. Japan and Korea, for instance -- countries where women are still restricted to highly traditional roles -- have virtually the same female arrest rates as Finland and The Netherlands, where women's rights are among the most advanced in the world.

In this area, as in so many others, it is virtually impossible to separate "constitutional" factors -- such as the greater physical strength of men -- from "social" factors, such as the encouragement of aggressive behavior in boys.

However, common sense suggests that even if men are more aggressive by nature than women, their aggression is promoted by strong, cultural reinforcement. While a democratic society ought not to tamper (even if such tampering were effective) with genuinely innate traits, it can and ought to change messages about what is socially acceptable male behavior. Indeed, recent laws making it easier to prosecute rape cases are one of the changes that might conceivably modify this particular manifestation of criminal male activity.

To some extent, Wilson and Herrnstein give short shrift to the question of punishment -- not because they regard it as unimportant but because their focus is on causes rather than cures for crime.

As might be expected, they do not denigrate the role of punishment as a tool for altering certain kinds of criminal behavior. While tbe true psychopath will not be deterred by the likelihood of a prison term, a more effective criminal justice system would certainly alter the calculations of those who rob people on the streets because they believe (correctly) that there is little chance of their being caught or sent to jail.

Nor do Wilson and Herrnstein regard deterrence as the only justification for punishment. They believe that retribution -- not because it will deter future crimes but because it is just -- is a legitimate and important function of punishment.

"An offender has violated an implicit social contract that ties the members of a community together," they note. "The criminal is therefore said to 'owe a debt' to society. Punishment as retribution balances the books."

Crime and Human Nature is a book that will be useful to anyone who wishes to have access, in one volume, to most of the important criminological studies of the past 50 years. Its main virtues are fairness and common sense. Its main fault is the fact that most of its conclusions will already have occurred to anyone who possesses a modicum of common sense.