Confessions

When the liberal Justices on the Warren Court were replaced by more conservative jurists, many predicted that Miranda would be quickly overruled. The Burger and Rehnquist Courts did cut back on some aspects of the decision but it has not yet been overruled, and in some important respects, its protections have actually been expanded.

Discussion of Miranda doctrine can usefully be organized around two questions: What factors trigger Miranda protection? And what rules govern the use of confessions after Miranda has been triggered?

Unlike Sixth Amendment rights, which are triggered by a formal charge, Miranda protection comes into play whether or not the suspect has been charged, so long as she is interrogated while in custody. With regard to the interrogation requirement, at least one subsequent case has actually expanded on the protection that Miranda first provided. In Rhode Island v. Innis (446 U.S. 291 (1980)), the Court held that Miranda applied not just to express questioning, but also to the "functional equivalent" of express questioning, which the Court defined as "any words or actions on the part of the police (other than those normally attendant to arrest and custody) that the police should know are reasonably likely to elicit an incriminating response from the suspect (446 U.S. at 301). This test is subtly different from the test for Sixth Amendment purposes, which focuses on whether the police "deliberately elicited" an incriminating response. Although it extends Miranda beyond express questioning, the test has the odd property of permitting the police to use techniques designed to get the defendant to confess so long as they are unlikely to succeed. For example, in Innis, the police appealed to Innis's fear that a still unrecovered gun might be found and used by an innocent child. Reasoning that the police had asked no direct questions and that there was no reason to suppose that this appeal would be effective, the Court held that Innis's statements were admissible.

In Illinois v. Perkins (496 U.S. 292 (1990)), the Supreme Court held that even if the interaction is likely to elicit incriminating statements, conversations with undercover agents are not within the purview of Miranda. A comparison with Sixth Amendment doctrine is again instructive. Massiah and Henry make clear that once a defendant has been formally charged with an offense, interaction with an undercover government agent may well violate the right to counsel. In contrast, the Perkins Court reasoned that Miranda was concerned with the risk that the pressure of station house interrogation would "compel" the suspect to speak and that this risk was absent when the suspect was unaware that he was conversing with a government agent.

The second element that triggers Miranda protection is custody. The Miranda Court defined "custody" as interference with a suspect's "freedom of action in any significant way." In the years since Miranda, the Court has made plain that a defendant can be in custody even if he is not at the station house. For example, the Court has applied Miranda to individuals under arrest at their home and to prisoners in jail for unrelated offenses. On the other hand, in an important qualification of the Miranda right, the Court held that an individual stopped briefly on the street for investigative purposes is not in custody in the Miranda sense. Moreover, even if the questioning occurs at the station house, the defendant is not necessarily in custody. According to the Court, the relevant question is neither what the police intend, nor what the suspect thinks. Instead, the question is whether a reasonable person would believe that he has the freedom to leave.

Once Miranda has been triggered, the next question is under what circumstances confessions can be utilized. For purposes of analysis, we can distinguish between three situations: cases where the warnings are not administered; cases where the warnings are administered, but the suspect does not expressly assert his rights; and cases where the warnings are administered, but the defendant does assert his rights.

In the first class of cases, the Miranda Court was unequivocal: in the absence of warnings, the suspect's statements were per se inadmissible. It is with regard to this situation that subsequent cases have cut back most severely on Miranda. Later Courts have fashioned a wide variety of exceptions to Miranda's per se rule. For example, in cases decided in 1971 and 1975, the Court held that even if a defendant has not been warned of his rights, or has been warned and has asserted his rights, his statements can be used against him in cross-examination to throw doubt on the truth of his testimony if he chooses to take the stand.

Similarly, in New York v. Quarles (467 U.S. 649 (1984)), a case decided in 1984, the Court recognized a "public safety" exception to Miranda. The case arose when a woman told the police that she had just been raped by a man who had entered a supermarket. An officer chased after Quarles in the supermarket and stopped him. A frisk revealed that he was wearing a shoulder holster that was empty. The officer asked Quarles where the gun was, and Quarles responded by nodding in the direction of some empty cartons and saying "the gun is over there." The Court held that "the need for answers to questions in a situation posing a threat to the public safety outweighs the need for the prophylactic rule protecting the privilege against self-incrimination" (467 U.S. at 653) and allowed this statement and the gun to be used against the defendant.

The Court has also been generous in permitting the admission of other evidence gained as a result of statements secured in violation of Miranda so long as the Miranda-defective statement itself is excluded. For example, in Michigan v. Tucker (417 U.S. 433 (1974)), the Court held that the testimony of another witness, discovered through a Miranda-defective statement, could be admitted at trial. Similarly, in Oregon v. Elstad (470 U.S. 289 (1985)), the Court held that a suspect's second confession could be used as evidence, even though it was the product of a first confession that had been secured in violation of Miranda.

Most of these cases have a common structure. The Court has tended to characterize Miranda warnings as "procedural safeguards" that are "not themselves rights protected by the Constitution but . . . instead measures to insure that the right against compulsory self-incrimination [is] protected." Having removed the mantle of constitutional necessity from Miranda, the Court then typically balances the benefits and costs of exclusion and concludes that the costs outweigh the benefits.

This technique has provided the Court with a means of restricting Miranda's reach without disowning the decision itself, but it has achieved this objective at the cost of considerable irony. It was, after all, Miranda's opponents who initially complained that the Court lacked constitutional authority to impose the warnings requirement on the states without a constitutional mandate. In contrast, Miranda's defenders insisted that the Court could legitimately place a gloss on the Constitution in order to secure self-incrimination rights. Now, critics and defenders seem to have switched places. The Justices who have participated in Miranda retrenchment have argued that the Court can appropriately require the police to obey procedural requirements not directly mandated by the Constitution, while the Justices criticizing retrenchment have insisted that Miranda can only be justified if the warnings are directly required by the Constitution.

In cases where warnings are given and the defendant does not assert his rights, the modern Court has hewed more closely to Miranda's requirements. Miranda itself provided that the defendant could waive his rights in these circumstances, so long as the waiver was knowing, intelligent, and voluntary. It is fair to say that more recent decisions have not interpreted this requirement generously, but neither have they disowned it. The Court has not required the police to go out of their way to provide more information than contained in the warnings, but neither has it completely forsaken inquiry into the legitimacy of the waiver.

Perhaps surprisingly, in the third class of cases, where the defendant has claimed his rights, the Court has actually gone beyond the Miranda requirements. At first, it appeared that here, too, the Court might cut back. In Michigan v. Mosely (423 U.S. 96 (1975)), the defendant received Miranda warnings and invoked his right to remain silent. Two hours later, a different detective administered the warnings again and asked Mosely about a separate crime. Mosely executed a waiver and made incriminating statements. The Court upheld his resulting conviction, noting that the defendant's right to cut off questioning about the first offense had been "scrupulously honored."

However, six years later, in Edwards v. Arizona (451 U.S. 477 (1981)), the Court took a very different approach to invocation of the right to counsel. Edwards invoked his right to counsel, and the officers ceased questioning him. The next morning, two different detectives came to see Edwards, informed him of his Miranda rights again, and engaged in a colloquy with him, resulting in Edwards's confession. The Court reversed Edwards's conviction. In doing so, it created a new per se rule on top of the Miranda rule: In cases where a suspect invokes his Miranda right to counsel, "a valid waiver of the right cannot be established by showing only that he responded to further police-initiated custodial interrogation even if he has been advised of his rights. [Such a suspect] is not subject to further interrogation by the authorities until counsel has been made available to him, unless the accused himself initiates further communication, exchanges, or conversations with the police" (451 U.S. at 483). Moreover, in Arizona v. Roberson (486 U.S. 675 (1988)), the Court went beyond even the Edwards requirements by applying the rule to a suspect questioned about a unrelated crime by an officer who was unaware of the first invocation of the counsel right. And in Minnick v. Mississippi (498 U.S. 146 (1990)), the Court held that the Edwards rule applied even in a case where the defendant had been allowed to talk to his lawyer prior to police questioning.