Miranda's hidden costs.

The Warren Court's creation of a constitutional right not to be questioned has impeded law-enforcement efforts in ways that even the professionals scarcely comprehend.

CURIOUSLY absent from the debate within Congress about how to combat historically unprecedented levels of violent crime in the United States has been any serious discussion of the single most damaging legacy of the Supreme Court's criminal-justice revolution of the 1960s: the Miranda rule. While intensive debate has focused upon other aspects of the Court's revolution -- for example, the exclusionary rule and novel forms of habeas-corpus review --Miranda today seems little more than an anachronistic remnant of the era of "Impeach Earl Warren" billboards.

One possible explanation for this development is that the impact of Miranda, while extraordinarily detrimental to the criminal-justice system, is largely a hidden one, while the costs of such innovations as the exclusionary rule and habeas corpus are highly visible, in the form of kilos of cocaine being removed from the courtroom and repetitive criminal appeals inundating the justice system. Indeed, the costs of Miranda are so obscured that even many law-enforcement officers are only vaguely aware of them. Estimating the costs of Miranda requires that we engage in the difficult exercise of comparing the present reality with the reality which might exist "but for" the rule.

For tens of millions of Americans, the Miranda warnings have been learned from countless televised police dramas in which arrested suspects have been apprised by conscientious police officers that "You have the right to remain silent; what you say may be used against you; you have a right to an attorney; and you have a right to a free attorney if you cannot afford one." Few of those who have long since memorized these innocuous words have paused to consider how they have eroded the ability of the criminal-justice system to carry out its responsibilities.

In its 1966 decision in Miranda v. Arizona, the Supreme Court by a 5 to 4 vote determined that the Fifth Amendment's prohibition against a person's being "compelled in any criminal case to be a witness against himself" required what have become known as the Miranda warnings whenever a witness in custody is subject to questioning by the police. The decision also required that police obtain a "waiver of rights" from a suspect -- that is, an affirmative agreement from the suspect that he would like to talk to police. Also, if at any time the suspect indicated a wish to stop talking or to see a lawyer, police had to stop questioning immediately. Failure to deliver the warnings, obtain the waiver, or cut off questioning when requested automatically bars the use at trial of statements obtained from the suspect.

Underlying Miranda was a deep suspicion on the part of the Court majority about any custodial interrogation of criminal suspects. For the 175 years preceding Miranda, it had never been thought that the police were violating a suspect's constitutional rights merely by questioning him in the absence of an attorney. No, we are not referring to a police officer's beating the hapless suspect with a rubber hose or policemen working in shifts to keep the suspect from sleep until he finally confesses to a crime. Such tactics were condemned uniformly well before Miranda. To satisfy the requirements of the Fifth Amendment, a confession needed to be voluntary, and circumstances which called that voluntariness into doubt served to undermine the admissibility of a confession. Confessions occur for any number of reasons, including the simple desire to cleanse one's soul and the more complicated desire to explain extenuating circumstances. Under the voluntariness standard, police were not precluded from asking the suspect what knowledge he possessed about the dead body they had discovered buried in his backyard.

This old understanding of the Fifth Amendment's prohibition against coerced self-incrimination was sharply transformed by Miranda. The Court acknowledged that "it might not find the defendant's statements to have been involuntary in traditional terms." In other words, the Miranda decision was amending the Constitution. In place of the previous understanding, the Court effectively provided a criminal suspect with a right not to be questioned. If he was questioned prior to the warnings, any statements would be suppressed; if he was questioned after the warnings and after he had requested an attorney, any statements that were made prior to the attorney's arrival again would be suppressed. After the attorney's arrival, it was certain that there would be no statements at all.

In dissent, Justice Byron White observed, "In some unknown number of cases the Court's rule will return a killer, a rapist, or other criminal to the streets and to the environment which produced him to repeat his crime whenever it pleases him. As a consequence there will not be a gain, but a loss, in human dignity." Justice White was prescient, although it can hardly be imagined that the majority of Justices in Miranda disagreed with his assessment. Such a result was logically certain under Miranda.

The insidiousness of Miranda is that, by and large, the violent predators placed back on the streets are not suspects to whom police have failed to give proper warnings. After nearly three decades, the police not surprisingly have learned Miranda by rote and only rarely fail to administer its warnings or follow its waiver and questioning-cutoff rules. Relatively few criminal cases involve the suppression of evidence obtained by police after a failure to comply with Miranda. This is what many observers, including some law-enforcement officers themselves, mean when they contend that the system has "learned to live" with Miranda.

No, the enormous cost of Miranda is entailed not in the lapses of the system but in its successes. It is when the warnings are properly administered and waiver rules are followed that it wreaks its greatest harm. For what it has done is to substantially dry up access to a hugely important category of criminal evidence -- confession evidence. It is almost as if the Supreme Court had told law-enforcement officials that, henceforth, they were no longer going to be able to use fingerprint evidence. No one would doubt that in "some unknown number of cases" individuals who would otherwise have been correctly identified as criminals would avoid prosecution.

Principally, Miranda has eroded the supply of information available to law enforcement by introducing the criminal's defense attorney to the process at the earliest possible stage. This is done, in a sense, by the suspect asserting his Miranda right to have an attorney present at questioning even before formal charges have been filed. The effect of this is to preclude entirely the questioning of many suspects because police recognize that such questioning would be pointless. By effectively insulating the suspect who invokes his rights or asks for a lawyer from any questioning, no matter how restrained or reasonable, Miranda has assured that far fewer confessions will be induced by questioning. For the Miranda majority, this was cause for celebration, not concern. But this is an odd -- not to say dangerous -- view of the world. As an earlier Supreme Court said, "the Constitution is not at all offended when a guilty man stubs his toe. On the contrary, it is decent to hope that he will." The principal legacy of Miranda is the creation of an environment in which everything possible has been done to avoid such self-inflicted injuries. Given the relatively modest intellectual faculties of many violent criminals, the incidence of such injuries had always been significant.

If the criminal suspect incriminates himself through police methods that do not involve compulsion, that is a good thing. It is a good thing because it results in accurate fact-finding by the criminal-justice system; it avoids the possibility that an innocent person might be charged with a crime he did not commit; and it promotes public confidence that the police have caught the right person. In other words, it is a good thing because it promotes justice through procedures which most Americans would view as fair.

The most compelling evidence of the desirability of confessions, and of the extent to which Justice White's warnings have been borne out, comes from the before-and-after studies done in the immediate wake of the decision in 1966. One leading study, done in Philadelphia, reported that, before Miranda, an estimated 45 per cent of all criminal suspects made confessions to police officers; following Miranda, that figure dropped to approximately 20 per cent. Another study, done in New York City, found that confession rates fell from 49 per cent to 15 per cent. In Pittsburgh, the confession rate among suspected robbers and murderers fell from 60 per cent to 30 per cent. Adverse effects on confessions were also reported in Chicago, Kansas City, Brooklyn, and New Orleans. The best-estimate consensus among all the studies done on the impact of Miranda is that a lost confession occurs in approximately one of every six, or 16 per cent, of all criminal cases in the United States.

The other leading methodology for such calculations, albeit less exact, is to compare the American confession rate after Miranda with confession rates in countries that follow different approaches to regulating police interrogations. Such comparisons fully confirm the conclusions of the before-and-after studies. Since Miranda, American police appear to obtain confessions in perhaps 40 per cent of all cases. In other countries, police are far more successful. In Great Britain in the 1970s and early 1980s, police, following the "Judges' Rules," gave only a very limited advice of rights. Confession rates were estimated to be 61 to 85 per cent, well above reported American rates. In Canada, confession rates also appear to be substantially higher than in the post-Miranda United States.

Such declines in confession rates as have occurred in the United States since Miranda are extraordinarily harmful to the interests of effective law enforcement. Confession evidence, because it comes from the perpetrator himself, is compelling; but it doesn't stand alone -- it is virtually always subject to corroboration by physical or other evidence. According to the available studies, in about one-quarter (24 per cent) of all criminal cases, confessions or other self-incriminating statements by a suspect are indispensable to a criminal conviction; in many more cases, perhaps an equal number, it can be surmised that they make some difference in terms of the severity of the offense for which a defendant is convicted.

A rough calculation, then, can be made as to the real-world cost of the Miranda rules. Multiplying the estimated 16-point reduction in the confession rate after Miranda by the estimated 24 per cent of cases in which a confession was necessary for conviction produces a figure of 4 per cent of all criminal cases that will be "lost" or never successfully prosecuted because of the rules. That figure may not sound very high, but the cost in absolute numbers of criminal cases is staggering. For FBI-index crimes, each year Miranda results in approximately 28,000 "lost" cases for violent crimes (murder, rape, aggravated assault, and robbery) and 79,000 "lost" cases for serious property crimes (burglary, larceny, and car theft). The bare numbers do not begin to convey the human costs in murders that go unpunished, rapists who remain at large, and treasured heirlooms that are never recovered.

Additionally, the leverage of prosecutors would be reduced in a roughly equal number of cases, resulting in plea bargains more favorable to the defendant and less favorable to the public. Compare these figures with the likely gains expected from other crime-control measures, such as midnight basketball leagues or even long-overdue habeas corpus reform. It seems improbable that any other single needed change in the criminal-justice system would yield anywhere near the number of successful prosecutions that would result from repealing Miranda.

Further, all these figures must be weighed in the light of other figures from the Bureau of Justice Statistics indicating that roughly two-thirds of all violent crime in the United States is committed by repeat offenders. Miranda does not merely defeat justice in the immediate case, but prematurely returns to the streets individuals in this high-risk category.

IF SUCH a cost in "lost" prosecutions were compelled by the clear language of the Constitution or by the dictates of sound public policy, then society might intelligently choose to suffer the tragic consequences. However, the strictures of Miranda are not required by the Constitution -- as the Supreme Court recognized in Miranda itself -- nor are they improvements over alternative methods of achieving fairness and due process in the context of custodial questioning. Miranda instead is a "prophylactic" rule (by its own terms) designed to establish protections against violations of the Fifth Amendment. As Chief Justice Warren remarked in his opinion for the court, "[T]he Constitution does not require any specific code of procedures for protecting the privilege against self-incrimination during custodial interrogation. Congress and the States are free to develop their own safeguards for the privilege, so long as they are fully as effective [as the Miranda rules]." In a later decision, the Court reiterated that the Miranda rights "were not themselves rights protected by the Constitution but were instead measures to ensure that the right against compulsory self- incrimination was protected."

Responding to this invitation, Congress in 1968 enacted new legislation concerning police interrogation, adopting the view that such legislation would be "fully as effective" in maintaining the guarantees of the Fifth Amendment. The new law restored the traditional "voluntariness" test and identified the Miranda warnings as mere factors to be considered by the courts in deciding whether or not a confession was genuinely voluntary.

Because of the hostility of the incumbent attorney general, Ramsey Clark, who believed that it was unconstitutional, the new congressional enactment quickly fell into desuetude. United States Attorneys were ordered not to defend confessions unless they satisfied the Miranda standards. Although later attorneys general, including John Mitchell and Ed Meese, were cognizant of the statute and sensitive to the costs of Miranda, no serious efforts were undertaken to reverse the Johnson Administration policy or to secure any determination of the constitutionality of the law. A recommendation by the Justice Department's Office of Legal Policy in 1987 that an aggressive effort be made to test the law was never adopted as the result of opposition by other agencies within the Department.

THE time has never been better for a test case raising the 1968 law and challenging Miranda. Just last year, Justice Scalia wrote a scathing concurrence chastising the Justice Department for its failure to enforce the law. He observed that that failure "may have produced -- during an era of intense national concern about the problem of run-away crime -- the acquittal and non-prosecution of many dangerous felons, enabling them to continue their depredations upon our citizens." He promised to rule on the statute at the next available opportunity. Congress should make certain that the Justice Department gives him the opportunity by directing it to begin enforcing the statute and reminding the Executive Branch of its constitutional duty to "take care that the laws be faithfully executed." Congress need not enact a new law; it need only ensure that a law already on the books is presented to the Court.

A favorable Court ruling seems quite a realistic prospect. It has long been understood that the Congress has the final say on rules of evidence for federal cases. If the statute admitting voluntary confessions does not confront any constitutional barrier (as the Court has now plainly held), on what possible grounds could the enactment of the people's representatives be struck down? As the Office of Legal Policy observed in recommending that the legitimacy of the Miranda-repeal statute be asserted, "There is every reason to believe that an effort to correct this situation would be successful. . . . It is difficult to see how we could fail to make our case."

While the statute highlights the problems of Miranda's continuing vitality in federal cases, equally perplexing is its applicability to the far greater number of state criminal prosecutions. The Supreme Court's acknowledgment that the Miranda rules are not derived from the Fifth Amendment leads one to ask how the federal judiciary possesses the authority to impose Miranda upon the states. It is an extraordinarily novel proposition of constitutional law that the states are limited in their criminal procedures not merely by the dictates of the Constitution but also by superintending or "prophylactic" rules invented by the federal judiciary. That such a proposition has not yet been challenged can be explained only by the states' tendency to rely upon the Justice Department for leadership on constitutional matters of this sort, as well, perhaps, as by an understandable failure even by the law- enforcement community to recognize the full extent of Miranda's hidden costs.

Ironically, there are better means of enforcing the very protections toward which Miranda is directed. However, as is the case generally with the exercise of uniform national policies, Miranda petrified efforts by the states, which were widespread in the 1960s, to experiment with their custodial interrogation procedures and search for alternatives that might better protect not only society's interest in apprehending criminals but also suspects' interests in preventing coercive questioning.

Perhaps the most effective replacement for Miranda would simply be to videotape or record police interrogations. About one-sixth of all police departments in the country already videotape at least some confessions, and a recent study by the National Institute of Justice concluded that, in addition to providing safeguards for the suspect, videotaping also resulted in the improvement of police interrogation practices, rendered confessions more convincing to judges and juries, and assisted prosecutors in negotiating more favorable plea bargains and guilty pleas. Videotaping would also provide more protection for innocent suspects caught up in the criminal-justice system.

For Miranda is not particularly well tailored to protecting a suspect's rights to be free from coercion. Justice Harlan's point in his Miranda dissent has never been effectively answered: "The new rules are not designed to guard against police brutality or other unmistakably banned forms of coercion. Those who use third-degree tactics and deny them in court are equally able and destined to lie as skillfully about warnings and waivers." It is not clear why police using rubber hoses before Miranda would have thought it necessary to shelve them afterward. Furthermore, once a valid Miranda waiver is obtained, police are relatively free to proceed as they like.

No legacy of the Warren Court has been more devastating to the first civil right of individuals, the right to be protected from attack. Congressional conservatives may choose to place serious procedural reform off-limits. They may do this, in part, because of the burdens of leadership required in order to explain the relationship of procedure to the substantive ability of the justice system to protect society. They may do this because it is easier to deal with public-policy problems whose costs are more visible. However, if effective reform is to be undertaken, unsettling such settled areas of the law as Miranda will be required. Until that time, society can do little more than continue to count Justice White's "unknown number" of killers, rapists, and other criminals who go free because of the criminal-justice innovations of the Warren Court.