Fifth Amendment — Self-Incrimination

Amdt5.4.1 Historical Background on Self-Incrimination

The source of the Self-Incrimination Clause was the maxim “nemo tenetur seipsum accusare,” that “no man is bound to accuse himself.” The maxim is but one aspect of two different systems of law enforcement which competed in England for acceptance; the accusatorial and the inquisitorial. In the accusatorial system, which predated the reign of Henry II but was expanded and extended by him, first the community and then the state by grand and petit juries proceeded against alleged wrongdoers through the examination of others, and in the early years through examination of the defendant as well. The inquisitorial system, which developed in the ecclesiastical courts, compelled the alleged wrongdoer to affirm his culpability through the use of the oath ex officio. Under the oath, an official had the power to make a person before him take an oath to tell the truth to the full extent of his knowledge as to all matters about which he would be questioned; before administration of the oath the person was not advised of the nature of the charges against him, or whether he was accused of crime, and was also not informed of the nature of the questions to be asked.¹

The use of this oath in Star Chamber proceedings, especially to root out political heresies, combined with opposition to the ecclesiastical oath ex officio, led over a long period of time to general acceptance of the principle that a person could not be required to accuse himself under oath in any proceeding before an official tribunal seeking information looking to a criminal prosecution, or before a magistrate investigating an accusation against him with or without oath, or under oath in a court of equity or a court of common law.² The precedents in the colonies are few in number, but following the Revolution six states had embodied the privilege against self-incrimination in their constitutions,³ and the privilege was one of those recommended by several state ratifying conventions for inclusion in a federal bill of rights.⁴ James Madison’s version of the Clause read “nor shall be compelled to be a witness against himself,” but a House amendment inserted “in any criminal case” into the provision.⁵

Amdt5.4.2 Early Doctrine on Self-Incrimination

By the latter part of the eighteenth century, English and early American courts had determined that coerced confessions were potentially excludable from admission at trial because they were untrustworthy.¹ For much of the nineteenth century, the Supreme Court invoked unreliability as the basis for excluding such confessions without mentioning the constitutional bar against self-incrimination.²

In the 1897 case of Bram v. United States, the Court suggested that the Fifth Amendment imposed separate restrictions on a confession’s admissibility. These restrictions focused on the confession’s voluntariness as an indicator of its trustworthiness as evidence. The Court wrote that in criminal trials in federal court, “wherever a question arises whether a confession is incompetent because not voluntary, the issue is controlled by that portion of the Fifth Amendment to the Constitution of the United States, commanding that no person ‘shall be compelled in any criminal case to be a witness against himself.’”³

Although the Supreme Court approved this Fifth Amendment interpretation in subsequent cases⁴ and reaffirmed Bram itself,⁵ the Court held in 1912 that a confession should not be excluded merely because the authorities had not warned a suspect of his right to remain silent.⁶ “In other cases, the Court expressed doubts as to whether the Fifth Amendment’s protection against self-incrimination—rather than a common-law principle that forced confessions were untrustworthy—required exclusion of involuntary confessions from federal criminal trials.”⁷ Because the Supreme Court had not yet ruled that the Self-Incrimination Clause applies to states through the Fourteenth Amendment, admissibility of confessions in state courts continued to be governed under due process standards developed from common-law principles. It was only in the 1960s, after the Court extended the Self-Incrimination Clause to the states, that a divided Court reaffirmed and extended the 1897 Bram ruling and imposed on both federal and state trial courts new rules for admitting or excluding confessions and other admissions made to police during custodial interrogation.⁸

Amdt5.4.3 General Protections Against Self-Incrimination Doctrine and Practice

The Court has settled upon the principle that the Clause serves two interrelated interests: the preservation of an accusatorial system of criminal justice, which goes to the integrity of the judicial system, and the preservation of personal privacy from unwarranted governmental intrusion.¹ To protect these interests and to preserve these values, the privilege “is not to be interpreted literally.” Rather, the “sole concern [of the privilege] is, as its name indicates, with the danger to a witness forced to give testimony leading to the infliction of penalties affixed to the criminal acts.”² Furthermore, “[t]he privilege afforded not only extends to answers that would in themselves support a conviction . . . but likewise embraces those which would furnish a link in the chain of evidence needed to prosecute . . .”³

The privilege against self-incrimination parries the general obligation to provide testimony under oath when called upon, but it also applies in police interrogations. In all cases, the privilege must be supported by a reasonable fear that a response will be incriminatory. The issue is a matter of law for a court to determine,⁴ and therefore, with limited exceptions, one must claim the privilege to benefit from it.⁵ Otherwise, silence in the face of questioning may be insufficient to invoke the privilege because it may not afford an adequate opportunity either to test whether information withheld falls within the privilege or to cure a violation through a grant of immunity.⁶ A witness who fails to claim the privilege explicitly when an affirmative claim is required is deemed to have waived it, and waiver may be found where the witness has answered some preliminary questions but desires to stop at a certain point.⁷ However, an assertion of innocence in conjunction with a claim of the privilege does not obviate the right of witnesses to invoke it, as their responses still may provide the government with evidence it may later seek to use against them.⁸

Although individuals must have reasonable cause to apprehend danger and cannot be the judge of the validity of their claims, a court that would deny a claim of the privilege must be “perfectly clear, from a careful consideration of all the circumstances in the case, that the individual is mistaken, and that the answer[s] cannot possibly have such tendency to incriminate.”⁹ To reach a determination, furthermore, a trial judge may not require a witness to disclose so much of the danger as to render the privilege nugatory. As the Court observed:

[I]f the witness, upon interposing his claim, were required to prove the hazard . . . he

would be compelled to surrender the very protection which the privilege is designed to

guarantee. To sustain the privilege, it need only be evident from the implications of the

question, in the setting in which it is asked, that a responsive answer to the question or

an explanation of why it cannot be answered might be dangerous because injurious

disclosure could result.¹⁰

The privilege against self-incrimination is a personal one and cannot be used by or on behalf of any organization, such as a corporation. Thus, a corporation cannot object on self-incrimination grounds to a subpoena of its records and books or to the compelled testimony of those corporate agents who have been given personal immunity from criminal prosecution.¹¹ Nor may a corporate official with custody of corporate documents that incriminate him personally resist their compelled production on the assertion of his personal privilege.¹²

A witness has traditionally been able to claim the privilege in any proceeding whatsoever in which testimony is legally required when his answer might be used against him in that proceeding or in a future criminal proceeding or when it might be exploited to uncover other evidence against him.¹³ Incrimination is not complete once guilt has been adjudicated, and hence the privilege may be asserted during the sentencing phase of trial.¹⁴ Conversely, there is no valid claim on the ground that the information sought can be used in proceedings which are not criminal in nature,¹⁵ and there can be no valid claim if there is no criminal prosecution¹⁶ The Court in recent years has also applied the privilege to situations, such as police interrogation of suspects, in which there is no legal compulsion to speak.¹⁷

What the privilege protects against is compulsion of “testimonial” disclosures. Thus, the clause is not offended by such non-testimonial compulsions as requiring a person in custody to stand or walk in a police lineup, to speak prescribed words, to model particular clothing, or to give samples of handwriting, fingerprints, or blood.¹⁸ A person may be compelled to produce specific documents even though they contain incriminating information.¹⁹ If, however, the existence of specific documents is not known to the government, and the act of production informs the government about the existence, custody, or authenticity of the documents, then the privilege is implicated.²⁰ Application of these principles resulted in a holding that the Independent Counsel could not base a prosecution on incriminating evidence identified and produced as the result of compliance with a broad subpoena for all information relating to the individual’s income, employment, and professional relationships.²¹

The protection is against “compulsory” incrimination, and traditionally the Court has treated within the Clause only those compulsions which arise from legally enforceable obligations, culminating in imprisonment for refusal to testify or to produce documents.²² The compulsion need not be imprisonment, but can also be termination of public employment²³ or disbarment of a lawyer²⁴ as a legal consequence of a refusal to make incriminating admissions. The degree of coercion may also prove decisive, the Court having ruled that moving a prisoner from a medium security unit to a maximum security unit was insufficient to compel him to incriminate himself in spite of the attendant loss of privileges and the harsher living conditions.²⁵ However, although it appears that prisoners²⁶ and probationers²⁷ have less protection than others do, the Court has not developed a clear doctrinal explanation to identify the differences between permissible and impermissible coercion.²⁸

It has long been the rule that a defendant who takes the stand on his own behalf does so voluntarily and cannot then claim the privilege to defeat cross-examination on matters reasonably related to the subject matter of his direct examination,²⁹ and that such a defendant may be impeached by proof of prior convictions.³⁰ But, in Griffin v. California,³¹ the Court refused to permit prosecutorial or judicial comment to the jury upon a defendant’s refusal to take the stand on his own behalf, because such comment was a “penalty imposed by courts for exercising a constitutional privilege” and “[i]t cuts down on the privilege by making its assertion costly.”³² Prosecutors’ comments violating the Griffin rule can nonetheless constitute harmless error.³³ Nor may a prosecutor impeach a defendant’s trial testimony through use of the fact that upon his arrest and receipt of a Miranda warning he remained silent and did not give the police the exculpatory story he told at trial.³⁴ But where the defendant took the stand and testified, the Court permitted the impeachment use of his pre-arrest silence when that silence had in no way been officially encouraged, through a Miranda warning or otherwise.³⁵

Further, the Court held inadmissible at the subsequent trial a defendant’s testimony at a hearing to suppress evidence wrongfully seized, because use of the testimony would put the defendant to an impermissible choice between asserting his right to remain silent and invoking his right to be free of illegal searches and seizures.³⁶ The Court also proscribed the introduction at a second trial of the defendant’s testimony at his first trial, given to rebut a confession which was subsequently held inadmissible, since the testimony was in effect “fruit of the poisonous tree” and had been “coerced” from the defendant through use of the confession.³⁷ Potentially most far-reaching was a holding that invalidated the penalty structure of a statute under which defendants could escape a possible death sentence by entering a guilty plea; the statute “needlessly encourage[d]” waivers of defendant’s Fifth Amendment right to plead not guilty and his Sixth Amendment right to a jury trial.³⁸

Although this “needless encouragement” test assessed the nature of the choice required to be made by defendants against the strength of the governmental interest in the system requiring the choice, the Court soon developed another test stressing the voluntariness of the choice. A guilty plea entered by a defendant who correctly understands the consequences of the plea is voluntary unless coerced or obtained under false pretenses; moreover, there is no impermissible coercion where the defendant has the effective assistance of counsel.³⁹ The Court in an opinion by Justice John Marshall Harlan then formulated still another test in holding that a defendant in a capital case in which the jury in one process decides both guilt and sentence could be put to a choice between remaining silent on guilt or admitting guilt and being able to put on evidence designed to mitigate the possible sentence. The pressure to take the stand in response to the sentencing issue, said the Court, was not so great as to impair the policies underlying the Self-Incrimination Clause, policies described in this instance as proscription of coercion and of cruelty in putting the defendant to an undeniably “hard” choice.⁴⁰ Similarly, the Court held that requiring a defendant to give notice to the prosecution before trial of his intention to rely on an alibi defense and to give the names and addresses of witnesses who will support it does not violate the Clause.⁴¹ Nor does it violate a defendant’s self-incrimination privilege to create a presumption upon the establishment of certain basic facts from which the jury may infer the defendant’s guilt unless he rebuts the presumption.⁴²

The obligation to testify is not relieved by this Clause, if, regardless of whether incriminating answers are given, a prosecution is precluded,⁴³ or if the result of the answers is not incrimination but rather harm to reputation or exposure to infamy or disgrace.⁴⁴ The Clause does not prevent a public employer from discharging an employee who, in an investigation specifically and narrowly directed at the performance of the employee’s official duties, refuses to cooperate and to provide the employer with the desired information on grounds of self-incrimination.⁴⁵ But it is unclear under what other circumstances a public employer may discharge an employee who has claimed his privilege before another investigating agency.⁴⁶

Finally, the rules established by the Clause and the judicial interpretations apply against the states to the same degree that they apply against the Federal Government,⁴⁷ and neither sovereign can compel discriminatory admissions that would incriminate the person in the other jurisdiction.⁴⁸ There is no “cooperative internationalism” that parallels the cooperative federalism and cooperative prosecution on which application against states is premised, and consequently concern with foreign prosecution is beyond the scope of the Self-Incrimination Clause.⁴⁹

Amdt5.4.4 Required Records Doctrine

Although the privilege is applicable to an individual’s papers and effects,¹ it does not extend to corporate persons; hence corporate records, as has been noted, are subject to compelled production.² In fact, however, the Court has greatly narrowed the protection afforded in this area to natural persons by developing the “required records” doctrine. That is, it has held “that the privilege which exists as to private papers cannot be maintained in relation to ‘records required by law to be kept in order that there may be suitable information of transactions which are the appropriate subjects of governmental regulation and the enforcement of restrictions validly established.’”³ This exception developed out of, as Justice Felix Frankfurter showed in dissent, the rule that documents which are part of the official records of government are wholly outside the scope of the privilege; public records are the property of government and are always accessible to inspection. Because government requires certain records to be kept to facilitate the regulation of the business being conducted, so the reasoning goes, the records become public at least to the degree that government could always scrutinize them without hindrance from the record-keeper. “If records merely because required to be kept by law ipso facto become public records, we are indeed living in glass houses. Virtually every major public law enactment—to say nothing of State and local legislation—has record-keeping provisions. In addition to record-keeping requirements, is the network of provisions for filing reports. Exhaustive efforts would be needed to track down all the statutory authority, let alone the administrative regulations, for record-keeping and reporting requirements. Unquestionably they are enormous in volume.”⁴

“It may be assumed at the outset that there are limits which the government cannot constitutionally exceed in requiring the keeping of records which may be inspected by an administrative agency and may be used in prosecuting statutory violations committed by the record-keeper himself.”⁵ But the only limit that the Court suggested in Shapiro was that there must be “a sufficient relation between the activity sought to be regulated and the public concern so that the Government can constitutionally regulate or forbid the basic activity concerned, and can constitutionally require the keeping of particular records, subject to inspection by the Administrator.”⁶ That there are limits established by the Self-Incrimination Clause itself rather than by a subject matter jurisdiction test is evident in the Court’s consideration of reporting and disclosure requirements implicating but not directly involving the required-records doctrine.

Amdt5.4.5 Immunity

Under the Fifth Amendment, the government cannot compel a person to be “a witness against himself . . . .,” although a person may waive the privilege against self-incrimination by declining to assert it, specifically disclaiming it, or testifying on the same matters prior to asserting the privilege.¹ In addition, Congress has passed immunity statutes, which allow “the person presiding over the proceeding” to compel a witness, who has asserted his or her privilege against self-incrimination, to testify, provided that “no testimony or other information compelled under the order (or any information directly or indirectly derived from such testimony or other information) may be used against the witness in any criminal case, except a prosecution for perjury, giving a false statement, or otherwise failing to comply with the order.”²

As the Supreme Court has recognized immunity statutes “seek a rational accommodation between the imperatives of the privilege [of self-incrimination] and the legitimate demands of government to compel citizens to testify.”³ Parliament appears to have enacted the first immunity statute in 1710,⁴ which, in turn, was widely copied in the colonies. Congress enacted the first federal immunity statute in 1857, providing for immunization of any person who testified before a congressional committee from prosecution for any matter “touching which” he had testified.⁵

The Supreme Court’s decision in Counselman v. Hitchcock soon rendered Congress’s immunity statute unenforceable.⁶ In Counselman, the Court held that an analogous limited-immunity statute was unconstitutional because it did not confer an immunity coextensive with the privilege it replaced. The Court’s reasoning in Counselman was ambiguous; it identified two faults in the statute. First, the statute did not proscribe “derivative” evidence.⁷ Second, it prohibited only future use of the compelled testimony.⁸ The latter language accentuated a division between adherents of “transactional” immunity and of “use” immunity which has continued to the present.⁹

Following Counselman, Congress enacted a statute that required transactional immunity in exchange for compelled testimony.¹⁰ The Court sustained this law in Brown v. Walker.¹¹ In 1956, the Court broadly reaffirmed Walker.¹² Because the immunity acts passed after Walker were generally transactional immunity statutes,¹³ the question of the constitutional sufficiency of use immunity did not arise. The dicta in cases dealing with immunity, the Court continued to assert the necessity of transactional immunity.¹⁴

The Court’s incorporation of the Self-Incrimination Clause against the states in 1964 raised new considerations. In particular, state officials lacked the power to confer immunity from federal prosecution.¹⁵ As a consequence, concerns arose that if states could not compel testimony because that they lacked authority to immunize a witness in a subsequent “foreign” prosecution, their law-enforcement efforts could suffer. To avoid this outcome, the Court emphasized the “use” restriction rationale of Counselman and announced that as a “constitutional rule, a state witness could not be compelled to incriminate himself under federal law unless federal authorities were precluded from using either his testimony or evidence derived from it.”¹⁶ After this decision, Congress enacted a statute replacing all prior immunity statutes and adopting a use-immunity restriction only.¹⁷ The Supreme Court upheld this statute in Kastigar v. United States.¹⁸

Amdt5.4.6 Withdrawal of Government Benefits

The line of cases begins with United States v. Sullivan,¹ in which a unanimous Court held that the Fifth Amendment did not privilege a bootlegger in not filing an income tax return because the filing would have disclosed the illegality in which he was engaged. “It would be an extreme if not an extravagant application of the Fifth Amendment to say that it authorized a man to refuse to state the amount of his income because it had been made in crime,” Justice Oliver Wendell Holmes stated for the Court.² However, “[i]f the form of return provided called for answers that the defendant was privileged from making he could have raised the objection in the return . . . .”³ Using its taxing power to reach gambling activities over which it might otherwise not have had jurisdiction,⁴ Congress enacted a complicated statute imposing an annual occupational tax on gamblers and an excise tax on all their wages, and coupled the tax with an annual registration requirement under which each gambler must file with the Internal Revenue Service (IRS) a declaration of his business with identification of his place of business and his employees and agents, filings which were made available to state and local law enforcement agencies. These requirements were upheld by the Court against self-incrimination challenges on the three grounds that (1) the privilege did not excuse a complete failure to file, (2) because the threshold decision to gamble was voluntary, the required disclosures were not compulsory, and (3) because registration required disclosure only of prospective conduct, the privilege, limited to past or present acts, did not apply.⁵

Constitutional limitations appeared, however, in Albertson v. SACB,⁶ which struck down under the Self-Incrimination Clause an order pursuant to statute requiring registration by individual members of the Communist Party or associated organizations. “In Sullivan the questions in the income tax return were neutral on their face and directed at the public at large, but here they are directed at a highly selective group inherently suspect of criminal activities. Petitioners’ claims are not asserted in an essentially noncriminal and regulatory area of inquiry, but against an inquiry in an area permeated with criminal statutes, where response to any of the form’s questions in context might involve the petitioners in the admission of a crucial element of a crime.”⁷

The gambling tax reporting scheme was next struck down by the Court.⁸ Because of the pervasiveness of state laws prohibiting gambling, said Justice John Marshall Harlan for the Court, “the obligations to register and to pay the occupational tax created for petitioner ‘real and appreciable,’ and not merely ‘imaginary and unsubstantial,’ hazards of self-incrimination.”⁹ Overruling Kahriger and Lewis, the Court rejected its earlier rationales. Registering per se would have exposed a gambler to dangers of state prosecution, so Sullivan did not apply.¹⁰ Any contention that the voluntary engagement in gambling “waived” the self-incrimination claim, because there is “no constitutional right to gamble,” would nullify the privilege.¹¹ And the privilege was not governed by a “rigid chronological distinction” so that it protected only past or present conduct, but also reached future self-incrimination the danger of which is not speculative and insubstantial.¹² Significantly, then, Justice Harlan turned to distinguishing the statutory requirements here from the “required records” doctrine of Shapiro. “First, petitioner . . . was not . . . obliged to keep and preserve records ‘of the same kind as he has customarily kept’; he was required simply to provide information, unrelated to any records which he may have maintained, about his wagering activities. This requirement is not significantly different from a demand that he provide oral testimony . . . . Second, whatever ‘public aspects’ there were to the records at issue in Shapiro, there are none to the information demanded from Marchetti. The Government’s anxiety to obtain information known to a private individual does not without more render that information public; if it did, no room would remain for the application of the constitutional privilege. Nor does it stamp information with a public character that the government has formalized its demands in the attire of a statute; if this alone were sufficient, the constitutional privilege could be entirely abrogated by any Act of Congress. Third, the requirements at issue in Shapiro were imposed in ‘an essentially non-criminal and regulatory area of inquiry’ while those here are directed to a ‘selective group inherently suspect of criminal activities.’ . . . The United States’ principal interest is evidently the collection of revenue, and not the punishment of gamblers, . . . but the characteristics of the activities about which information is sought, and the composition of the groups to which inquiries are made, readily distinguish this situation from that in Shapiro.”¹³

Most recent in this line of cases is California v. Byers,¹⁴ which indicates that the Court has yet to settle on an ascertainable standard for judging self-incrimination claims in cases where government is asserting an interest other than criminal law enforcement. Byers sustained the constitutionality of a statute which required the driver of any automobile involved in an accident to stop and give his name and address. The state court had held that a driver who reasonably believed that compliance with the statute would result in self-incrimination could refuse to comply. A plurality of the Court, however, determined that Sullivan and Shapiro applied and not the Albertson-Marchetti line of cases, because the purpose of the statute was to promote the satisfaction of civil liabilities resulting from automobile accidents and not criminal prosecutions, and because the statute was directed to all drivers and not to a group which was either “highly selective” or “inherently suspect of criminal activities.” The combination of a noncriminal motive with the general character of the requirement made too slight for reliance the possibility of incrimination.¹⁵ Justice Harlan concurred to make up the majority on the disposition of the case, disagreeing with the plurality’s conclusion that the stop and identification requirement did not compel incrimination.¹⁶ However, the Justice thought that, where there is no governmental purpose to enforce a criminal law and instead government is pursuing other legitimate regulatory interests, it is permissible to apply a balancing test between the government’s interest and the individual’s interest. When he balanced the interests protected by the Amendment—protection of privacy and maintenance of an accusatorial system—with the noncriminal purpose, the necessity for self-reporting as a means of securing information, and the nature of the disclosures required, Justice Harlan voted to sustain the statute.¹⁷ Byers was applied in Baltimore Dep’t of Social Services v. Bouknight¹⁸ to uphold a juvenile court’s order that the mother of a child under the court’s supervision produce the child. Although in this case the mother was suspected of having abused or murdered her child, the order was justified out of concern for the child’s safety—a “compelling reason[ ] unrelated to criminal law enforcement.”¹⁹ Moreover, because the mother had custody of her previously abused child only as a result of the juvenile court’s order, the Court analogized to the required records cases to conclude that the mother had submitted to the requirements of the civil regulatory regime as the child’s “custodian.”

Amdt5.4.7.1 Early Doctrine and Custodial Interrogation

By the latter part of the eighteenth century English and early American courts had developed a rule that coerced confessions were potentially excludable from admission at trial because they were testimonially untrustworthy.¹ The Supreme Court at times continued to ground exclusion of involuntary confessions on this common law foundation of unreliability without any mention of the constitutional bar against self-incrimination. Consider this dictum from an 1884 opinion: “[V]oluntary confession of guilt is among the most effectual proofs in the law, . . . [b]ut the presumption upon which weight is given to such evidence, namely, that one who is innocent will not imperil his safety or prejudice his interests by an untrue statement, ceases when the confession appears to have been made either in consequence of inducements of a temporal nature, held out by one in authority, touching the charge preferred, or because of a threat or promise by or in the presence of such person, which, operating upon the fears or hopes of the accused, in reference to the charge, deprives him of that freedom of will or self-control essential to make his confession voluntary within the meaning of the law.”² Subsequent cases followed essentially the same line of thought.³

Then, language in the 1897 case of Bram v. United States opened the door to eventually extending the doctrinal basis for analyzing the admissibility of a confession beyond the common-law test that focused on voluntariness as an indicator of the confession’s trustworthiness as evidence.“In criminal trials, in the courts of the United States, wherever a question arises whether a confession is incompetent because not voluntary, the issue is controlled by that portion of the Fifth Amendment to the Constitution of the United States, commanding that no person ‘shall be compelled in any criminal case to be a witness against himself.’”⁴ However, though this approach⁵ and the case itself were subsequently approved in several cases,⁶ the Court would still hold in 1912 that a confession should not be excluded merely because the authorities had not warned a suspect of his right to remain silent,⁷ and more than once later opinions could doubt “whether involuntary confessions are excluded from federal criminal trials on the ground of a violation of the Fifth Amendment’s protection against self-incrimination, or from a rule that forced confessions are untrustworthy. . . .”⁸ One reason for this was that the Self-Incrimination Clause had not yet been made applicable to the states, thereby requiring that the admissibility of confessions in state courts be determined under due process standards developed from common-law principles. It was only after the Court extended the Self-Incrimination Clause to the states that a divided Court reaffirmed and extended the 1897 Bram ruling and imposed on both federal and state trial courts new rules for admitting or excluding confessions and other admissions made to police during custodial interrogation.⁹

Amdt5.4.7.2 Pre-Miranda Self-Incrimination Doctrine (1940s to 1960s)

From the 1940s to the 1960s, the Supreme Court decided a series of cases that explained when a confession’s admission in a criminal trial violates the Fifth Amendment’s self-incrimination doctrine. In its 1943 decision in McNabb v. United States,¹ the Supreme Court held that confessions obtained after an “unnecessary delay” in presenting a suspect for arraignment after arrest could not be admitted in criminal trials.² This rule, developed pursuant to the Court’s supervisory power over the lower federal courts³ and hence not applicable to the states,⁴ was designed to implement guarantees that the Federal Rules of Criminal Procedure provides to defendants.⁵ The rule was informed by concern over incommunicado interrogation and coerced confessions.⁶ Although the Court never specified a minimum time after which delay in presenting a suspect for arraignment could invalidate a confession, Congress in 1968 legislated a six-hour period for interrogation following arrest before the suspect must be presented.⁷

Supreme Court cases from this time period addressed when a confession would become inadmissible because it had been obtained through coercive interrogation tactics. Many of the early cases disclosed clear instances of coercion that the Court determined had produced involuntary confessions. For example, the Court had little difficulty concluding that physical torture was coercive.⁸ Moreover, in its first confession case arising from a state court proceeding, the Supreme Court set aside a conviction based solely on confessions obtained through repeated whippings of the defendant with ropes and studded belts.⁹ However, the Court also condemned other overtly coercive tactics that did not amount to physical torture. For example, in Chambers v. Florida,¹⁰ the Court held that five days of prolonged questioning and incommunicado detention made subsequent confessions involuntary.

Although the Court did not hold that prolonged questioning by itself made a resulting confession involuntary,¹¹ it increasingly found coercion present even in intermittent questioning over a period of days of incommunicado detention.¹² In Ashcraft v. Tennessee,¹³ the Court held that a confession was inadmissible when it was obtained after almost 36 hours of continuous questioning under powerful electric lights by multiple officers, experienced investigators, and highly trained lawyers. Similarly, Ward v. Texas voided a conviction based on a confession obtained after three days of questioning during which the defendant was driven from county to county and told falsely of a danger of lynching.¹⁴ In Stein v. New York,¹⁵ however, the Court affirmed convictions of experienced criminals who had confessed after twelve hours of intermittent questioning over a period of 32 hours of incommunicado detention. The majority stressed that the correct approach was to balance “the circumstances of pressure against the power of resistance of the person confessing. What would be overpowering to the weak of will or mind might be utterly ineffective against an experienced criminal.”¹⁶

In resolving cases involving confessions during interrogation, the Court attempted by considering the “totality of the circumstances” to determine whether a confession was “voluntary” and admissible or “coerced” and inadmissible. The Court attempted to balance law enforcement’s need to question suspects against concerns about undue coercion.¹⁷ Although the Court has often focused on the nature of the coercion without regard to the individual characteristics of the suspect,¹⁸ the Court has occasionally determined that some suspects are susceptible to even mild coercion because of their age or intelligence.¹⁹ In some cases, a single factor indicated that the confession was involuntary.²⁰ However, in other cases, the Court recited a number of contributing factors without ranking any factor above the others, including the defendant’s age and intelligence; whether the defendant was held incommunicado, denied requested counsel, or deprived of access to friends; and whether the authorities employed trickery in obtaining a confession.²¹ The Court also held that confessions induced through the exploitation of some illegal action, such as an illegal arrest²² or an unlawful search and seizure,²³ were inadmissible.

Amdt5.4.7.3 Miranda and Its Aftermath

In Miranda v. Arizona, the Supreme Court held that prosecutors may not use statements obtained during a custodial interrogation unless the interrogation was conducted pursuant to certain procedural safeguards. Specifically, the Court concluded that such statements are inadmissible at trial unless the individual subject to interrogation was informed of his or her right to remain silent, that any statements could be used against the subject in subsequent proceedings, and of his or her right to an attorney.¹ The Miranda Court regarded police interrogation as inherently coercive. The Court explained that the relevant “Miranda warnings” were necessary to ensure that suspects were not stripped of their ability to make a free and rational choice between speaking and not speaking.² Although the Miranda decision became highly controversial, the Court has continued to adhere to it.³ However, the Court has created exceptions to the Miranda warnings over the years, and referred to the warnings as “prophylactic”⁴ and “not themselves rights protected by the Constitution.”⁵

In Dickerson v. United States,⁶ the Court addressed a foundational issue, finding that Miranda was a “constitutional decision” that could not be overturned by statute, and consequently that 18 U.S.C. § 3501, which provided for a less strict “voluntariness” standard for the admissibility of confessions, could not be sustained.

Consistent application of Miranda’s holding on warnings to state proceedings necessarily implied a constitutional basis for Miranda, the Court explained, because federal courts “hold no supervisory authority over state judicial proceedings.”⁷ Moreover, Miranda itself had purported to guide law enforcement agencies and courts.⁸ However, even if Miranda is rooted in the Constitution, the Court has indicated that this does not mean a precise articulation of its required warnings is “immutable.”⁹

In addition to finding that Miranda had “constitutional underpinnings,” the Dickerson Court also rejected a request to overrule Miranda. “Whether or not we would agree with Miranda’s reasoning and its resulting rule, were we addressing the issue in the first instance,” Chief Justice William Rehnquist wrote for the seven-Justice majority, “the principles of stare decisis weigh heavily against overruling it now.” There was no special justification for overruling the decision; subsequent cases had not undermined the decision’s doctrinal underpinnings, but rather had “reaffirm[ed]” its “core ruling.” Moreover, Miranda warnings had “become so embedded in routine police practice [that they] have become part of our national culture.”¹⁰

As to the viability of Miranda claims in federal habeas corpus cases, the Court suggested in 1974 that most claims could be disallowed¹¹ but reversed course in 1993. The Court ruled in Withrow v. Williams that Miranda protects a fundamental trial right of the defendant, unlike the Fourth Amendment exclusionary rule addressed in Stone v. Powell.¹² Thus, claimed violations of Miranda merited federal habeas corpus review because they related to the correct ascertainment of guilt.¹³ The Miranda rule differed from the Mapp v. Ohio¹⁴ exclusionary rule because Mapp’s primary purpose was to deter future Fourth Amendment violations, which the Court opined would only be marginally advanced by allowing collateral review.¹⁵ A further consideration was that eliminating review of Miranda claims would not significantly reduce federal habeas review of state convictions, because most Miranda claims could be recast in terms of due process denials resulting from admission of involuntary confessions.¹⁶

The Court further explored the constitutional nature of Miranda in its 2022 case Vega v. Tekoh.¹⁷ In Vega, the Court reiterated that while Miranda was a constitutional decision that adopted constitutional rules, those rules were set forth by the Court as a way to safeguard constitutional rights under the Fifth Amendment.¹⁸ Therefore, a Miranda violation does not necessarily constitute a violation of the Constitution.¹⁹ The Court concluded that because a Miranda violation is not a violation of a constitutional right, it is not actionable under 42 U.S.C. § 1983, which requires someone suffer the deprivation of [a] right . . . secured by the Constitution.²⁰

Amdt5.4.7.4 Custodial Interrogation Standard

Law enforcement officers must give Miranda warnings prior to “questioning initiated by law enforcement officers after a person has been taken into custody or otherwise deprived of his freedom of action in any significant way.”¹ Such warnings are thus required when a person is (1) taken into custody, and (2) subject to interrogation. The Supreme Court has explained that whether a person is “in custody” depends on the results of a two-part test that considers whether a reasonable person in the suspect’s shoes would feel that he could freely exercise his right against self-incrimination and the degree to which the suspect’s freedom of action is restricted.

First, whether a person is “in custody” during questioning depends on the degree of coercive pressure imposed on him. The Court applies an objective, context-specific test that considers the degree of intimidation that a reasonable person in the suspect’s shoes would feel if he were to freely exercise his right against self-incrimination. A police officer’s subjective and undisclosed view that a person being interrogated is a criminal suspect is not relevant for Miranda purposes, nor is the subjective view of the person being questioned.² However, age may weigh in favor of requiring Miranda warnings if the detainee is a juvenile.³

Second, the Supreme Court has considered whether various restrictions on a person’s freedom of action constitute taking that person into custody for purposes of Miranda. The Court has determined that, for example, an ordinary traffic stop does not to amount to Miranda “custody.”⁴ Moreover, interrogating a prison inmate about previous outside conduct does not necessarily amount to custody, even if the inmate is isolated from the general prison population for questioning.⁵ The fact that a suspect may be present in a police station does not necessarily mean, absent further restrictions, that questioning is custodial.⁶ By itself, the fact that the suspect is in his home or other familiar surroundings will not ordinarily lead to a conclusion that the inquiry was custodial.⁷ However, questioning a person upon arrest in his home may be custodial.⁸ When a person has been subjected to Miranda custody that custody ends when he is free to resume his normal life activities after questioning.⁹ Nevertheless, a break in custody may not end all Miranda implications for subsequent custodial interrogations.¹⁰

In addition to requiring that a person be taken into custody to trigger Miranda warnings, such warnings must precede custodial interrogation. It is not necessary under Miranda that the police ask a question in order to “interrogate” the suspect, as demonstrated in Rhode Island v. Innis.¹¹ There, police had apprehended the defendant as a murder suspect but had not found the weapon used. While he was being transported to police headquarters in a squad car, the officers did not question the defendant, who after receiving Miranda warnings had wanted to consult a lawyer. However, the officers discussed among themselves that a school for children with disabilities was near the crime scene, and that they hoped to find the weapon before it injured a child. The defendant then took them to the weapon’s hiding place.

Unanimously rejecting a contention that only express questioning violates Miranda, the Court said: We conclude that the Miranda safeguards come into play whenever a person in custody is subjected to either express questioning or its functional equivalent. That is to say, the term ‘interrogation’ under Miranda refers not only to express questioning, but also to 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. The latter portion of this definition focuses primarily upon the perceptions of the suspect, rather than the intent of the police. This focus reflects the fact that the Miranda safeguards were designed to vest a suspect in custody with an added measure of protection against coercive police practices, without regard to objective proof of the underlying intent of the police.¹²

The Court, however, concluded that the officers’ conversation was not the functional equivalent of questioning and the evidence was admissible.¹³ A later Court applied Innis in Arizona v. Mauro¹⁴ to hold that a suspect who had requested an attorney was not “interrogated” when the police instead brought the suspect’s wife, who also was a suspect, to speak with him in the police’s presence. The majority emphasized that the suspect’s wife had asked to speak with her husband; therefore, the meeting was not a police-initiated ruse designed to elicit a response from the suspect. Furthermore, the meeting could not be characterized as a police attempt to use the coercive nature of confinement to extract a confession that would not have been given in an unrestricted environment.

In Estelle v. Smith,¹⁵ the Court held that a court-ordered jailhouse interview by a psychiatrist seeking to determine the defendant’s competency to stand trial constituted “interrogation” with respect to testimony on issues of guilt and punishment. Thus, the psychiatrist’s conclusions about the defendant’s dangerousness were inadmissible at the capital sentencing phase of the trial because the defendant had not received his Miranda warnings prior to the interview. That a psychiatrist designated to conduct a neutral competency examination had questioned the defendant, rather than a police officer, was “immaterial,” the Court concluded, because the psychiatrist’s testimony at the penalty phase changed his role from one of neutrality to that of the prosecution’s agent.¹⁶

Amdt5.4.7.5 Miranda Requirements

Miranda requires that before a suspect in custody is interrogated, he must be given full warnings (or the equivalent) of his rights. Specifically, the suspect must receive express warnings of his right to remain silent; that anything he says may be used as evidence against him; that he has a right to counsel; and that, if he cannot afford counsel, he is entitled to an appointed attorney.¹ In a later decision, the Court held that it is unnecessary for the police to give the warnings as a verbatim recital of the words in the Miranda opinion itself, so long as the words used “fully conveyed” to a defendant his rights.²

Once a warned suspect asserts his right to silence and requests counsel, the police must scrupulously respect this assertion. The Miranda Court stated that once a warned suspect “indicates in any manner, at any time prior to or during questioning, that he wishes to remain silent, the interrogation must cease.” Further, if the suspect requests the assistance of counsel during interrogation, questioning must cease until he has counsel.³

Subsequently, the Court has often barred the police from continuing (or reinitiating) interrogation with a suspect requesting counsel until counsel is present, except when the suspect himself initiates further communications. In Edwards v. Arizona,⁴ initial questioning ceased as soon as the suspect requested counsel, and the police returned the suspect to his cell. Questioning resumed the following day only after different police officers had confronted the suspect and again warned him of his rights; the suspect agreed to talk and thereafter incriminated himself. Nonetheless, the Court held, “when an accused has invoked his right to have counsel present during custodial interrogation, a valid waiver of that 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. We further hold that an accused . . . , having expressed his desire to deal with the police only through counsel, 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.”⁵ The Edwards rule bars police-initiated questioning stemming from a separate investigation as well as questioning relating to the crime for which the suspect was arrested.⁶ It also applies to interrogation by officers of a different law enforcement authority.⁷

On the other hand, the Edwards rule requiring that a lawyer be provided to a suspect who had requested one in an earlier interrogation does not apply once there has been a meaningful break in custody. The Court in Maryland v. Shatzer⁸ characterized the Edwards rule as a judicially prescribed precaution against using the coercive pressure of prolonged custody to badger a suspect who has previously requested counsel into talking without one. However, after a suspect has been released to resume his normal routine for a sufficient period to dissipate the coercive effects of custody, a period set at 14 days by the Shatzer Court, the rationale for solicitous treatment ceases. If the police take the suspect into custody again, the options for questioning him are no longer limited to suspect-initiated talks or providing counsel. Rather, the police may issue new Miranda warnings and proceed accordingly.⁹ The Court has not extended the Edwards rule explicitly to other aspects of the Miranda warnings.¹⁰

Amdt5.4.7.6 Miranda Exceptions

A properly warned suspect may waive his Miranda rights and submit to custodial interrogation. Miranda recognized that a suspect may voluntarily and knowingly give up his rights and respond to questioning, but the Court also cautioned that the prosecution bore a “heavy burden” to establish that a valid waiver had occurred.¹ The Court continued: “[a] valid waiver will not be presumed simply from the silence of the accused after warnings are given or simply from the fact that a confession was in fact eventually obtained.”² Subsequent cases indicated that determining whether a suspect has waived his Miranda rights is a fact-specific inquiry not easily susceptible to per se rules. According to these cases, resolution of the issue of waiver “must be determined on ‘the particular facts and circumstances surrounding that case, including the background, experience, and conduct of the accused.’”³ Under this line of cases, a waiver need not always be express, nor does Miranda impose a formalistic waiver procedure.⁴

In Berghuis v. Thompkins, citing the societal benefit of requiring an accused to invoke Miranda rights unambiguously, the Court refocused its Miranda waiver analysis on whether a suspect understood his rights.⁵ There, a suspect refused to sign a waiver form, remained largely silent during the ensuing 2-hour and 45-minute interrogation, but then made an incriminating statement. The five-Justice majority found that the suspect had failed to invoke his right to remain silent and had also implicitly waived the right. According to the Court, although a statement following silence alone may be inadequate to waive Miranda rights, the prosecution may show an implied waiver by demonstrating that a suspect understood the Miranda warnings given to him and subsequently made an uncoerced statement.⁶ Furthermore, once a suspect has knowingly and voluntarily waived his Miranda rights, police officers may continue questioning until and unless the suspect clearly invokes his rights later.⁷

The admissions of an unwarned or improperly warned suspect may not be used directly against him at trial, but the Court has permitted some use for other purposes, such as impeachment. Prosecutors cannot introduce a defendant’s confession or other incriminating admissions obtained in violation of Miranda against him at trial to establish guilt⁸ or determine the sentence, at least in bifurcated trials in capital cases.⁹ On the other hand, the “fruits” of such an unwarned confession or admission may be used in some circumstances if the statement was voluntary.¹⁰

The Court, in opinions that more narrowly construe Miranda, has broadened the permissible impeachment purposes for which unlawful confessions and admissions may be used.¹¹ Thus, in Harris v. New York,¹² the Court held that the prosecution could use statements, obtained in violation of Miranda, to impeach the defendant’s testimony if he voluntarily took the stand and denied commission of the offense. Subsequently, in Oregon v. Hass,¹³ the Court permitted prosecutors to impeach the defendant using a statement the defendant made after police had ignored his request for counsel following his Miranda warning. Such impeachment material, however, must still meet the standard of voluntariness associated with the pre-Miranda tests for the admission of confessions and statements.¹⁴

The Court has created a “public safety” exception to the Miranda warning requirement for serious offenses. In New York v. Quarles,¹⁵ the Court held admissible a recently apprehended suspect’s response in a public supermarket to the arresting officer’s demand to know the location of a gun that the officer had reason to believe the suspect had just discarded or hidden in the supermarket. The Court, in an opinion by Justice William Rehnquist,¹⁶ declined to place officers in the “untenable position” of having to make instant decisions as to whether to proceed with Miranda warnings and thereby increase the risk to themselves or to the public or whether to dispense with the warnings and run the risk that resulting evidence will be excluded at trial. While acknowledging that the exception itself would “lessen the desirable clarity of the rule,” the Court predicted that confusion would be slight: “[w]e think that police officers can and will distinguish almost instinctively between questions necessary to secure their own safety or the safety of the public and questions designed solely to elicit testimonial evidence from a suspect.”¹⁷ However, no such compelling justification was offered for a Miranda exception for lesser offenses, and protecting the rule’s “simplicity and clarity” counseled against creating one.¹⁸ The Court stated: “[A] person subjected to custodial interrogation is entitled to the benefit of the procedural safeguards enunciated in Miranda, regardless of the nature or severity of the offense of which he is suspected or for which he was arrested.”¹⁹