The Public worries that criminal trials, especially those involving murderers, have been hamstrung by the introduction of a number of implausible new legal stratagems. These range from the so-called Twinkie defense—the assertion that excessive ingestion of junk food diminishes judgment—to the claim advanced by lawyers for Eric and Lyle Menendez (with the effect of producing a hung jury in their first trial) that fear of sexual abuse is sufficient to reduce the charge for killing one’s parents, to arguments that a woman may shoot a brutal husband even when he is asleep.

Americans have never been entirely comfortable with the not-guilty-by-reason-of-insanity verdict pronounced upon John Hinckley after he shot President Ronald Reagan in 1981; that discomfort has been heightened by what people view as an indefensible effort to extend the narrowly defined legal category of insanity to include various psychological states like temporary insanity, diminished capacity, and a multitude of “syndromes” including premenstrual, postpartum, post-traumatic, and the like. The emergence of these exculpatory psychological concepts suggests to many that essential notions of personal responsibility have been eroded and the courts’ stern task of judging behavior has been weakened by the deployment of dubious theories of social causation. Instead of promoting a dispassionate review of objective evidence, our legal system increasingly encourages us to explain criminal conduct on the basis of conflicting theories presented by rival expert witnesses speaking psychobabble.¹

Many legal scholars, however, see the matter quite differently. The law reviews are filled with learned and subtle discussions of every novel defense claim considered by the appellate courts. The law professors who read and write these essays are usually inclined to dismiss public anxiety over such legal innovations as unwarranted. The insanity defense, they point out, is rarely raised and is even more rarely successful. Moreover, a person found to be insane may spend more time in an institution than one convicted of murder. The Twinkie defense, though it made for provocative headlines at the time, probably played no role in the conviction of Dan White on manslaughter charges after he shot to death both the mayor and a city supervisor of San Francisco in 1978.

Indeed (the law professors continue), few of the new defenses based upon the mental condition of the defendant have allowed killers to walk away free; at most they have produced verdicts of manslaughter rather than murder, and even this occurs only infrequently. The first trial of the Menendez brothers was an anomaly, not at all representative of what happens in the great majority of homicide prosecutions. To be sure, the battered-woman syndrome has been introduced into many trials of women who killed their husbands, but such killings are also unusual; the syndrome rarely leads to any outcome more questionable than a lenient sentence based on a recognition of grave prior brutality.

Clearly, there is a gap between what the public and what the law professors believe. Although there is merit on both sides, I think the public one should dominate. But, paradoxically, the public itself is of two minds.

Studies of public opinion show that the average citizen wants the system to be tough on criminals by imposing harsher sentences. Studies of juries, by contrast, suggest that the average juror aims to do justice by understanding the defendant and by devising a verdict that will take into account his subjective state and the circumstances of his offense. Thus, two-thirds of adults questioned in Illinois expressed the belief that judges were too lenient in sentencing convicted burglars; but when asked to design an appropriate sentence, most of these same people recommended penalties that were even less severe than the minimum sentence a judge could have imposed. In their classic study, The American Jury (1966), Harry Kalven and Hans Zeisel compared the decisions of juries with the views of judges familiar with the facts of each case. Though the two usually agreed, in those instances where they disagreed the jury was much more likely than the judge to have favored acquittal.

_____________

Norman J. Finkel, a professor of psychology at Georgetown University, begins his new study, Commonsense Justice: Jurors’ Notions of the Law² by reviewing these and other examples of the tension between what people say in their capacity as citizens and what they do in their capacity as jurors. His data come from experiments in which people have been asked to pretend to be jurors and to react to the scenario of a criminal case.

That there are obviously limitations to this method is something Finkel carefully acknowledges. Make-believe juries are a bit unrealistic: while some “jurors” in these studies hear evidence read aloud by others, meet to discuss cases, and try hard to duplicate what a real jury goes through, others merely stare at a written transcript. Still, whether or not such studies produce accurate results, their central finding is a compelling one: jurors are interested in explanations. They seek, in Finkel’s words, to “account for the facts by inferring causal and intentional links among particular facts.”

In one experiment, for example, Finkel asked mock jurors to sentence a man who had been arrested for writing a $100 check against a nonexistent bank account. To the stock of information the jurors knew about the defendant, Finkel then added successive facts, telling them ultimately that the man had previously been convicted of 7 other crimes. As the jury was exposed to such incriminating information, the average sentence it imposed rose from 10 months to 30 years. But then, in the last stages of the experiment, the jury heard either a psychiatrist speaking on behalf of the defendant—which made the sentence fall from 30 to 5 years—or both a psychiatrist testifying for the defendant and a prosecutor testifying on the dangers of recidivism—which brought the sentence down to a level of about 15 years. In either case, only a small percentage of the jurors sentenced the defendant to life imprisonment without parole, the requirement of the state from which the case had been drawn.

In a similar study conducted by Finkel, mock jurors were asked to pass judgment on a woman charged with killing her abusive husband, but the facts were altered as they were presented to different groups of volunteers. To some, the husband was described as having advanced on his wife with a weapon. Others were told he had gone to sleep after threatening that as soon as he awoke she would get “what she deserved.” Not surprisingly, four out of every five mock jurors voted to acquit when the woman shot her husband as he advanced on her menacingly. But almost half also voted to acquit even when she killed him while he was asleep.

_____________

These two cases, and others like them in the book, rest comfortably with what I take Finkel’s own views to be. He is a critic of mandatory life imprisonment for repeat offenders, and so he welcomes the fact that jurors worry about the degree of guilt of a repeat felon in a state with mandatory life-sentence laws. He is concerned about the problem of battered women, and so he embraces the fact that half of the mock jurors voted to acquit a woman who shot her abusive husband while he was asleep. For Finkel, what the defendant intended is central, and he stresses the great importance of a jury’s desire to achieve a proportionality that will reconcile a defendant’s actions with his motives and circumstances.

Finkel’s argument runs against the famous and, to my mind, still valid lessons of Oliver Wendell Holmes, who tried to show, in The Common Law (1881) and elsewhere, that subjectivism—that is, seeing the world through the eyes of the defendant—is wrong. A jury has a natural inclination to judge the person as well as the act; it is out of a desire to check this inclination that the written law strives to draw a bright line between actions and motives. Strictly speaking, the motive for a crime is irrelevant. It is as wrong to steal in order to buy cocaine as it is to steal to give to the Salvation Army. A man who kills his wife is guilty of murder whether he does so to obtain her insurance or to end her painful disease.

In practice, of course, motives are rarely irrelevant. Prosecutors refer to motives in deciding what charges to bring, and juries refer to them in deciding what penalty to impose. Thus, written law and human sentiments are partially at war one with another. But the war ought not end with the surrender of one side. The law draws a clear line; that same line reflects, or ought to reflect, the moral sentiments that people bring to judging crime. As Finkel rightly points out, a pure commitment to objectivity—which would lead to the same punishment being meted out to people who killed from very different motives—risks stripping the law of its moorings in widely shared standards of justice. But forsaking objective law in favor of the shared moral dispositions of people acting as jurors risks abandoning the neutral principles of the law, thus punishing—or, alternatively, excusing—people for their thoughts, motives, and character, and weakening the social-control features of the law by telling would-be offenders that they can get away with crime if they come up with good stories.

_____________

To grasp what is at stake here, consider the possibility that Finkel had undertaken his research in, say, Mississippi in the 1920’s. There, I suspect, studies would have shown white jurors treating whites and blacks very differently, excusing the former if they killed the latter but severely punishing the latter if they killed the former. Under such circumstances, Finkel’s commitment to proportionality between intention and punishment might well have taken a different turn, leading him to wonder whether such disparate outcomes based upon race should be tolerated. Would not a book written by him then have given greater weight to the principles of neutral law and objective standards of culpability than to moral sentiments?

In short, we need to keep in mind these two broad functions of the law: assessing individual actions and affirming shared social understandings. I agree with Finkel that the average citizen contemplating a new law and the average juror judging a particular application of that law will often reach very different conclusions. But tax burdens are also assessed very differently by the citizen as voter and the same citizen as taxpayer, and so is Medicare by the citizen as observer and the same citizen as participant. We want the public to be of two minds about certain things, and not the least of these is justice.

This is certainly true with respect to the insanity defense. Consider the M’Naghten rule, which states that a person is legally insane if by a “defect of reason” he is unable to “know the nature and quality of his act” or, if he does know these things, he does not know that what he did “was wrong.” I happen to like this formulation for all the reasons that the attorney Joseph Livermore and the psychologist Paul Meehl offered nearly 30 years ago. It is a very good rule because it recognizes insanity as a defense, links it to the character of practical reasoning, and clearly implies (and this is how I think most juries interpret it) that you are sane unless you are gripped by hallucinations or delusions.

Finkel, however, distrusts the M’Naghten rule because it admits of no shade of intermediate status: it asks jurors to imagine a person who is either entirely equipped with right reason or entirely deficient in it. At one level he is right to suggest that such a sharp distinction makes no sense: people vary in the strength of their moral faculties, with some being weaker and others hardier. But at another level it makes perfectly good sense: one either hallucinates and suffers from delusions or one does not. If we want the criminal-justice system to reflect all the gradations developed by social science, we could ask it to array defendants along a continuum ranging from obviously insane to clearly sane. But that is not what we want or need from the system; we want it to defend justice by holding everybody to a high standard of personal conduct, while allowing a jury to find someone not guilty by giving it the power to apply a narrow test of insanity.

Finkel proposes a different test, one more closely suited to how jurors like to think. He wants them to ask, in order, the following questions: did the defendant’s actions cause harm to befall the victim? If yes, then was the defendant suffering from a “disability of the mind”? If yes, then what was the degree of mental disability: partial or total? And if partial, was the defendant himself to blame (for example, by abusing alcohol or drugs) for bringing it about?

Although Finkel may well be correct that this is how jurors want to think, he does not make a convincing argument that this is how we should let them think. My own view is that considerations of partial mental disability are an opening for jurors to place themselves so fully in the defendant’s shoes that they will forgive defendants they like and punish those they dislike.

_____________

Being a juror is a difficult task, but making it easier can be a mistake. The explanation-hungry, story-seeking, context-searching temperament of a juror confronting a case must be held sharply in bounds by the skeptical, judgment-prone, verdict-seeking attitude of a nonjuror, an ordinary citizen, facing the same case.

American courts have of late been somewhat quick to embrace social science at the expense of law. It is a tendency that ought to be resisted. Finkel’s book is a useful volume for a social scientist. I hope lawyers do not read it.

¹ See, for example, “Getting Away With Murder” by Walter Berns, COMMENTARY, April 1994.

² Harvard University Press, 400 pp., $39.95.