View PDF Version

No. 99-5525

In the Supreme Court of the United States

CHARLES THOMAS DICKERSON, PETITIONER

v.

UNITED STATES OF AMERICA

ON WRIT OF CERTIORARI
TO THE UNITED STATES COURT OF APPEALS
FOR THE FOURTH CIRCUIT

REPLY BRIEF FOR THE UNITED STATES

JANET RENO
Attorney General
SETH P. WAXMAN
Solicitor General
Counsel of Record
Department of Justice
Washington, D.C. 20530-0001
(202) 514-2217



The centerpiece of the argument in support of the judgment is the assertion that the Miranda doctrine is non-constitutional in character, so that Congress was free to supersede this Court's holdings with a statute that restores the due process "voluntariness" inquiry. Brief of Court-Appointed Amicus Curiae Paul G. Cassell 4-28. As we show below, that proposition is untenable. In the alternative, amicus argues (Br. 28-48) that, despite Congress's manifest intention to turn the clock back and restore the voluntariness test in 18 U.S.C. 3501, that statute and a variety of unrelated provisions do not return the law to its pre-Miranda state, but in fact provide adequate additional safeguards for Fifth Amendment rights; and, if that is not so, Miranda should be reconsidered and rejected.

None of those propositions is sound. This Court's consistent application of Miranda to the States over the past 34 years establishes its constitutional character, such that it is not subject to plenary revision and overruling by Congress. The purpose and effect of Section 3501 is not to provide substitutes for the Miranda safeguards, but to reject the need for any such safeguards. And the burden required to justify the overruling of Miranda, and to return litigation over confessions wholly to the voluntariness test that this Court found inadequate 34 years ago, is a heavy one, and it has not been met in this case.

Miranda rests on the proposition that, in the distinctive context of custodial interrogation, systemically adequate safeguards are required to protect a suspect's rights under the Self-Incrimination Clause. Miranda warnings are not the only permissible safeguards, this Court has made clear, but some safeguards must exist. Section 3501 rejects that premise. That premise, however, has long been a central principle of the Court's constitutional criminal jurisprudence, and we do not find justification to urge this Court to overrule its precedents that have stood for 34 years. As Chief Justice Burger explained in 1980, "[t]he meaning of Miranda has become reasonably clear and law enforcement practices have adjusted to its strictures; I would neither overrule Miranda, disparage it, nor extend it at this late date." Rhode Island v. Innis, 446 U.S. 291, 304 (Burger, C.J., concurring).

I. CONGRESS MAY NOT "OVERRULE" MIRANDA BY REINSTATING A PURE VOLUNTARINESS TEST

Amicus contends (Br. 4-28) that this Court need not formally overrule Miranda in order to uphold Section 3501 because Miranda is defeasible by Congress at will. That contention cannot be reconciled with this Court's practices in applying Miranda or with bedrock principles of criminal procedure.

A. Miranda And The States

Amicus seeks to reconcile this Court's consistent application of Miranda to the States with his view that Miranda is a "nonconstitutional measure[]" (Br. 17) by positing that this Court has power, in the absence of contrary legislation, to impose rules on the States that are not constitutionally required. Amicus's theory is fundamentally inconsistent with this Court's holdings on its role in reviewing state convictions.

1. Review of state convictions. Because this Court's authority in state cases "is limited to enforcing the commands of the United States Constitution," Mu'Min v. Virginia, 500 U.S. 415, 422 (1991), the Court could not have applied Miranda to the States during the past 34 years without concluding that the Constitution required such application.

Amicus takes issue (Br. 18 n.12) with this Court's statement in Mu'Min. He asserts instead (Br. 17) that, in the absence of contrary legislation, "the Court has some authority to impose on the states nonconstitutional measures designed to protect constitutional rights," and that this premise is "at the heart of [amicus's] explanation of the legal basis of Miranda." In amicus's view, because Miranda's exclusionary rule "is not a requirement of the Fifth Amendment" (Br. 10), it follows that Congress has the authority to "modify" it (Br. 12). By "modify," amicus apparently means "eliminate": in his view, Congress may abrogate Miranda's holding and mandate that voluntariness alone is the only prerequisite for admissibility. Nothing in this Court's cases supports that view of Congress's power over rules that this Court has applied to the States.

In order to uphold Section 3501 on the theory that Miranda is a "nonconstitutional measure" that is defeasible at the will of Congress, this Court would have to repudiate core principles governing its review of state criminal cases. This Court has frequently been guided by the proposition that it "do[es] not establish procedural rules for the States, except when mandated by the Constitution." Barker v. Wingo, 407 U.S. 514, 523 (1972); accord Victor v. Nebraska, 511 U.S. 1, 17 (1994) ("[W]e have no supervisory power over the state courts.").1 The principle that the Court may not reverse state court decisions except for violations of the Constitution or a federal statute is one that Members of the Court have recognized to be "obvious." See Carter v. Kentucky, 450 U.S. 288, 307-308 (1981) (Rehnquist, J., dissenting) ("[S]ince the result of the Court's decision is to reverse the judgment of the Supreme Court of Kentucky, the decision must obviously rest upon the fact that the decision of that court is inconsistent with the United States Constitution."). To accept amicus's theory of Miranda's application to the States would not only contradict that principle; it would vastly enlarge this Court's power to reverse state convictions based on rules that are not constitutionally compelled.

2. Application on habeas corpus. This Court's recent application of Miranda on habeas corpus also conflicts with amicus's view that Miranda is common law rather than constitutional doctrine. In Withrow v. Williams, 507 U.S. 680, 690 (1993), the Court fully acknowledged Miranda's "prophylactic" character, yet concluded that "in protecting a defendant's Fifth Amendment privilege against self-incrimination, Miranda safeguards a fundamental trial right." Id. at 691 (internal quotation marks omitted); see id. at 689 (describing Miranda's "now-familiar measures in aid of a defendant's Fifth Amendment privilege"); id. at 692 (discussing "the Fifth Amendment 'trial right' protected by Miranda"). The Court nowhere suggested that it regarded Miranda as one of the "laws of the United States" that may be enforceable on habeas corpus. 28 U.S.C. 2254(a).2

B. Miranda And The Constitution

Amicus offers his radical theory to explain this Court's application of Miranda to the States because he concludes that this Court's later cases have stripped Miranda of a constitutional basis. See Br. 28 (Miranda is an "extraconstitutional" rule). Relying on this Court's statements that the Miranda rules are "not themselves rights protected by the Constitution," New York v. Quarles, 467 U.S. 649, 654 (1984), and that they "sweep[] more broadly than the Fifth Amendment itself," Oregon v. Elstad, 470 U.S. 298, 306 (1985), amicus contends (Br. 4-9) that the Miranda doctrine cannot be constitutionally based, and, accordingly, Congress is free to require the admission of unwarned confessions without providing any substitute protective framework. Amicus misunderstands the source and nature of Miranda.3

Miranda responded to this Court's conclusion-after 30 years of grappling with the due process voluntariness test on a case-by-case basis-that additional protections, above and beyond the totality-of-the-circumstances test, are required to provide safeguards in view of the coercive pressures of custodial interrogation.4 The Court thus crafted a prophylactic framework to provide systemic protection for rights guaranteed by the Self-Incrimination Clause. See Michigan v. Tucker, 417 U.S. 433, 443-444 (1974). As this Court has since explained:

Both waiver of rights and admission of guilt are consistent with the affirmation of individual responsibility that is a principle of the criminal justice system. It does not detract from this principle, however, to insist that neither admissions nor waivers are effective unless there are both particular and systemic assurances that the coercive pressures of custody were not the inducing cause.

Minnick v. Mississippi, 498 U.S. 146, 155 (1990) (emphasis added). Far from concluding that developments since Miranda have deprived that case of constitutional support, the Court reaffirmed, nearly two decades later, that Miranda "strikes the proper balance between society's legitimate law enforcement interests and the protection of the defendant's Fifth Amendment rights." Moran v. Burbine, 475 U.S. 412, 424 (1986); see also Elstad, 470 U.S. at 317 (the Court "in no way retreats from the bright-line rule of Miranda").

The Court's recognition that Miranda warnings are not, in all circumstances, required to satisfy the Constitution (see Amicus Br. 7-8) in no way undermines the proposition that the Miranda framework rests on this Court's authority to interpret and apply the Constitution. Indeed, it is precisely because the Miranda cases create a prophylactic framework that this Court may shape the governing legal rules, applying the Miranda doctrine where its purposes are best served and declining to do so where its purposes would be less well-served or where it would impose special or unusual costs. This Court has done so by, among other things, permitting the use of unwarned statements for impeachment, see Harris v. New York, 401 U.S. 222, 224 (1971); Oregon v. Hass, 420 U.S. 714, 722 (1975), and dispensing with the warnings where public safety so requires, Quarles, 467 U.S. at 651. Sustaining Miranda, however, would hardly require overruling Quarles, Harris, and Hass, among other cases. See Amicus Br. 8-9 & nn.4-5. This Court's Miranda cases have held that the Constitution generally requires a set of procedural safeguards that serve systemic values, but the Court has also recognized that such safeguards need not be imposed inflexibly and without regard to competing concerns.5

Nor is Miranda invalidated by the Court's conclusion (384 U.S. at 467) that either the Miranda rules or some equally adequate alternative is necessary to protect the Fifth Amendment privilege. See Amicus Br. 6-7, 11-12. In that respect, Miranda is supported, rather than undercut, by Smith v. Robbins, 120 S. Ct. 746 (2000). Robbins explained that the procedure outlined in Anders v. California, 386 U.S. 738 (1967), for the withdrawal of court-appointed appellate counsel who views his client's claims as frivolous was not "an independent constitutional command" but instead constituted "'a prophylactic framework' that we established to vindicate the constitutional right to appellate counsel." 120 S. Ct. at 757. The decision in Anders, the Court noted in Robbins, "simply erect[ed] 'safeguards.'" Ibid.6 Robbins thus held that, when counsel for an indigent defendant concludes that the appeal is frivolous and seeks to withdraw, States must have "procedures [that] adequately safeguard a defendant's right to appellate counsel." Id. at 753; see also id. at 760. Yet the Court also made clear that no single set of procedures is required-the test is whether a particular procedure "reasonably ensures that an indigent's appeal will be resolved in a way that is related to the merit of that appeal." Id. at 759.7 Finally, Robbins held that, even where a State provides adequate procedures for assuring that appointed counsel will not improperly withdraw, a defendant may still claim that he was deprived of the underlying constitutional right to counsel, under the test of Strickland v. Washington, 466 U.S. 668 (1984). 120 S. Ct. at 763-764.8

In all of those respects, Robbins's constitutional analysis runs parallel to Miranda: in both settings, a prophylactic set of procedures is required to protect rights; a legislature may experiment with alternative procedures, subject to this Court's review for adequacy; but a legislature may not simply eliminate all such procedures in favor of correcting actual violations on a case-by-case basis. Whether or not rules such as Miranda and Anders are properly described as "constitutional common law" (Amicus Br. 10 n.7), they "cannot be overturned by mere congressional disapproval. * * * Congress may override preemptive lawmaking based on the Constitution, but only if the federal courts independently conclude that Congress has enacted a statute that provides roughly the same degree of protection for constitutional policies as the federal common law rule." Thomas W. Merrill, The Common Law Powers of Federal Courts, 52 U. Chi. L. Rev. 1, 57-58 (1985).

C. A Plenary Congressional Power To Overrule Miranda Has No Case Law Support

Amicus seeks support for his theory of congressional power to modify or overrule this Court's constitutional holdings in several other branches of this Court's jurisprudence. None of those lines of cases assists him.

1. Bivens. In Bivens v. Six Unknown Agents, 403 U.S. 388 (1971), this Court, exercising the traditional authority of courts "to adjust their remedies so as to grant the necessary relief," id. at 392, recognized a cause of action for damages against federal officials for the violation of Fourth Amendment rights. The Court drew on the judicial power, authorized in the grant of general federal question jurisdiction in 28 U.S.C. 1331: "In the absence of * * * a congressional directive, the federal courts must make the kind of remedial determination that is appropriate for a common-law tribunal." Bush v. Lucas, 462 U.S. 367, 378 (1983).

The Court has held that Congress may displace Bivens actions, at least when it provides "constitutionally adequate" alternatives. Bush, 462 U.S. at 378 n.14. Assuming that Congress may completely eliminate a Bivens remedy in a particular context (a question the Court has reserved), Bivens does not furnish an analogy to Miranda. Bivens was not based on a constitutional requirement deriving from the Fourth Amendment, but on the Court's traditional authority under Section 1331 to imply a cause of action for damages in federal court to remedy a completed violation of law. Congressional power over the implication of such federal causes of action for damages has always been recognized as plenary. By contrast, Miranda and the cases following it are based on the proposition that the Self-Incrimination Clause requires safeguards to prevent violations in the distinctive context of custodial interrogation. Plenary congressional authority over whether the Constitution requires safeguards would be inconsistent with this Court's cases from Marbury v. Madison, 5 U.S. (1 Cranch) 137 (1803), to City of Boerne v. Flores, 521 U.S. 507 (1997).9

2. Act of State and Commerce Clause Cases. Amicus asserts that in cases applying the Act of State doctrine, the Court has recognized its authority to a craft a rule applicable to the States that is "subject to revision and restriction by Congress as a judicially developed, nonconstitutional rule, even while the Court's authority to craft it in the first instance is not seriously in doubt." Br. 18; see also Br. 18 n.11. The Act of State doctrine is not comparable to Miranda. It recognizes limitations on judicial inquiry into foreign governmental action. The authority of federal courts to adopt such a rule of decision and apply it to the States arises from constitutional grants of authority that "reflect[] a concern for uniformity in this country's dealings with foreign nations and indicat[e] a desire to give matters of international significance to the jurisdiction of federal institutions." See Banco Nacional de Cuba v. Sabbatino, 376 U.S. 398, 427 n.25 (1964). Because the doctrine protects powers reserved to the federal government, it necessarily binds the States. And because the ultimate constitutional responsibility for foreign affairs resides in the Executive and Legislative Branches, those branches have substantial latitude to modify the otherwise applicable judicial rule. Cf. First Nat'l City Bank v. Banco Nacional de Cuba, 406 U.S. 759, 767-768 (1972) (plurality opinion) (addressing authority of Executive Branch to supersede Act of State doctrine). By contrast, Miranda does not rely on or protect constitutional grants of authority to the political branches of the federal government. Therefore, those branches have no authority to abrogate this Court's decisions in Miranda and its progeny.

The dormant Commerce Clause cases cited by amicus (Br. 18 n.11) are irrelevant for a similar reason: that doctrine derives from and protects the plenary power of Congress to regulate commerce. See South-Central Timber Dev., Inc. v. Wunnicke, 467 U.S. 82, 87-88 (1984) (dormant Commerce Clause is "a self-executing limitation on the power of the States," but it is "clear that Congress may redefine the distribution of power over interstate commerce by permit[ting] the states to regulate the commerce in a manner which would otherwise not be permissible") (Opinion of the Court; internal quotation marks omitted).

II. SECTION 3501 IMPERMISSIBLY RETURNS THE LAW TO ITS PRE-MIRANDA STATE

Amicus argues (Br. 28) that, even if the Court may require some "prophylaxis" beyond the voluntariness standard, Section 3501, "taken together with the legal landscape that surrounds it, provides more than adequate protection to safeguard suspects from police compulsion." Section 3501, however, was intended to-and does-simply return the law to its pre-Miranda state. See U.S. Br. 13-20; J.A. 197. Moreover, there are no changes in the surrounding "legal landscape" that compensate for Section 3501's abrogation of Miranda's core holdings.

1. Section 3501. Section 3501 requires a court to admit a confession "if it is voluntarily given," 18 U.S.C. 3501(a), and it directs the court to "take into consideration all the circumstances surrounding the giving of the confession" in determining voluntariness, 18 U.S.C. 3501(b). Section 3501 thereby codifies pre-Miranda law, which in virtually identical terms required that "all the circumstances attendant upon the confession must be taken into account" in determining its voluntariness. Reck v. Pate, 367 U.S. 433, 440 (1961). Amicus does point out (Br. 33) that Section 3501 requires consideration of the defendant's knowledge of "the nature of the offense with which he was charged or * * * suspected," a factor that the Court subsequently concluded is not relevant to the voluntary waiver of Fifth Amendment rights. Colorado v. Spring, 479 U.S. 564, 577 (1987). Section 3501, however, does not require courts to give specific weight to that (or any other) factor. And to the extent that the defendant's awareness of the charges might be relevant under Section 3501, although irrelevant under the Fifth Amendment, it would not serve any purpose in protecting constitutional values. See U.S. Br. 18 n.13.

Amicus also argues (Br. 32) that "[t]he incentives to [give Miranda warnings] that Section 3501 provides are much stronger than those in pre-Miranda law." He observes (Br. 32) that Section 3501 requires that a court, in determining voluntariness, "shall take into consideration all the circumstances surrounding the giving of the confession, including" whether a suspect "was advised or knew" of his right to remain silent and whether he "had been advised * * * of his right to the assistance of counsel," 18 U.S.C. 3501(b). Before Miranda, however, this Court had very frequently adverted to those same factors as important elements in the totality-of-the-circumstances voluntariness test. See U.S. Br. 15-16 nn.10-11 (citing cases). Amicus does not explain how a direction to courts that they "shall" consider all the circumstances, including two particular factors, differs in any significant way from pre-existing law, which required that courts "must" consider all the circumstances, Reck, 367 U.S. at 440 including (under this Court's cases) those same two factors.

2. Other remedies. Amicus also contends (Br. 34) that, apart from Section 3501, "the legal incentives for non-coercive police questioning today are almost unrecognizably greater than when Miranda was decided," and that those incentives, together with Section 3501, create a "constitutionally adequate alternative to the Miranda rules," Br. 37. Amicus fails, however, to substantiate that contention.

a. Amicus argues that Bivens actions are now available against federal law enforcement authorities, although "[w]hen Miranda was written, it was quite difficult as a practical matter to obtain damages in federal court from federal law enforcement officers who violated Fifth Amendment rights." Br. 34. Three of the four consolidated cases decided in this Court's Miranda decision were state cases. Damages against state law enforcement officials for violations of constitutional rights had been recognized since this Court's decision in Monroe v. Pape, 365 U.S. 167 (1961), five years before Miranda. That remedy did not affect the result in Miranda or this Court's later cases. Amicus offers no reason to assign greater weight to that remedy now.

In any event, Bivens and Section 1983 actions are of little utility in protecting against "the compulsion inherent in custodial surroundings," 384 U.S. at 458, which is the focus of the Miranda rules. While the use of force to obtain a confession may be deterred by the prospect of a damages award under Bivens or Section 1983, Miranda does not address the use of violence. The subtler coercive pressures addressed by Miranda are far less likely to result in damages. Damages would also be rare because, when a confession is excluded from evidence, there would be no Fifth Amendment violation, see United States v. Verdugo-Urquidez, 494 U.S. 259, 264 (1990); and when a confession is admitted and the conviction upheld, there would be no sustainable claim.10

b. Amicus's reliance (Br. 34-35) on the Federal Tort Claims Act, 28 U.S.C. 2680(h), is also misplaced, since psychological coercion in interrogation would not, standing alone, constitute an "assault, battery, false imprisonment, false arrest, abuse of process, or malicious prosecution." Ibid.11 In short, the remedies noted by amicus cannot make up for Section 3501's overt return to the pre-Miranda law governing the voluntariness of a confession.

III. THE SHOWING REQUIRED TO OVERRULE MIRANDA HAS NOT BEEN MADE

Our opening brief submitted that, applying settled principles of stare decisis, there is insufficient justification for overruling Miranda. The amici in support of the judgment largely avoid the term "overrule"; they instead request this Court to "modify" Miranda by "abandon[ing] the irrebuttable presumption that confessions obtained without compliance with the Miranda procedures are always involuntary." Cassell Br. 40. Because the government already bears the burden of establishing the voluntariness of a confession by a preponderance of the evidence, see Lego v. Twomey, 404 U.S. 477, 489 (1972), adopting a rebuttable presumption is equivalent to overruling Miranda outright. That course is not warranted.

A. The Proportionality Of Miranda

Amicus argues (Br. 40-48) that Miranda's irrebuttable presumption must be abandoned because it is not "congruent and proportional," see Boerne, 521 U.S. at 508, to an underlying constitutional right and is therefore beyond this Court's authority. That conclusion is mistaken.

The underlying point of the "congruence and proportionality" analysis, applied in Boerne in analyzing Congress's power under Section 5 of the Fourteenth Amendment, is to inquire whether remedial measures devised by Congress permissibly protect an underlying constitutional right or instead impermissibly redefine the right. See Kimel v. Florida Bd. of Regents, 120 S. Ct. 631, 644-645 (2000). This Court has never said that such an analysis applies to the Court's own formulation of prophylactic constitutional standards. But, even if it were to do so now, the Court's prior holding that either Miranda or some adequate alternative is necessary to effectuate the Self-Incrimination Clause would satisfy that analysis.

Miranda itself extensively analyzed the nature of custodial interrogation, the utility of warnings in promoting a voluntary decision by the suspect, and the impact on law enforcement of a warnings-and-waiver rule before concluding that such a rule (or an equally effective legislative safeguard) was necessary to protect the privilege. 384 U.S. at 445-491. In later cases, the Court has firmly adhered to the view that "the Court in Miranda was impelled to adopt" the irrebuttable presumption, Garner v. United States, 424 U.S. 648, 657 (1976), and that "protection of the privilege against self-incrimination during pretrial questioning requires application of special 'procedural safeguards,'" Pennsylvania v. Muniz, 496 U.S. 582, 589 (1990). In Moran v. Burbine, 475 U.S. at 424, 426-427, the Court reviewed the competing concerns implicated by custodial interrogation and reaffirmed Miranda as a "carefully drawn approach."

Amicus suggests (Br. 41-42) that under the "congruence and proportionality" test, a prophylactic rule is justifiable only if it operates where the underlying constitutional norm has been-or likely has been-violated. Amicus is incorrect. This Court has frequently recognized prophylactic rules, but it has never suggested that they must satisfy that standard. See U.S. Br. 44-47. For example, the Court in Anders and Robbins did not justify the prophylactic rule at issue there (see pp. 7-8, supra) by stating that, as an empirical matter, most instances in which appointed appellate counsel unilaterally withdraw involve appeals of arguable merit rather than frivolous appeals. A sensible approach to this Court's prophylactic rules instead considers the importance of the right, the efficacy of competing approaches, and the costs and benefits of a safeguard.12

B. The Costs And Benefits Of Miranda

The amici in support of the judgment believe that the costs of Miranda's exclusionary rule outweigh any of its benefits. E.g., Cassell Br. 46. We do not minimize the costs of excluding probative evidence, either in general or in any particular case. The question whether those costs are justified, however, involves a weighing of competing interests and an assessment of the actual impact of the rule. See, e.g., United States v. Janis, 428 U.S. 433 (1976).

1. The Exclusionary Rule. Miranda serves at least three purposes. First, it provides procedures to dispel the inherent potential for compulsion in custodial interrogation. See Moran v. Burbine, 475 U.S. at 425, 426-427. Second, it provides a defined legal standard for the courts in place of the less determinate and more subjective voluntariness inquiry. See Tucker, 417 U.S. at 442-443. Third, it provides guidance for the police and prosecutors. Spring, 479 U.S. at 577 n.9. It achieves those goals through application of an irrebuttable presumption of compulsion only in the government's case-in-chief, and then only where not overborne by public safety interests. See pp. 4-6, supra.

This Court was fully aware in Miranda that its holding would preclude the use of some confessions. And the Court has since enforced the doctrine in cases in which the result has been to reverse convictions for serious crimes. See, e.g., Minnick v. Mississippi, supra; Arizona v. Roberson, 486 U.S. 675 (1988); Edwards v. Arizona, 451 U.S. 477 (1981). Accordingly, the observation that Miranda violations will foreclose the use of some evidence cannot, standing alone, justify overruling that decision.

While the exclusionary rule does impose certain costs, it also enables this Court to define and enforce legal requirements. By doing so, the Court facilitates the admission of the great number of confessions in which the police comply with Miranda. In the absence of such a restraining legal rule, there may be temptation to interrogate suspects without providing Miranda warnings, in the hope that a court could subsequently be persuaded that a resulting confession was voluntary. If the courts disagree, prosecutions may be seriously compromised. Miranda thus helps to ensure that statements made by suspects in custodial interrogation will ultimately be found admissible at trial.13 We therefore agree with an American Bar Association committee that concluded that "[a]lthough the Miranda decision has sparked heated controversy on a political level, the restrictions it imposes are not considered troublesome by either police or prosecutors." ABA Special Comm. on Crim. Justice in a Free Society, Criminal Justice in Crisis 27 (1988). See also ibid. ("The Committee finds that Miranda does not have a significant impact on law enforcement's ability to solve crime or to prosecute criminals successfully.").14

2. Federal Prosecutions. With respect to the costs of the Miranda rule, our submission is informed by experience in federal prosecutions. Federal courts rarely order the suppression of statements under Miranda. Between 1989 and 1999, approximately 720,000 federal prosecutions were brought; during that period, according to the Justice Department's records, federal courts suppressed approximately 78 statements under Miranda-i.e., one out of every 9,300 federal prosecutions.15 Suppression can impose a cost in a particular case that numbers cannot adequately convey. Nonetheless, the infrequency of suppression under Miranda in federal prosecutions weighs against overruling that decision.16

Amicus Cassell (Br. 24) states that, "contrary to the impression conveyed in the government's brief [at 34], the actual views of federal law enforcement agencies that have been lodged with the Court reveal serious difficulties with Miranda's exclusionary rule." That is incorrect. In general, the letters we have lodged with the Court conclude that the core mandates of Miranda do not hinder law enforcement efforts in any significant way, although certain extensions of Miranda have caused some difficulties. U.S. Br. 35. For example, the FBI General Counsel stated that "[t]he FBI has very little difficulty complying with the relatively simple mandates of the Miranda decision," although "[p]roblems do occur * * * when those dictates are complicated by the additional protections afforded custodial subjects by [Edwards, Roberson, and Minnick]." The Treasury Department law enforcement agencies reported that they "favor the current legal framework and do not find that the issuance of Miranda warnings hinders their investigations," "would not support any modification to current practices," and "would [not] support replacing Miranda with the §3501 voluntariness standard."17 And the DEA reported that it is "unable to cite any incidents in which the requirement to provide Miranda warnings to a suspect adversely impacted on a case." Chief Counsel Memorandum (Feb. 22, 2000).18

Amicus Cassell asserts (Br. 25) that the Justice Department "has for many years supported the constitutionality of Section 3501." Amicus relies (Br. 25-26) principally on recommendations to the Attorney General (which did not result in formal policies) and recollections of certain former Justice Department officials in after-the-fact congressional testimony. Those sources do not express official Justice Department policy. The more telling and indisputable fact is that, "with limited exceptions [Section 3501] has been studiously avoided by every Administration, not only in this Court but in the lower courts, since its enactment more than 25 years ago." Davis v. United States, 512 U.S. 452, 463-464 (1994) (Scalia, J., concurring). That position implies both a recognition of Section 3501's conflict with Miranda and the absence of sufficient reason to request this Court to overrule it.

3. The alternatives to Miranda. Miranda stated that "Congress and the States are free to develop their own safeguards for the privilege, so long as they are fully as effective as those described [in Miranda] in informing accused persons of their right of silence and in affording a continuous opportunity to exercise it." 384 U.S. at 490. The precise standard of constitutional adequacy that an alternative safeguard must satisfy is not before the Court in this case. If Congress or a state legislature were to provide for a set of limited warnings coupled with video or audio taping of the interrogation, or for some other alternative set of safeguards, such as pre-interrogation access to a magistrate, counsel, or other competent advisor, the Court would be faced with the question whether the Miranda rules remained applicable in the jurisdiction at issue. Section 3501, however, as its supporters intended, see U.S. Br. 18-20, seeks to return the law to its pre-Miranda state without providing any alternative. That approach would require overruling Miranda.

In considering the sea change in the law that is being proposed, it is worth recalling that the Miranda Court arrived at its solution only after concluding that the "totality of the circumstances" voluntariness test, as the sole protection for the Fifth Amendment rights of a custodial suspect, had failed. The "totality" test had been found inadequate not because of bad faith by police, or because of unwillingness of courts to grapple with the inquiry. It was inadequate because a "totality" test, without more, provided insufficient guidance to the police, left inadequate means for this Court to unify and expound the law, and resulted in an uncertain legal rule that could not secure the vital constitutional rights at stake. There has been no showing that a return to that regime would be successful today.

Respectfully submitted.

JANET RENO
Attorney General

SETH P. WAXMAN
Solicitor General

APRIL 2000

1 See also Smith v. Phillips, 455 U.S. 209, 221 (1982); Harris v. Rivera, 454 U.S. 339, 344-345 (1981) (per curiam)); Doyle v. Ohio, 426 U.S. 610, 617 n.8 (1976); Cicenia v. Lagay, 357 U.S. 504, 508-509 (1958); Gallegos v. Nebraska, 342 U.S. 55, 63-64 (1951) (plurality); Townsend v. Burke, 334 U.S. 736, 738 (1948); McNabb v. United States, 318 U.S. 332, 340 (1943).

2 As authority for the view that application of Miranda on habeas implements the "laws of the United States," amicus cites (Br. 21-22), a treatise by Professor Yackle. In fact, however, Professor Yackle principally maintains that Miranda violations are cognizable on federal habeas, because to hold otherwise would be to "eliminate" the "substantive [constitutional] right[], not merely [a] judge-made remed[y], * * * from the scope of habeas corpus." L. Yackle, Post Conviction Remedies § 96, at 370 (1981).

3 Amicus's defense of Section 3501 would also leave intact Miranda's application to the States. A theory that imposes Miranda on the States but not on the federal government has no constitutional coherence. Perhaps amicus would posit that States too could opt out of Miranda by enacting the voluntariness test, but that theory violates the Supremacy Clause: a State may not nullify federal law.

4 Commentators recognize the inability of the "totality" test, standing alone, to secure the constitutional rights at stake. See, e.g., 2 W. LaFave & J. Israel, Criminal Procedure § 6.2(d), at 467 (1999) (noting the Court's failure to "articulate a clear and predictable definition of voluntariness"); S. Saltzburg & D. Capra, American Criminal Procedure: Cases and Commentary 514 (5th ed. 1996) ("The word 'voluntary' hardly offered clear guidance to law enforcement officers and to lower court judges. It had to be defined anew in every case. * * * In fact, each [of the Supreme Court's] totality of the circumstances decision[s], it might be argued, caused a greater division among lower trial and appellate courts."); see also U.S. Br. 40-41 & n.32 (discussing pre-Miranda cases).

5 Amicus finds (Br. 6-7) this Court's statements that the Miranda safeguards are not themselves constitutional rights to be irreconcilable with a conclusion that Miranda has a "constitutional basis." In Elstad, however, the Court explained that "[w]hen police ask questions of a suspect in custody without administering the required warnings, Miranda dictates that the answers received be presumed compelled and that they be excluded from evidence at trial in the State's case in chief." 470 U.S. at 317. That presumption explains how unwarned statements can be excluded from evidence consistent with the Fifth Amendment's text that a defendant may not be "compelled * * * to be a witness against himself." The Miranda conclusive presumption is a legal rule, but that does not distinguish it from voluntariness determinations. The "totality of the circumstances" test also resolves a question of law about the permissible "techniques for extracting the statements"; it is not a question of historical fact. Miller v. Fenton, 474 U.S. 104, 116 (1985).

6 This Court's description of Miranda is to the same effect: Miranda provided a set of "procedural safeguards" that are "prophylactic" in character. Michigan v. Tucker, 417 U.S. at 444, 446.

7 Miranda similarly contemplated legislative alternatives: "[W]e cannot say that the Constitution necessarily requires adherence to any particular solution for the inherent compulsions of the interrogation process. * * * We encourage Congress and the States to continue their laudable search for increasingly effective ways of protecting the rights of the individual while promoting efficient enforcement of our criminal laws." 384 U.S. at 467.

8 Miranda has the same feature: even when the warnings have been given, the suspect may still raise a claim that subsequent statements were involuntary under the totality-of-the-circumstances test. See Withrow v. Williams, 507 U.S. at 693-694. In practice, however, once a defendant waives his rights after adequate warnings, the voluntariness claim is virtually sure to fail. See U.S. Br. 36-38.

9 Amicus also cites Mapp v. Ohio, 367 U.S. 643 (1961), to exemplify his view that this Court has "authority to improvise measures to assist in the protection of constitutional rights where neither the Constitution nor the legislature has specified a particular mechanism for protecting those rights." Br. 10. He does not expressly assert that Congress could, without providing an adequate alternative, simply eliminate the exclusionary rule. But, because the exclusionary rule is not a "personal constitutional right" of the defendant, United States v. Leon, 468 U.S. 897, 906 (1984), amicus's theory about Miranda necessarily implies that result. In effect, amicus would have this Court cede to Congress plenary authority to define the measures that are required to protect constitutional rights.

10 18 U.S.C. 241 and 242, also cited by amicus (Br. 35), are likewise unhelpful. Those remedies were available before Miranda, see Williams v. United States, 341 U.S. 97 (1951), and provide minimal protection to the Fifth Amendment privilege in the settings relevant in Miranda cases.

11 Amicus also cites (Br. 35 & n.25) 28 U.S.C. 530B, which generally provides that government attorneys are subject to state ethics rules. Section 530B could not influence questioning in the vast majority of custodial interrogations, which are conducted by law enforcement agents without the involvement or direction of government attorneys.

12 The ultimate conclusion on the need for and efficacy of a prophylactic rule to effectuate a constitutional right is not a question of legislative policy or factfinding in this setting any more than it would be in other constitutional contexts. Amicus relies (Br. 26-27) on the 1968 Senate committee report on Section 3501 concluding that Miranda harms law enforcement. Although congressional factfinding is entitled to deference, Congress is not free simply to substitute its view for this Court's judgment regarding the ultimate constitutional balance to be drawn. See Boerne, 521 U.S. at 519; see also H. Monaghan, Foreword: Constitutional Common Law, 89 Harv. L. Rev. 1, 34 n.176, 42 n.217 (1975); Y. Kamisar, Can (Did) Congress Overrule Miranda?, 85 Cornell L. Rev. 883, 916-929 (2000) (forthcoming).

13 Some amici have pointed out that Miranda itself does not always provide a "bright line" test. Certainly, Miranda requires application of judgment. But this Court has regularly recognized that Miranda provides far better guidance to police and courts than a multi-factor "totality" test, see U.S. Br. 34 n.24, and experience has not suggested otherwise.

14 We acknowledge that some members of the law enforcement community participating in this case contend that Miranda is harmful to law enforcement. With respect, we do not agree that the claimed harms are sufficient to justify overruling of Miranda. Other amici in the law enforcement community have filed briefs in agreement with our view.

15 The figures are reproduced and explained in the Appendix, infra.

16 Amicus has acknowledged that only a "tiny fraction of cases * * * go forward but are later lost because of a Miranda suppression motion." P. Cassell, Miranda's Social Costs: An Empirical Reassessment, 90 Nw. U. L. Rev. 387, 394 (1996). In his writings, he has argued instead that the "typical" cost of Miranda is the result of "diminishing the confession rate and thus reducing the evidentiary strength of the prosecution's case." Id. at 437. That claim is highly debatable. See U.S. Br. 32 n.23. But since amicus now urges (Br. 31, 32 n.23) that under Section 3501 "federal law enforcement officers will almost certainly continue to give [Miranda warnings]," replacing Miranda with Section 3501 would not increase the number of confessions, and it would therefore do nothing to alleviate what amicus has consistently claimed is Miranda's primary cost to law enforcement.

17 The Acting Chief of the Office of Tax Crimes stated that his office "finds no advantage by either modifying [its] current procedures concerning Miranda or adopting Title 18 U.S.C. §3501." The Commissioner of Customs stated that "I do not believe that the Miranda warnings have compromised Customs officers' ability to question suspects or obtain confessions admissible in court. The clear rule of Miranda provides firm guidance for our officers, and I strongly recommend that it not be changed." And the ATF, through its Assistant Director, reported that "[m]ost agents have grown accustomed to the Miranda process, and it generally does not disturb the progress of many investigations. * * * Section 3501 has the potential to cloud an area of the law that has become second nature to the investigator. Miranda has provided a good balance between the rights of the accused and the conduct of law enforcement."

18 Amicus relies (Br. 25) on an earlier letter from a DEA official that states that the giving of Miranda warnings in "a stressful environment" following an arrest "can have a chilling effect" on a suspect's willingness to cooperate. Memorandum from Richard Fiano, DEA Chief of Operations 2 (Oct. 13, 1999). The DEA has made clear, however, that "regardless of any change in the Miranda rule which may result in this case, DEA will continue, as a policy matter, to counsel its Agents to provide Miranda warnings." Chief Counsel Memorandum (Feb. 22, 2000); accord DEA Deputy Chief Counsel Gleason Memorandum (undated, sent Oct. 1, 1997).

APPENDIX