10 Chap. L. Rev. 579 (No PDF)
Chapman Law Review
Spring 2007
Articles
AGAINST ORTHODOXY: MIRANDA IS NOT PROPHYLACTIC AND THE CONSTITUTION IS NOT PERFECT
Lawrence Rosenthal [tippy title=”*” header=”off”]Associate Professor of Law, Chapman University School of Law. I am immensely grateful for the comments provided by Katherine Darmer, Steven Duke, Scott Howe, and Yale Kamisar. Any errors that remain are my own despite their best efforts to set me straight. Special thanks are due to Tom Bell and the Committee on Teaching, Evaluation and Scholarship at Chapman University School of Law for arranging a workshop presentation of this article at which I received many valuable suggestions. My thanks also to Andrew Bugman and Michael Riddell for highly capable research assistance.[/tippy]
Copyright (c) 2007 Chapman Law Review; Lawrence Rosenthal
Introduction
In the four decades since the decision in Miranda v. Arizona, [tippy title=”1″ header=”off”]1384 U.S. 436 (1966).[/tippy] two points of consensus have emerged about that decision. In the discussion that follows, I mean to take on both.
The first area of agreement is that Miranda’s rationale for requiring its now-famous warnings is wrong, or at least dramatically overstated. In Miranda, the Court, applying to police interrogation the Fifth Amendment’s admonition that “[n]o person shall . . . be compelled in any criminal case to be a witness against himself,” [tippy title=”2″ header=”off”]U.S. Const. amend. V.[/tippy] concluded that “without proper safeguards the process of in-custody interrogation of persons suspected or accused of crime contains inherently compelling pressures which work to undermine the individual’s will to resist and to compel him to speak where he would not otherwise do so freely.” [tippy title=”3″ header=”off”]384 U.S. at 467.[/tippy] That view of the inherently coercive nature of custodial interrogation, however, did not survive.
In Michigan v. Tucker, [tippy title=”4″ header=”off”]417 U.S. 433 (1974).[/tippy] the Court first labeled Miranda warnings as “prophylactic standards.” [tippy title=”5″ header=”off”]Id. at 446. For a largely accurate prediction of the implications of Tucker, see Geoffrey R. Stone, The Miranda Doctrine in the Burger Court, 1977 Sup. Ct. Rev. 99, 115-25.[/tippy] By this the Court meant that “Miranda warnings . . . are ‘not themselves rights protected by the Constitution but [are] instead measures to insure that the right against compulsory self-incrimination [is] protected.”’ [tippy title=”6″ header=”off”]New York v. Quarles, 467 U.S. 649, 654 (1984) (quoting Tucker, 417 U.S. at 444 (alterations in original)).[/tippy] Accordingly, “[t]he Miranda exclusionary rule . . . sweeps more broadly than the Fifth Amendment itself.” [tippy title=”7″ header=”off”]Oregon v. Elstad, 470 U.S. 298, 306 (1985).[/tippy] Thus, the Court has habitually characterized Miranda warnings as “prophylactic.” [tippy title=”8″ header=”off”]E.g., United States v. Patane, 542 U.S. 630, 638-41 (2004) (plurality opinion); Chavez v. Martinez, 538 U.S. 760, 770-73 (2003) (opinion of Thomas, J.); Davis v. United States, 512 U.S. 452, 457-58 (1994); Withrow v. Williams, 507 U.S. 680, 690-91 (1993); Duckworth v. Eagan, 492 U.S. 195, 203 (1989); Arizona v. Roberson, 486 U.S. 675, 680-81 (1988); Connecticut v. Barrett, 479 U.S. 523, 528 (1987); Elstad, 470 U.S. at 308; New York v. Quarles, 467 U.S. 649, 654 (1984); Edwards v. Arizona, 451 U.S. 477, 491-92 (1981) (Powell, J., concurring in result). There is no small debate over the meaning of the term “prophylactic” in this context. See, e.g., Yale Kamisar, Miranda Thirty-Five Years Later: A Close Look at the Majority and Dissenting Opinions in Dickerson, 33 Ariz. St. L.J. 387, 411 n.147 (2001). For present purposes, it will suffice to utilize that term in the sense that it has appeared in the Court’s post-Miranda decisions–a rule is “prophylactic” if it grants relief without a finding that a litigant’s own constitutional rights have been violated. See, e.g., Elstad, 470 U.S. at 306-07. Perhaps the best explication of this meaning of prophylaxis has been offered by Brian Landsberg: “I use the term ‘prophylactic rules’ to refer to those risk-avoidance rules that are not directly sanctioned or required by the Constitution, but that are adopted to ensure that the government follows constitutionally sanctioned or required rules.” Brian K. Landsberg, Safeguarding Constitutional Rights: The Uses and Limits of Prophylactic Rules, 66 Tenn. L. Rev. 925, 926 (1999).[/tippy] This is despite the fact that Miranda itself never characterized its holding as prophylactic, and within three years of Miranda, the Court expressly reaffirmed that decision and added that the use of “admissions obtained in the absence of the required warnings was a flat violation of the Self-Incrimination Clause of the Fifth Amendment as construed in Miranda.” [tippy title=”9″ header=”off”]See Orozco v. Texas, 394 U.S. 324, 326 (1969).[/tippy]
This view of Miranda has had dramatic results. While it is settled that under the Fifth Amendment, compelled statements may not be used for any purpose–including impeachment of the declarant’s subsequent testimony, [tippy title=”10″ header=”off”]See, e.g., Braswell v. United States, 487 U.S. 99, 117 (1988); New Jersey v. Portash, 440 U.S. 450, 458-59 (1979); Mincey v. Arizona, 437 U.S. 385, 401-02 (1978).[/tippy] or as a source of investigative leads [tippy title=”11″ header=”off”]See, e.g., United States v. Hubbell, 530 U.S. 27, 29-30, 38 (2000); Michigan v. Harvey, 494 U.S. 344, 351 (1990); Hoffman v. United States, 341 U.S. 479, 486 (1951).[/tippy] — Miranda’s prophylactic nature is thought to permit the use of unwarned statements made during custodial interrogation for impeachment, [tippy title=”12″ header=”off”]See, e.g., Oregon v. Hass, 420 U.S. 714 (1975).[/tippy] as well as physical evidence or subsequent Mirandized confessions that are obtained as a consequence of antecedent Miranda violations. [tippy title=”13″ header=”off”]See Patane, 542 U.S. at 639-44; id. at 644-45 (Kennedy, J., concurring in the judgment); Elstad, 470 U.S. at 307-09.[/tippy] The prophylactic characterization of Miranda has also led the Court to conclude that its requirements are inapplicable to contexts in which the costs of exclusion are deemed to be particularly high, such as when police officers are facing an exigency that threatens public safety. [tippy title=”14″ header=”off”]See New York v. Quarles, 467 U.S. 649, 655-58 (1984).[/tippy] Where this freewheeling cost-benefit analysis may lead in the future is anybody’s guess.
For their part, Miranda’s advocates do not spend much time defending its conception of unwarned custodial interrogation as inherently coercive. Even Stephen Schulhofer, perhaps Miranda’s most vigorous proponent, concedes that Miranda rests on what he characterizes as a “conclusive presumption” that custodial interrogation involves compulsion within the meaning of the Fifth Amendment. [tippy title=”15″ header=”off”]See Stephen J. Schulhofer, Reconsidering Miranda, 54 U. Chi. L. Rev. 435, 446-53 (1987). If there is any more vigorous Miranda advocate than Schulhofer, it is Yale Kamisar, who has endorsed this position as well. See Yale Kamisar, Can (Did) Congress “Overrule” Miranda?, 85 Cornell L. Rev. 883, 943-50 (2000).[/tippy] Instead, Miranda’s defenders argue for the propriety of prophylactic constitutional law, [tippy title=”16″ header=”off”]See, e.g., Mitchell N. Berman, Constitutional Decision Rules, 90 Va. L. Rev. 1, 114-66 (2004); Evan H. Caminker, Miranda and Some Puzzles of “Prophylactic” Rules, 70 U. Cin. L. Rev. 1 (2001); David Huitema, Miranda: Legitimate Response to Contingent Requirements of the Fifth Amendment, 18 Yale L. & Pol’y Rev. 261, 280-89 (2000); Yale Kamisar, Confessions, Search and Seizure and the Rehnquist Court, 34 Tulsa L.J. 465, 471-76 (1999); Kamisar, supra note 8, at 410-25; Susan R. Klein, Miranda Deconstitutionalized: When the Self-Incrimination Clause and the Civil Rights Act Collide, 143 U. Pa. L. Rev. 417, 481-88 (1994); David A. Strauss, The Ubiquity of Prophylactic Rules, 55 U. Chi. L. Rev. 190 (1988); Charles D. Weisselberg, Saving Miranda, 84 Cornell L. Rev. 109, 181-88 (1998).[/tippy] despite vigorous protests from Miranda opponents. [tippy title=”17″ header=”off”]See, e.g., Dickerson v. United States, 530 U.S. 428, 457-61 (2000) (Scalia, J., dissenting); Joseph D. Grano, Prophylactic Rules in Criminal Procedure: A Question of Article III Legitimacy, 80 Nw. U. L. Rev. 100 (1985). These attacks have not left Miranda’s supporters unmoved; George Thomas, for example, so despairs of any justification for Miranda as a prophylactic rule of Fifth Amendment law that he has come to advocate for Miranda as a due process requirement instead. See George C. Thomas III, Separated at Birth But Siblings Nonetheless: Miranda and the Due Process Notice Cases, 99 Mich. L. Rev. 1081, 1093-117 (2001).[/tippy] David Strauss, for example, defends prophylactic constitutional law by arguing that when there is a sufficiently high risk of a constitutional violation, the Court has frequently concluded that a prophylactic rule is warranted. [tippy title=”18″ header=”off”]See Strauss, supra note 16, at 192-95; David A. Strauss, Miranda, the Constitution, and Congress, 99 Mich. L. Rev. 958, 966-68 (2001).[/tippy] To demonstrate that Miranda is no innovation in constitutional law, he likens Miranda to what he characterizes as a prophylactic rule of First Amendment law forbidding discrimination on the basis of content when regulating speech or other expressive activities because of the risk that content regulation will be motivated by a censorial hostility to disfavored ideas. [tippy title=”19″ header=”off”]See Strauss, supra note 16, at 195-204; Strauss, supra note 18, at 963-65.[/tippy]
Professor Strauss’s analogy to First Amendment jurisprudence is contestable. It is unclear whether the First Amendment rules on which Professor Strauss relies are properly characterized as “prophylactic”; perhaps they are more fairly characterized as “bright-line” rules selected for ease of administration. [tippy title=”20″ header=”off”]For a quite helpful effort to explicate the difference between prophylactic and bright-line rules, see Landsberg, supra note 8, at 950-51.[/tippy] It is equally unclear that strict scrutiny of laws that regulate the content of speech is fairly characterized as prophylactic because such laws can be invalidated without proof of a censorial motive; after all, the First Amendment forbids all laws “abridging the freedom of speech,” [tippy title=”21″ header=”off”]U.S. Const. amend. I.[/tippy] not merely “censorship.” It may be that all inadequately justified restrictions on speech should be deemed to violate the First Amendment regardless of the presence of a censorial motive, whether likely or actual; the principal significance of content regulation is that it demonstrates that the government’s justification for a regulation is suspect when it is not applied uniformly. [tippy title=”22″ header=”off”]In City of Ladue v. Gilleo, 512 U.S. 43 (1994), for example, the Court observed that “[e]xemptions from an otherwise legitimate regulation of a medium of speech may be noteworthy for a reason quite apart from the risks of viewpoint and content discrimination: They may diminish the credibility of the government’s rationale for restricting speech in the first place.” Id. at 52. Indeed, in what was probably the high-water mark of the Court’s suspicion of content regulation, Police Dep’t of Chicago v. Mosley, 408 U.S. 92 (1972), the Court applied the rule against content discrimination to invalidate a prohibition on picketing near schools that exempted labor picketing, see id. at 99-102, yet surely no one thought that Chicago was attempting to censor all views except those of teachers’ unions. Instead, the exemption was used to undermine Chicago’s proffered justification for the regulation on the ground that “Chicago itself has determined that peaceful labor picketing during school hours is not an undue interference with school.” Id. at 100. On the question whether the First Amendment is properly understood as prohibiting only censorial motives, compare, e.g., Elena Kagan, Private Speech, Public Purpose: The Role of Governmental Motive in First Amendment Doctrine, 63 U. Chi. L. Rev. 413 (1996) (identifying censorial motive as the linchpin for First Amendment analysis), and Jed Rubenfeld, The First Amendment’s Purpose, 53 Stan. L. Rev. 767 (2001) (same), with John Fee, Speech Discrimination, 85 B.U. L. Rev. 1103 (2005) (rejecting motive as dispositive), Richard A. Posner, Pragmatism Versus Purposivism in First Amendment Analysis, 54 Stan. L. Rev. 737 (2002) (same), and Eugene Volokh, Speech as Conduct: Generally Applicable Laws, Illegal Courses of Conduct, “Situation-Altering Utterances,” and the Uncharted Zones, 90 Cornell L. Rev. 1277 (2005) (same). Although in Dickerson, the parties defending Miranda took Professor Strauss’s view that many rules of constitutional law are properly understood as prophylactic, the Court refrained from endorsing that position, and Justice Scalia’s dissent argued that of all the assertedly prophylactic rules pressed on the Court, only the rule that an increased sentence imposed after a successful appeal should be deemed vindictive and hence violative of due process could properly be characterized as prophylactic, and for that reason was an anomaly. See Dickerson v. United States, 530 U.S. 428, 457-61 (2000) (Scalia, J., dissenting). Even this concession may be too generous; the presumption of vindictiveness upon resentencing may simply define what constitutes adequate proof of a vindictive motive rather than prophylaxis as that concept was used in Miranda. See, e.g., Alabama v. Smith, 490 U.S. 794, 798-801 (1989). For a contrary argument that the doctrines pressed upon in Dickerson are properly characterized as prophylactic, see Kamisar, supra note 8, at 410-25.[/tippy] Even if Professor Strauss is right to characterize the First Amendment skepticism about content discrimination as prophylactic, the Court has consistently applied the rule against content discrimination whenever an unacceptable censorial risk is thought to be present. [tippy title=”23″ header=”off”]See, e.g., R.A.V. v. City of St. Paul, 505 U.S. 377, 387-90 (1992).[/tippy] When it comes to Miranda, however, the Court has been willing to limit its reach and remedial force even in circumstances that it acknowledges contain the very risk of coercive interrogation that gave rise to the Miranda rule. [tippy title=”24″ header=”off”]See, e.g., New York v. Quarles, 467 U.S. 649, 655-57 (1984).[/tippy] No arguably prophylactic First Amendment doctrine has been circumscribed in a similar way. [tippy title=”25″ header=”off”]Professor Strauss argues that the limitations on the scope of Miranda are no different than the limitations that the Court has placed on First Amendment doctrines, noting, as an example, that “the constitutional rules governing defamation of public officials are different from the rules governing defamation of private individuals, which are in turn different from the rules governing defamation that addresses no subject of public interest.” Strauss, supra note 18, at 968. But the threat that defamation liability will stifle public discussion and debate is thought to be particularly great when the allegedly defamatory statement concerns a public official or a matter of public concern. See, e.g., R.A.V., 505 U.S. at 421 (Stevens, J., concurring in the judgment); Milkovich v. Lorain Journal Co., 497 U.S. 1, 14-16 (1990); Phila. Newspapers, Inc. v. Hepps, 475 U.S. 767, 771-75 (1986); Dun & Bradstreet, Inc. v. Greenmoss Builders, Inc., 472 U.S. 749, 758-61 (1985) (plurality opinion). Thus, as with First Amendment content discrimination doctrine that limits strict scrutiny to circumstances in which the threat to First Amendment values is thought to be high, see R.A.V., 505 U.S. at 387-90, the Court has also granted enhanced First Amendment protection against defamation liability only in circumstances in which the threat to First Amendment values posed by potential defamation liability is thought to be high. In the Miranda context, in contrast, as noted above, even when the precise danger that gave rise to Miranda is present–the coercive pressures of custodial interrogation–the Court has nevertheless limited both the scope and the remedial efficacy of Miranda.[/tippy]
Moreover, even assuming the prevalence of prophylactic rules of constitutional law, in the particular world of Fifth Amendment prophylaxis, Miranda is unique. Prophylactic Fifth Amendment rules are not unknown; for example, even though the Fifth Amendment, by its terms, prohibits only compelling a “witness” to testify in a “criminal case,” [tippy title=”26″ header=”off”]U.S. Const. amend. V.[/tippy] a seemingly prophylactic Fifth Amendment rule requires that even those asked to testify in noncriminal proceedings, who fear that their statements could later be used to incriminate them, must be given immunity to ensure that there is no later use of that testimony in a criminal case. [tippy title=”27″ header=”off”]See United States v. Patane, 542 U.S. 630, 638 (2004) (plurality opinion); Chavez v. Martinez, 538 U.S. 760, 770-73 (2003) (plurality opinion); id. at 777-78 (Souter, J., concurring in the judgment).[/tippy] When testimony is immunized in this fashion, however, the Fifth Amendment is thought to require that it not be used for any purpose in a subsequent criminal case. [tippy title=”28″ header=”off”]See, e.g., Lefkowitz v. Cunningham, 431 U.S. 801, 804-06 (1977); Lefkowitz v. Turley, 414 U.S. 70, 84-85 (1973).[/tippy] In contrast, Miranda, as we have seen, does not carry a prohibition of the use of unwarned statements for impeachment or investigative leads. [tippy title=”29″ header=”off”]See supra text accompanying notes 12-13.[/tippy] Thus, the label “prophylactic” has made Miranda a particularly lame constitutional duck.
The second point of agreement is that Miranda has turned out to be a failure combating the coercive nature of custodial interrogation. [tippy title=”30″ header=”off”]See, e.g., Welsh S. White, Miranda’s Waning Protections: Police Interrogation Practices after Dickerson 139-86 (2001); Louis Michael Seidman, Brown and Miranda, 80 Cal. L. Rev. 673, 740, 745-46 (1992); William J. Stuntz, Miranda’s Mistake, 99 Mich. L. Rev. 975 (2001); George C. Thomas III, Miranda’s Illusion: Telling Stories in the Police Interrogation Room, 81 Tex. L. Rev. 1091 (2003); Welsh S. White, Miranda’s Failure to Restrain Pernicious Interrogation Practices, 99 Mich. L. Rev. 1211 (2001).[/tippy] Despite Miranda, coerced confessions are said to be ubiquitous. [tippy title=”31″ header=”off”]See, e.g., Richard A. Leo, False Confessions: Causes, Consequences, and Solutions, in Wrongly Convicted: Perspectives on Failed Justice 36, 44-47 (Saundra D. Westervelt & John A. Humphrey eds., 2001); Hugo Adam Bedau & Michael L. Radelet, Miscarriages of Justice in Potentially Capital Cases, 40 Stan. L. Rev. 21, app. B, tbls.1-2 & 5-8 (1987); Steven A. Drizin & Richard A. Leo, The Problem of False Confessions in the Post-DNA World, 82 N.C. L. Rev. 891, 901-62 (2004); Samuel R. Gross et al., Exonerations in the United States: 1989 Through 2003, 95 J. Crim. L. & Criminology 523, 544-46 (2005); Richard A. Leo et al., Bringing Reliability Back In: False Confessions and Legal Safeguards in the Twenty-First Century, 2006 Wis. L. Rev. 479, 512-520; Richard A. Leo & Richard J. Ofshe, The Consequences of False Confessions: Deprivations of Liberty and Miscarriages of Justice in the Age of Psychological Interrogation, 88 J. Crim. L. & Criminology 429 (1998).[/tippy] Thus, we are told that stronger medicine is needed, such as videotaping custodial interrogation, [tippy title=”32″ header=”off”]See, e.g., Yale Kamisar, Brewer v. Williams –A Hard Look at a Discomfiting Record, in Police Interrogation and Confessions: Essays in Law and Policy 113, 132-37 (1980); White, supra note 30, at 190-95; Albert W. Alschuler, Constraint and Confession, 74 Denv. U. L. Rev. 957, 977-78 (1997); Paul G. Cassell, Miranda’s Social Costs: An Empirical Reassessment, 90 Nw. U. L. Rev. 387, 486-97 (1996); Daniel Donovan & John Rhodes, Comes a Time: The Case for Recording Interrogations, 61 Mont. L. Rev. 223 (2000); Drizin & Leo, supra note 31, at 966-1001; Mark A. Godsey, Reformulating the Miranda Warnings in Light of Contemporary Law and Understandings, 90 Minn. L. Rev. 781, 808-09, 815 (2006); Leo, supra note 31, at 49-50; Richard A. Leo, The Impact of Miranda Revisited, 86 J. Crim. L. & Criminology 621, 681-92 (1996); Richard A. Rosen, Reflections on Innocence, 2006 Wis. L. Rev. 237, 261-66; Stephen J. Schulhofer, Miranda, Dickerson, and the Puzzling Persistence of Fifth Amendment Exceptionalism, 99 Mich. L. Rev. 941, 955 (2001); Christopher Slobogin, Toward Taping, 1 Ohio St. J. Crim. L. 309 (2003); Stuntz, supra note 30, at 981; Wayne T. Westling, Something Is Rotten in the Interrogation Room: Let’s Try Video Oversight, 34 J. Marshall L. Rev. 537 (2001).[/tippy] requiring counsel during interrogation, [tippy title=”33″ header=”off”]See, e.g., Charles J. Ogletree, Are Confessions Really Good for the Soul?: A Proposal to Mirandize Miranda, 100 Harv. L. Rev. 1826 (1987); Schulhofer, supra note 32, at 955.[/tippy] strengthening constitutional regulation of the admissibility of confessions, [tippy title=”34″ header=”off”]See, e.g., Kamisar, Kauper’s “Judicial Examination of the Accused” Forty Years Later–Some Comments on a Remarkable Article, in Police Interrogation and Confessions, supra note 32, at 77; Sharon L. Davies, The Reality of False Confessions–Lessons of the Central Park Jogger Case, 30 N.Y.U. Rev. L. & Soc. Change 209, 231-52 (2006); Richard J. Ofshe & Richard A. Leo, The Decision to Confess Falsely: Rational Choice and Irrational Action, 74 Denv. U. L. Rev. 979, 1115-22 (1997); Stuntz, supra note 30, at 995-98.[/tippy] forbidding interrogation techniques thought to be particularly likely to produce false or coercive confessions, [tippy title=”35″ header=”off”]See, e.g., White, supra note 30, at 196-215; Alschuler, supra note 32, at 967-78; Miriam S. Gohara, A Lie for a Lie: False Confessions and the Case for Reconsidering the Legality of Deceptive Interrogation Techniques, 33 Fordham Urb. L.J. 791 (2006); Mark A. Godsey, Rethinking the Involuntary Confession Rule: Toward a Workable Test for Identifying Compelled Self-Incrimination, 93 Cal. L. Rev. 465, 515-39 (2005); Alan Hirsch, Threats, Promises, and False Confessions: Lessons of Slavery, 49 How. L.J. 31, 54-59 (2005); Deborah Young, Unnecessary Evil: Police Lying in Interrogations, 28 Conn. L. Rev. 425 (1996).[/tippy] or abolishing custodial interrogation entirely. [tippy title=”36″ header=”off”]See, e.g., Donald A. Dripps, Foreword: Against Police Interrogation–And the Privilege Against Self-Incrimination, 78 J. Crim. L. & Criminology 699, 723-28 (1988); Irene Merker Rosenberg & Yale L. Rosenberg, A Modest Proposal for the Abolition of Custodial Confessions, 68 N.C. L. Rev. 69, 109-14 (1989).[/tippy]
My task will be to demonstrate that both points of consensus are wrong. On the first, I will argue that Miranda was quite right to conclude that custodial interrogation inherently involves compulsion within the meaning of the Fifth Amendment. Thus, Miranda is not prophylactic–its warnings are required only when a suspect is compelled to incriminate himself, and they ensure that incriminating statements are received in evidence only when a suspect has validly waived the right to be free from compelled self-incrimination. On the second, I will argue that Miranda should be measured by whether it has produced greater compliance with the Fifth Amendment, and on that score, Miranda is a resounding success. Miranda’s required warnings succeed in producing valid waivers of Fifth Amendment rights, and therefore prevent what would otherwise be unconstitutional interrogations. Although suspects may frequently misgauge their own interests in deciding whether to submit to custodial interrogation, the Fifth Amendment does not protect suspects from themselves–it is not aimed at “[p]reventing foolish (rather than compelled) confessions,” to use Justice Scalia’s typically memorable formulation. [tippy title=”37″ header=”off”]Dickerson v. United States, 530 U.S. 428, 449 (2000) (Scalia, J., dissenting).[/tippy] Perhaps Miranda is a failure from the standpoint of those who think that the Constitution condemns any tactic that might smack of overreaching or risk convicting the innocent, but the Constitution does not demand perfection. The Fifth Amendment, in particular, prohibits only compulsion; and once a suspect validly waives his right to be free from compelled self-incrimination, the Fifth Amendment does not protect a suspect from his own improvident decision to submit to interrogation. As for the Due Process Clause, although it is hardly indifferent to the risk of wrongful conviction, it surely does not demand that the risk of error in criminal litigation be zero–a risk of error inheres in virtually any investigative tactic, from eyewitness testimony to fingerprints. For the most part, however, the critics of police interrogation as currently practiced cannot demonstrate that the tactics that they would forbid under the rubric of due process pose any greater risk of convicting the innocent than any others. We are far from the day when an empirical case can be made that interrogation tactics that are considered appropriate under current law should be condemned because they create what is thought to be an undue risk of error–a truly prophylactic approach to constitutional criminal procedure. In short, for constitutional purposes, a confession obtained after a valid waiver of Fifth Amendment rights is good enough.
I. Miranda Is Not Prophylactic: A Constitutional Defense of Miranda
The Supreme Court’s first encounter with the intersection of the Fifth Amendment and custodial interrogation was Bram v. United States. [tippy title=”38″ header=”off”]168 U.S. 532 (1897).[/tippy] That case involved the interrogation of Bram, the first officer on an American ship, who was suspected of murdering the ship’s captain, the captain’s wife, and the second mate. [tippy title=”39″ header=”off”]Id. at 534-37.[/tippy] After Charles Brown, who was at the ship’s wheel on the night of the murders, told his shipmates that he had seen Bram murder the captain, Bram was put in irons and subsequently placed in police custody when the ship reached Halifax. [tippy title=”40″ header=”off”]Id. at 535-37.[/tippy] In Halifax, Bram was brought before a police detective, who stripped Bram of his clothing, which he then searched. [tippy title=”41″ header=”off”]Id. at 538.[/tippy] At that point, the detective later testified that he told Bram:
“Bram, we are trying to unravel this horrible mystery.” I said: “Your position is rather an awkward one. I have had Brown in this office and he made a statement that he saw you do the murder.” He said: “He could not have seen me; where was he?” I said: “He states he was at the wheel.” “Well,” he said, “he could not see me from there.” I said: “Now, look here, Bram, I am satisfied that you killed the captain from all I have heard from Mr. Brown. But,” I said, “some of us here think you could not have done all that crime alone. If you had an accomplice, you should say so, and not have the blame of this horrible crime on your own shoulders.” He said: “Well, I think, and many others on board the ship think, that Brown is the murderer; but I don’t know anything about it.” [tippy title=”42″ header=”off”]Id. at 539.[/tippy]
This testimony was subsequently offered and received at Bram’s trial as a false effort at exculpation reflecting Bram’s consciousness of guilt. [tippy title=”43″ header=”off”]Id. at 541-42.[/tippy]
If Bram’s interrogation sounds familiar, it should. The tactics in play in Bram are not dissimilar to the police tactics considered in Miranda itself. The police manuals that the Court famously reviewed essentially advised interrogators to induce suspects to believe that it was in the suspect’s interest to confess by minimizing or excusing the suspect’s conduct, leading him to believe that his conviction is a certainty, or assuring him that his cooperation will be rewarded. [tippy title=”44″ header=”off”]Miranda v. Arizona, 384 U.S 436, 449-55 (1966). See also, e.g., Kamisar, What Is an “Involuntary” Confession?, in Police Interrogation and Confessions, supra note 32, at 1; White, supra note 30, at 25-36.[/tippy] Even after Miranda, these tactics continue to be used; while the requisite warnings are now given, interrogators still try to convince suspects that they will be better off if they confess. [tippy title=”45″ header=”off”]See, e.g., White, supra note 30, at 76-101; Leo, supra note 31, at 39-41; Richard A. Leo, Miranda’s Revenge: Police Interrogation as a Confidence Game, 30 Law & Soc’y Rev. 259 (1996); Richard A. Leo & Welsh S. White, Adapting to Miranda: Modern Interrogators’ Strategies for Dealing with the Obstacles Posed by Miranda, 84 Minn. L. Rev. 397, 431-50 (1999). See also Fred E. Inbau et al., Criminal Interrogation and Confessions 232-80 (4th ed. 2001).[/tippy]
This was certainly the approach that I took when I participated in interrogations during my years as a prosecutor, usually in an effort to “flip” a potential cooperator upon arrest. I never met a suspect who was eager to confess–the suspects I encountered uniformly wished to stay out of jail. In my experience, suspects cooperated with interrogation only if they perceived it to be in their interest to do so. So before beginning an interrogation, I would make a few “let-me-tell-you-why-I’m-here” remarks, which I carefully prefaced with an admonition that I wanted the suspect not to respond so as to avoid engaging in unwarned custodial interrogation. [tippy title=”46″ header=”off”]See Rhode Island v. Innis, 446 U.S. 291, 300-02 (1980).[/tippy] In those remarks, I would convey to the suspect that he was in a lot of trouble, characterizing things in as dire a fashion as a plausible view of the evidence would allow. Then, I would give the requisite warnings, and if I obtained a waiver, I would stress my ability to help the suspect if he cooperated. Thus, I tried to use the threat of sanctions to induce cooperation, just as the manuals instruct.
The question posed by Bram–and by my own interrogations–is whether an implicit threat to send someone to prison for as long as possible unless he agrees to submit to interrogation is a form of compulsion within the meaning of the Fifth Amendment. To me, this has never seemed like a very hard question; nor did the Court see it as particularly difficult in Bram. In that case, the Court concluded that Bram’s statements were inadmissible because they “must necessarily have been the result of either hope or fear, or both, operating on the mind.” [tippy title=”47″ header=”off”]Bram, 168 U.S. at 562 (emphasis added).[/tippy] The Court explained:
It cannot be doubted that . . . the result was to produce upon [Bram’]s mind the fear that if he remained silent it would be considered an admission of guilt . . . and it cannot be conceived that the converse impression would not also have naturally arisen, that by denying there was hope of removing suspicion from himself. [tippy title=”48″ header=”off”]Id.[/tippy]
There are those who believe that the Fifth Amendment’s prohibition on compelled self-incrimination was intended to do no more than codify the common law privilege against compelled oaths. [tippy title=”49″ header=”off”]See, e.g., Akhil Reed Amar, The Constitution and Criminal Procedure: First Principles 70-88 (1997); Joseph D. Grano, Confessions, Truth, and the Law 123-43 (1993); 3 John Henry Wigmore, Evidence in Trials at Common Law § 823 (Aspen Law & Business 1970) (1904); William T. Pizzi & Morris B. Hoffman, Taking Miranda’s Pulse, 58 Vand. L. Rev. 813, 842-44 (2005).[/tippy] On this view, Bram went wrong in extending the Fifth Amendment to unsworn interrogation and improperly conflating the privilege against compelled self-incriminatory testimony with the common law rule that forbade the reception of an involuntary confession in evidence. [tippy title=”50″ header=”off”]See, e.g., Godsey, supra note 35, at 477-88. Other commentators, however, argue that the Fifth Amendment incorporated elements of both the privilege and the evidentiary rule. See, e.g., Leonard W. Levy, Origins of the Fifth Amendment: The Right Against Self-Incrimination 325-32, 405-32 (1968); Laurence A. Benner, Requiem for Miranda: The Rehnquist Court’s Voluntariness Doctrine in Historical Perspective, 67 Wash. U. L.Q. 59, 92-109 (1989); Thomas Y. Davies, Farther and Farther from the Original Fifth Amendment: The Recharacterization of the Right against Self-Incrimination as a “Trial Right” in Chavez v. Martinez, 70 Tenn. L. Rev. 987 (2003); Lawrence Herman, The Unexplored Relationship between the Privilege against Compulsory Self-Incrimination and the Involuntary Confession Rule (Part II), 53 Ohio St. L.J. 497, 529-50 (1992); Stephen A. Saltzburg, Miranda v. Arizona Revisited: Constitutional Law or Judicial Fiat, 26 Washburn L.J. 1, 4-8 (1986).[/tippy] But once one agrees that the Fifth Amendment extends to admitting the results of an unsworn interrogation into evidence at a subsequent trial, on the ground that the declarant who was under compulsion to confess becomes a “witness” within the meaning the Fifth Amendment, [tippy title=”51″ header=”off”]Even before Bram, the Court had taken this position in Boyd v. United States, 116 U.S. 616, 634-35 (1886). For arguments in support of this conclusion, see, e.g., Kamisar, Equal Justice in the Gatehouses and Mansions of American Procedure (1965), in Police Interrogation and Confessions, supra note 32, at 27 [hereinafter Kamisar, Equal Justice]; Kamisar, A Dissent from the Miranda Dissents: Some Comments on the “New” Fifth Amendment and the Old “Voluntariness” Test (1966), in Police Interrogation and Confessions, supra note 32, at 41, 41-68 [hereinafter Kamisar, A Dissent]; Albert W. Alschuler, A Peculiar Privilege in Historical Perspective: The Right to Remain Silent, 94 Mich. L. Rev. 2625, 2667-72 (1996); Herman, supra note 50, at 529-50. Indeed, even Justice Scalia and Justice Thomas, the two Miranda foes now sitting on the Court, see Dickerson v. United States, 530 U.S. 428, 461-65 (2000) (Scalia, J., joined by Thomas, J., dissenting), have concluded that the term “witness,” at least within the meaning of the Fifth Amendment, includes not only one who gives testimony but also anyone who furnishes evidence to the prosecution that it later uses at trial. See United States v. Hubbell, 530 U.S. 27, 49-56 (2000) (Thomas, J., joined by Scalia, J., concurring).[/tippy] it is hard to argue with the concept of compulsion embraced in Bram.
The appropriate starting point, of course, is to define compulsion for purposes of the Fifth Amendment. There is general agreement that the paradigmatic form of compulsion that was forbidden by the privilege against compelled self-incrimination as it was understood at the time of the Fifth Amendment’s ratification was the compelled oath. [tippy title=”52″ header=”off”]See, e.g., Grano, supra note 49, at 124-31; Levy, supra note 50, at 205-404; Godsey, supra note 35, at 477-85; John H. Langbein, The Historical Origins of the Privilege Against Self-Incrimination at Common Law, 92 Mich. L. Rev. 1047, 1072-85 (1994); Eben Moglen, Taking the Fifth: Reconsidering the Origins of the Constitutional Privilege Against Self-Incrimination, 92 Mich. L. Rev. 1086 (1994).[/tippy] An accused who declined to take the oath risked conviction, contempt, or other sanctions. [tippy title=”53″ header=”off”]See, e.g., Levy, supra note 50, at 130-33, 269; Lawrence Herman, The Unexplored Relationship Between the Privilege Against Compulsory Self-Incrimination and the Involuntary Confession Rule (Part I), 53 Ohio St. L.J. 101, 108, 134-35 (1992); Langbein, supra note 52, at 1073.[/tippy] The original meaning of compulsion, then, is the use of a threat of criminal sanctions to obtain testimony. Indeed, the most natural reading of the term “compulsion” is the threat of adverse consequences–such as the historically paradigmatic sanction of conviction or contempt–as a form of coercive pressure on a suspect to become a “witness” against himself. [tippy title=”54″ header=”off”]Mark Godsey, although a critic of Bram, provides a particularly helpful account in support of just this view of compulsion. See Godsey, supra note 35, at 491-97.[/tippy] The textual prohibition on compelled self-incrimination in the Fifth Amendment, however, is not limited to those who are compelled to take oaths, nor to those subjected to torture or other specified techniques for eliciting confessions. In particular, unlike the Sixth Amendment’s Compulsory Process Clause, which addresses only the compulsion that inheres in requiring a witness to testify at a trial, [tippy title=”55″ header=”off”]“In all criminal prosecutions the accused shall enjoy the right…to have compulsory process for obtaining witnesses in his favor….” U.S. Const. amend. VI.[/tippy] the Fifth Amendment is triggered by any form of compulsion, not merely the obligation to obey compulsory process requiring one to appear and testify at a judicial proceeding. This suggests that the holding in Bram was correct; Bram was subjected to compulsion–the threat of criminal sanctions–during his interrogation, and that threat was used to induce him to submit to interrogation.
To be sure, the threat of conviction or contempt for a refusal to take a mandatory oath is a more direct form of compulsion than the future risk of sanctions facing Bram, but again, the Fifth Amendment’s text prohibits any quantum or form of compulsion, not just compulsion through direct, immediate, and relatively certain sanctions to be imposed in an already pending proceeding. As the Court wrote in Bram,
A confession can never be received in evidence where the prisoner has been influenced by any threat or promise; for the law cannot measure the force of the influence used, or decide upon its effect upon the mind of the prisoner, and therefore excludes the declaration if any degree of influence has been exerted. [tippy title=”56″ header=”off”]Bram v. United States, 168 U.S. 532, 543 (1897) (quoting 3 Sir Wm. Oldnall Russell, A Treatise on Crimes and Misdemeanors 478 (Horace Smith & A.P. Perceval Keep eds., 6th ed. 1896)). For a fascinating discussion of the evolution of the rule against promises of leniency during interrogation between the seventeenth and nineteenth centuries and how it led some jurisdictions to a solution much like Miranda, see Wesley MacNeil Oliver, Magistrates’ Examinations, Police Interrogations, and Miranda-Like Warnings in the Nineteenth Century, 81 Tul. L. Rev. 777, 786-810 (2007).[/tippy] It was through the threat of a prosecution for murder that Bram was induced to accuse Brown while denying that Brown could have seen him, and that is compulsion in the same sense that persons were once compelled to testify under oath by the threat of contempt. As the Court observed, Bram could not have decided whether to respond to the detective’s queries without necessarily considering the peril he faced. Similarly, when I succeeded in obtaining cooperation from my interrogees, they necessarily had to consider the magnitude of sanctions that they faced if they failed to cooperate. Thus, whatever else might amount to compulsion within the meaning of the Fifth Amendment, Bram correctly concluded that threatening a suspect with criminal sanctions during custodial interrogation falls within the scope of the Fifth Amendment’s protection against compelled self-incrimination. [tippy title=”57″ header=”off”]The Supreme Court has occasionally, and without citing historical evidence, characterized torture as the chief evil against which the right against compelled self-incrimination was directed. See, e.g., McKune v. Lile, 536 U.S. 24, 41 (2002). Although the historical basis for this focus on torture as opposed to compulsory oaths is questionable, even under the Court’s own precedents, see, e.g., Miranda, 384 U.S. at 442-43, treating torture as the paradigm of compulsion does not blunt the force of the argument advanced above. Torture, like the sanctions that following from refusing to take a compulsory oath, was a form of punishment that followed from a refusal to cooperate with interrogation. The text of the Fifth Amendment, moreover, does not identify only a single form of compulsion that is forbidden. Instead, any form of compulsion–be it torture or any other type of punitive sanction based on a refusal to submit to interrogation–renders a subsequent statement inadmissible. That said, it is beyond the scope of the current project to attempt a comprehensive account of Fifth Amendment compulsion. For present purposes, it is sufficient to observe that the original meaning of the text supports Bram’s view that a threat of a criminal prosecution amounts to compulsion, whatever else might also constitute compulsion for purposes of the Fifth Amendment. For a more extended discussion of what types of adverse governmental actions might amount to compulsion within the meaning of the Fifth Amendment, see, for example, Grano, supra note 49, at 59-83; Ronald J. Allen & M. Kristin Mace, The Self-Incrimination Clause Explained and Its Future Predicted, 94 J. Crim. L & Criminology 243, 250-56 (2004); Godsey, supra note 35, at 491-97; Peter Westen & Stuart Mandell, To Talk, To Balk, or To Lie: The Emerging Fifth Amendment Doctrine of “Preferred Response,” 19 Am. Crim. L. Rev. 521, 535-40 (1982).[/tippy]
At the outset, I promised to defend the proposition that custodial interrogation always involves compulsion within the meaning of the Fifth Amendment. Even if I am right that Bram’s interrogation, or my own, involved Fifth Amendment compulsion, that does not mean that every instance of custodial interrogation involves compulsion within the meaning of the Fifth Amendment. Of course, once Bram had established that the law cannot endeavor to measure the effect of compulsion on the mind of a prisoner, it was perhaps only a small step to Miranda’s holding that custodial interrogation inherently involves compulsion, but it is surely possible to argue that custodial interrogation sometimes occurs in the absence of anything fairly characterized as compulsion. In his dissenting opinion in Miranda, Justice White made the point this way: “[U]nder the Court’s rule, if the police ask [an arrestee] a single question such as ‘Do you have anything to say?’ or ‘Did you kill your wife?’ his response, if there is one, has somehow been compelled . . . . Common sense informs us to the contrary.” [tippy title=”58″ header=”off”]384 U.S. at 533-34 (White, J., dissenting).[/tippy] Most commentators seem to find this point unanswerable. [tippy title=”59″ header=”off”]See, e.g., Grano, supra note 49, at 59-83; Gerald M. Caplan, Questioning Miranda, 38 Vand. L. Rev. 1417, 1446-58 (1985); Paul G. Cassell & Bret S. Hayman, Police Interrogation in the 1990s: An Empirical Study of the Effects of Miranda, 43 U.C.L.A. L. Rev. 839, 919 (1996); George C. Thomas III, An Assault on the Temple of Miranda, 85 J. Crim. L. & Criminology 807, 818-22 (1995) (reviewing Joseph D. Grano, Confessions, Truth, and the Law (1993)); Anthony P. Bigornia, Supreme Court Review, The Supreme Court’s Cost-Benefit Analysis of Federal Habeas Review of Alleged Miranda Violations, 84 J. Crim. L. & Criminology 915, 934-35 (1994). See also Schulhofer, supra note 15, at 448.[/tippy] But when a public official, with a badge and a gun, deprives a suspect of his liberty, places him in custody, and then asks, “do you have anything to say?” is it really the case that there is no compulsion to respond?
One thing that is unquestionably inherent in custodial interrogation is an assertion of the state’s power to deprive suspects of their liberty. When the state exercises this power, and then begins to interrogate the detainee, compulsion to respond to the interrogation is an inevitable result. After all, implicit in custodial interrogation is the threat that the detention and accompanying interrogation will be followed by a criminal prosecution with its attendant sanctions. That kind of threat, in turn, is the hallmark of Fifth Amendment compulsion, as we have seen. Of course, some detainees will have the fortitude to ignore their jailer’s questions, but if the interrogator’s questions go unanswered, there has been no Fifth Amendment violation because the suspect has not been compelled to become a witness against himself. [tippy title=”60″ header=”off”]In this respect, it bears remembering that Miranda does not prohibit unwarned interrogation per se, it only addresses the admission of evidence derived from the unwarned interrogation in a subsequent criminal case. See Chavez v. Martinez, 538 U.S. 760, 770-73 (2003) (plurality opinion); id. at 789 (Kennedy, J., concurring in the judgment). See also New York v. Quarles, 467 U.S. 649, 686 (1984) (Marshall, J., dissenting). Indeed, in Miranda itself, the Court framed its holding in terms of the admissibility of evidence: “[T]he prosecution may not use statements, whether exculpatory or inculpatory, stemming from custodial interrogation of the defendant unless it demonstrates the use of procedural safeguards effective to secure the privilege against self-incrimination.” 384 U.S. at 444. Thus, if an officer engages in unwarned interrogation and is ignored, there may be no compulsion within the meaning of the Fifth Amendment, but there is no Miranda violation either.[/tippy] When the suspect submits to interrogation, however, at least absent extraordinary circumstances, I see no plausible way to deny that the suspect has been compelled to respond to his captor’s questions within the meaning of the Fifth Amendment by virtue of the compulsive power of custody and the inherent threat that it will continue unless the jailer is somehow satisfied. As Bram recognizes, the hope of pleasing one’s jailer–convincing him that he is holding an innocent person, or at least to be lenient–cannot help but enter into the calculations of one subjected to custodial interrogation. Conversely, the threat that the jailer, if unsatisfied by the suspect, will ultimately seek criminal sanctions is equally implicit in any assertion of the state’s power to detain and interrogate. [tippy title=”61″ header=”off”]For an example of a suspect who seems to have suffered from a form of mental illness that instilled a desire to confess even before he was taken into custody, see Colorado v. Connelly, 479 U.S. 157, 161-62 (1986).[/tippy]
To be sure, one can imagine bizarre cases in which the use of custodial interrogation bears no causal relation to an ensuing confession, such as a suspect who suffers under some form of mental illness that produces a desire for confession or who otherwise decides to confess before he is taken into custody. It may be that in an appropriate case, the Court will carve out such cases from Miranda. The Court has already characterized Miranda’s holding as establishing a “presumption of coercion” “inherent in custodial interrogations,” [tippy title=”62″ header=”off”]United States v. Patane, 542 U.S. 630, 639 (2004) (plurality opinion); id. at 646 (Souter, J., dissenting); Arizona v. Roberson, 486 U.S. 675, 686 (1988); Oregon v. Elstad, 470 U.S. 298, 307 (1985).[/tippy] perhaps it will eventually hold that this presumption can be rebutted by proof that there was no causal relationship between an incriminating statement and custodial interrogation. [tippy title=”63″ header=”off”]Indeed, the Court has already held that there are some situations that involve technical custody but in which questioning is so routine and noncoercive that no compulsion in the constitutional sense is present. Thus, the Court has held that questioning during a routine traffic stop does not trigger Miranda, see Berkemer v. McCarty, 468 U.S. 420, 435-41 (1984), and that routine booking questions also fail to trigger Miranda, see Pennsylvania v. Muniz, 496 U.S. 582, 601-02 (1990). Indeed, even the officer’s question about the location of the suspect’s gun upon apprehending a suspect in the just-completed armed rape at issue in New York v. Quarles, 467 U.S. at 651-52, may not involve Fifth Amendment compulsion; the events unfolded so quickly and spontaneously that the suspect may well not have been subject to the kind of compulsion that is present when a suspect must necessarily contemplate the risks inhering in remaining silent while under interrogation. See id. at 655-58. Given the Court’s willingness to limit the scope of Miranda to contexts in which its presumption of compulsion holds, it is likely that Miranda will not reach cases in which the suspect confesses for reasons unrelated to the existence of custodial interrogation.[/tippy] But surely the potential for bizarre outliers in which there is no relationship between custodial interrogation and an ensuing confession is no reason to doubt that absent extraordinary circumstances, the existence of the custodial relation between interrogator and interrogee exerts compulsion within the meaning of the Fifth Amendment. If the worst that can be said of Miranda is that it envisions the usual case rather than the bizarre outlier, then it should be thought to fare pretty well in the canon of constitutional criminal procedure.
It is doubtless true that Justice White’s hypothetical interrogation involves only a bit of compulsion–far less than the compulsion that inheres in physical punishment [tippy title=”64″ header=”off”]See Brown v. Mississippi, 297 U.S. 278 (1936).[/tippy] or the other tactics considered impermissible when the Court was assessing confessions under the Due Process Clause’s “voluntariness” test. [tippy title=”65″ header=”off”]See, e.g., Lynumn v. Illinois, 372 U.S. 528 (1963) (inducing confession by threats that defendant would lose custody of her children); Payne v. Arkansas, 356 U.S. 560 (1958) (holding suspect three days incommunicado with little food and threats of lynching); Leyra v. Denno, 347 U.S. 556 (1954) (obtaining confessions after questioning ill defendant repeatedly over three days and using hypnosis to suggest guilt); Watts v. Indiana, 338 U.S. 49 (1949) (questioning suspect repeatedly and persistently over six days); Malinski v. New York, 324 U.S. 401 (1945) (stripping suspect naked for three hours and then questioning; questioning continued over three days during which defendant was held incommunicado); Ward v. Texas, 316 U.S. 547 (1942) (questioning repeatedly as suspect was moved from jail to jail over a three-day period and told of threats of lynching); Chambers v. Florida, 309 U.S. 227 (1940) (interrogating repeatedly over five days culminating in an all-night session). For discussions of the relative breadth of the due-process concept of voluntariness as compared to the Fifth Amendment’s concept of compulsion, see Catherine Hancock, Due Process Before Miranda, 70 Tul. L. Rev. 2195 (1996); Steven Penney, Theories of Confession Admissibility: A Historical View, 25 Am. J. Crim. L. 309 (1998).[/tippy] Once the Court left the vagaries of due process behind and held that the Fifth Amendment’s prohibition on compelled self-incrimination was applicable to the states, [tippy title=”66″ header=”off”]See Malloy v. Hogan, 378 U.S. 1 (1964).[/tippy] however, the pertinent constitutional question, even as applied to state prosecutions, was whether compulsion was used to obtain an incriminating statement. And, as we have seen, the Fifth Amendment forbids the admission of any confession obtained through compulsion, whether a little or a lot, as the Court held in Bram. Thus, Miranda’s holding on the coercive nature of custodial interrogation is not only defensible, it was not even much of an innovation in Fifth Amendment jurisprudence. [tippy title=”67″ header=”off”]Indeed, in Miranda itself, the Court observed that the applicability of the Fifth Amendment to custodial interrogation “could have been taken as settled in federal courts almost 70 years ago [in Bram].” 384 U.S. at 461. To be sure, Bram had been rarely invoked by the Court prior to Miranda, but the Court decided only a single case involving the application of the Fifth Amendment to custodial interrogation between Bram and Miranda, and in that case, it treated Bram as controlling. See Ziang Sung Wan v. United States, 266 U.S. 1 (1924).[/tippy]
Accordingly, there is little if any prophylaxis in the view that compulsion within the meaning of the Fifth Amendment is inherent in the process of custodial interrogation. Indeed, that view was not invented in Miranda; in fact the Court’s principal pre-Miranda advocate of unwarned interrogation, Justice Jackson, took nearly that position more than two decades before Miranda. Dissenting in Ashcraft v. Tennessee, [tippy title=”68″ header=”off”]322 U.S. 143 (1944).[/tippy] Justice Jackson objected to the majority’s concern with coercion in its analysis of the admissibility of a confession under the Due Process Clause because, in his view, coercion was always present in custodial interrogation:
The Court bases its decision on the premise that custody and examination of a prisoner for thirty-six hours is ‘inherently coercive.’ Of course it is. And so is custody and examination for one hour. Arrest itself is inherently coercive, and so is detention. When not justified, infliction of such indignities upon the person is actionable as a tort. Of course such acts put pressure upon the prisoner to answer questions[,] to answer them truthfully, and to confess if guilty. [tippy title=”69″ header=”off”]Id. at 161 (Jackson, J., dissenting). In a similar vein, Joseph Grano argued that one cannot assess the impact of any interrogation tactic without developing normative standards to define what tactics should be deemed permissible. See Grano, supra note 49, at 59-83.[/tippy] The careful reader will note that Justice Jackson used the term “coercive,” a concept that the Court had applied in its due process jurisprudence concerning the admissibility of confessions in state prosecutions before it held the Fifth Amendment applicable to the states, rather than the Fifth Amendment concept of compulsion. Still, it was but a small step from his acknowledgement that custodial interrogation is inherently coercive to Miranda’s conclusion that custodial interrogation involved compulsion as that concept is understood under the Fifth Amendment. [tippy title=”70″ header=”off”]For an account of the meaning of compulsion under the Fifth Amendment, see supra text accompanying notes 52-55. The term “coercion,” as it came to be used in the due process cases, had no especially precise meaning. It appeared to denote methods involving a degree of psychological or physical pressure on a suspect that gave rise to an unacceptable risk of a false confession, or methods that were inconsistent with normative standards that the Court was prepared to impose governing the amount of pressure that interrogators would be permitted to utilize. See Penney, supra note 65, at 341-46. Whatever its precise meaning, this due-process concept of coercion is plainly not far from the Fifth Amendment’s conception of compulsion. See Hancock, supra note 65, at 2223-32.[/tippy]
The account of Fifth Amendment compulsion that I have advanced is rejected by nearly all commentators, but for strikingly unpersuasive reasons. Albert Alschuler, for example, has written that “[a] person can influence another’s choice without compelling it; to do so she need only keep her persuasion within appropriate bounds of civility . . . .” [tippy title=”71″ header=”off”]Alschuler, supra note 51, at 2626.[/tippy] Perhaps so, but the “appropriate bounds of civility” surely does not include imprisoning the object of one’s attempts at persuasion; as Justice Jackson acknowledged, that form of “persuasion” goes into the realm of the tortious. [tippy title=”72″ header=”off”]See supra text accompanying note 69. Professor Grano, in contrast, argued that noncustodial interrogation frequently involves tactics every bit as compulsive as custodial interrogation. See Joseph D. Grano, Miranda’s Constitutional Difficulties: A Reply to Professor Schulhofer, 55 U. Chi. L. Rev. 174, 186 (1988). For noncustodial interrogation, however, there is force to Professor Alschuler’s claim that the police are undertaking what is properly characterized as persuasion as opposed to compulsion. In any event, even if Miranda can fairly be accused of under-enforcing the Fifth Amendment by limiting its holding to custodial interrogation, that accusation hardly undermines the thesis that Miranda is not prophylactic.[/tippy] Mere persuasion, under Bram or in ordinary parlance, is unaccompanied by actual or threatened deprivation of liberty.
Commentators also argue that Miranda’s concept of compulsion and waiver is inconsistent; they argue that if the threat of criminal sanctions were deemed compulsion, a defendant could never validly waive his Fifth Amendment rights because a waiver given while a suspect is subject to compulsion could never be voluntary. [tippy title=”73″ header=”off”]See, e.g., Donald A. Dripps, About Guilt and Innocence: The Origins, Development, and Future of Constitutional Criminal Procedure 81, 119-20 (2003); Joseph D. Grano, Selling the Idea to Tell the Truth: The Professional Interrogator and Modern Confessions Law, 84 Mich. L. Rev. 662, 671-72 (1986) (reviewing Fred E. Inbau et al., Criminal Interrogation and Confessions (1986)); Seidman, supra note 30, at 740, 744; Welsh S. White, Defending Miranda: A Reply to Professor Caplan, 39 Vand. L. Rev. 1, 10 (1986). Similarly, Professor Thomas argues that Fifth Amendment compulsion cannot include the threat of conviction or the Fifth Amendment would prevent a defendant from testifying at trial in his own defense in order to avoid conviction because such testimony is necessarily compelled. See Thomas, supra note 59, at 820-21. These arguments echo Justice White’s Miranda dissent:
But if the defendant may not answer without a warning a question such as “Where were you last night?” without having his answer be a compelled one, how can the Court ever accept his negative answer to the question of whether he wants to consult his retained counsel or counsel whom the Court will appoint?
384 U.S. at 536 (White, J., dissenting).[/tippy] But that does not follow.
Miranda applied settled principles of waiver as it held that to introduce the results of custodial interrogation into evidence, “a heavy burden rests on the government to demonstrate that the defendant knowingly and intelligently waived his privilege against self-incrimination and his right to retained or appointed counsel,” adding that “[t]his Court has always set high standards of proof for the waiver of constitutional rights.” [tippy title=”74″ header=”off”]384 U.S. at 475.[/tippy] The standards for waiver, however, do not require that the defendant face no risks if he chooses to assert his rights. We know this from the guilty-plea cases, which characterize a plea of guilty as a waiver, among other things, of the right against compelled self-incrimination, [tippy title=”75″ header=”off”]See, e.g., Florida v. Nixon, 543 U.S. 175, 187-88 (2004); United States v. Ruiz, 536 U.S. 622, 628-29 (2002); Cooper v. Oklahoma, 517 U.S. 348, 364 (1996); Godinez v. Moran, 509 U.S. 389, 397 n.7 (1993); Parke v. Raley, 506 U.S. 20, 28-29 (1992). This rule emerged in Boykin v. Alabama, 395 U.S. 238 (1969), in which the Court held that because a guilty plea involves a waiver of constitutional rights, including the right against compelled self-incrimination, there must be affirmative evidence of a knowing, intelligent, and voluntary waiver in the record. Id. at 242-43.[/tippy] but add that an accused can nevertheless make a knowing, voluntary, and intelligent decision to seek the advantages of a negotiated disposition despite the pressure created by a potentially greater sentence if the accused asserts the right to trial. [tippy title=”76″ header=”off”]See, e.g., United States v. Mezzanatto, 513 U.S. 196, 209-10 (1995); Alabama v. Smith, 490 U.S. 794, 802-03 (1989); Corbitt v. New Jersey, 439 U.S. 212, 225 (1978); Bordenkircher v. Hayes, 434 U.S. 357, 363-64 (1978); Chaffin v. Stynchcombe, 412 U.S. 17, 30-32 (1973); Brady v. United States, 397 U.S. 742, 749-55 (1970).[/tippy] Indeed, even if a plea bargain ultimately turns out to be a poor deal, the waiver is still considered valid. [tippy title=”77″ header=”off”]See, e.g., Bradshaw v. Stumpf, 545 U.S. 175, 186 (2005); Mabry v. Johnson, 467 U.S. 504, 508 (1984); Brady, 397 U.S. at 757.[/tippy] Similarly, when a defendant elects to waive his right to remain silent and testify at trial, he does so under the threat that the prosecution’s case, if left unrebutted, will likely result in conviction. This kind of pressure has also never been thought to amount to a violation of the Fifth Amendment. Rather, “it is not thought inconsistent with the enlightened administration of criminal justice to require the defendant to weigh such pros and cons in deciding whether to testify.” [tippy title=”78″ header=”off”]McGautha v. California, 402 U.S. 183, 215 (1971). Although some cases, in rather conclusory terms, characterize the pressure that a defendant experiences when deciding whether to testify as something other than “compulsion” within the meaning of the Fifth Amendment, see, e.g., Ohio Adult Parole Auth. v. Woodard, 523 U.S. 272, 286-87 (1998); Williams v. Florida, 399 U.S. 78, 83-85 (1970); that view is strikingly unpersuasive. As we have seen, Bram properly concluded that a suspect facing the threat of criminal prosecution if he does not speak is subjected to compulsion within the meaning of the Fifth Amendment. A view that denies the existence of similar compulsion during a criminal trial–when the threat of conviction is even closer at hand–comports with neither Bram, the historically understood meaning of compulsion, or common sense.[/tippy] Similarly, a suspect asked to waive his right to remain silent during custodial interrogation may decide to do so even though he faces the threat of a criminal prosecution.
It follows that Miranda warnings provide the ingredients for a valid waiver of Fifth Amendment rights; in this sense as well, Miranda worked no innovation, but merely applied settled law. Long before Miranda, it had been settled that “‘courts indulge every reasonable presumption against waiver’ of fundamental constitutional rights and . . . ‘do not presume acquiescence in the loss of fundamental rights.”’ [tippy title=”79″ header=”off”]Johnson v. Zerbst, 304 U.S. 458, 464 (1938) (citations omitted) (quoting Aetna Ins. Co. v. Kennedy, 301 U.S. 389, 393 (1937), and Ohio Bell Tel. Co. v. Public Utilities Comm’n, 301 U.S. 292, 307 (1937)).[/tippy] It had been equally settled that a valid waiver is “an intentional relinquishment or abandonment of a known right or privilege.” [tippy title=”80″ header=”off”]Id.[/tippy] Miranda applied these rules as it fashioned its warnings in order to guarantee that suspects receive the information necessary to make a knowing and intelligent waiver of Fifth Amendment rights. [tippy title=”81″ header=”off”]384 U.S. at 444, 475.[/tippy]
Miranda requires that an individual in custody be advised of his right to remain silent, that anything he says can be used against him, and of his right to have counsel present during interrogation. [tippy title=”82″ header=”off”]Id. at 444, 467-72.[/tippy] These rights are nowhere to be found in the Fifth Amendment itself, but this advice ensures that a defendant understands that he has no obligation to participate in custodial interrogation. The right to counsel, in particular, informs the suspect that if he wishes expert advice as he assesses whether to participate in custodial interrogation, it may be had. Advising the suspect of the availability of expert legal advice clearly provides a far more comprehensive offer of aid in assessing the risks of submitting to interrogation than any formulaic warning that the Court could have advised. The right to counsel is perhaps the most debatable of the Miranda rights, since that right seems to inhere in the Sixth and not the Fifth Amendment. [tippy title=”83″ header=”off”]See, e.g., Scott W. Howe, The Troubling Influence of Equality in Constitutional Criminal Procedure: From Brown to Miranda, Furman, and Beyond, 54 Vand. L. Rev. 359, 397-403, 434-38 (2001).[/tippy] But if one rigorously applies the rule that indulges every reasonable presumption against waiver, it is quite defensible to conclude that suspects cannot be expected to make knowing and intelligent decisions if they are unaware of the availability of expert legal advice. [tippy title=”84″ header=”off”]Precisely because the Miranda right to counsel can be defended only based upon a rather generous presumption against a knowing and intelligent waiver, it may be that there are some circumstances, such as interrogation overseas where competent counsel is not readily available, in which this rule for waiver becomes overgenerous and should be abandoned. For additional consideration of the applicability of Miranda to interrogation overseas, see M.K.B. Darmer, Beyond Bin Laden and Lindh: Confessions Law in an Age of Terrorism, 12 Cornell J.L. & Pub. Pol’y 319, 345-54 (2003); M.K.B. Darmer, Lessons from the Lindh Case: Public Safety and the Fifth Amendment, 68 Brook. L. Rev. 241, 271-86 (2002); Mark A. Godsey, Miranda’s Final Frontier–The International Arena: A Critical Analysis of United States v. Bin Laden and a Proposal for a New Miranda Exception Abroad, 51 Duke L.J. 1703, 1770-80 (2002).[/tippy]
Once advised of these rights, moreover, a suspect’s decision to submit to custodial interrogation is properly characterized as an intentional relinquishment of a known right. [tippy title=”85″ header=”off”]See North Carolina v. Butler, 441 U.S. 369, 372-75 (1979).[/tippy] The Fifth Amendment, after all, secures no more than a right to be free from compelled self-incrimination. The Miranda warnings ensure that the suspect knows that he need not participate in interrogation and is being asked to surrender that right. For purposes of the Fifth Amendment, no more is required to obtain a valid waiver, at least under the settled principles of waiver law that Miranda applied. As we have seen, a waiver is valid as long as a suspect intentionally relinquishes a known right, and the Miranda warnings ensure that a suspect knows that he has a right to remain silent and is facing a decision whether to relinquish his right when he is asked to waive. Thus, while Miranda does not eliminate the compulsion inherent in custodial interrogation, it instead produces a valid waiver of the right to be free from that compulsion.
Understanding Miranda warnings as a recipe for valid waiver explains as well the Court’s invitation for “potential alternatives for protecting the privilege” that are “at least as effective in apprising accused persons of their right of silence and in assuring a continuous opportunity to exercise it . . . .” [tippy title=”86″ header=”off”]384 U.S. at 467.[/tippy] There is no single form of words necessary for a valid waiver; any advice that enables a suspect to make a knowing and intelligent decision about Fifth Amendment rights will comport with constitutional standards. [tippy title=”87″ header=”off”]Indeed, the Court has tolerated some variation in the language of the required warnings that is not thought to dilute their efficacy. See Duckworth v. Eagan, 492 U.S. 195 (1989); California v. Prysock, 453 U.S. 355 (1981) (per curiam).[/tippy] That does not make the warnings prophylactic; it makes them one among many alternatives that could produce a valid waiver of constitutional rights. [tippy title=”88″ header=”off”]Thus, in some sense the warnings amount to a “safe harbor.” See Michael C. Dorf & Barry Friedman, Shared Constitutional Interpretation, 2000 Sup. Ct. Rev. 61, 81-85. The safe harbor, however, is within the law of waiver, and is not based on an absence of compulsion. The warnings do not eliminate the compulsion inherent in custodial interrogation by creating a safe harbor of voluntariness; if they did, then the Court would have held that the warnings without more could produce an admissible statement. To the contrary, the Court made plain that even with the requisite warnings, a statement is inadmissible unless the government can discharge its burden of proving a valid waiver. See 384 U.S. at 475-77. See also Tague v. Louisiana, 444 U.S. 469, 470-71 (1980) (per curiam); Butler, 441 U.S. at 373.[/tippy]
To be sure, many interrogators are adept at using some combination of threats and inducements to convince suspects to submit to interrogation–and even to confess–regardless of whether it was in the suspect’s interest to do so. I certainly tried to do just that when I was a prosecutor. Thus, I freely acknowledge that if they are to protect their own legal interests most efficaciously, suspects would probably be well advised to consult with an attorney before deciding to submit to interrogation. [tippy title=”89″ header=”off”]Even this rule of thumb is not unqualified. In my experience, both prosecutors and judges make a special effort to acknowledge cooperation that was offered from the start, without need of counsel or a negotiated arrangement. It is, in my view, at least a modest overstatement to claim that a suspect is never well-advised to cooperate with the authorities during uncounseled custodial interrogation.[/tippy] It is also probably true that many suspects have undeserved confidence in their ability to talk their way out of their troubles. [tippy title=”90″ header=”off”]See Thomas, supra note 30, at 1106-12. There is at least some empirical evidence to support this view. Using volunteer subjects under laboratory conditions, Saul Kassin and Rebecca Norwick found that the most likely reason simulated suspects gave for waiving their Miranda rights was their belief that they could convince their interrogators of their innocence. See Saul M. Kassin & Rebecca J. Norwick, Why People Waive their Miranda Rights: The Power of Innocence, 28 Law & Hum. Behav. 211 (2004). Although caution is necessary in treating laboratory experiments involving volunteers who face stakes that are much lower than those in the criminal justice system as indicative of the behavior of actual suspects, this supposition is corroborated by evidence that unsuccessful experience with the criminal justice system makes suspects less likely to submit to interrogation. Leo found that suspects with felony records were almost four times more likely to invoke their rights during custodial interrogation than suspects with no record, and almost three times more likely to invoke than suspects with only a misdemeanor record. See Richard A. Leo, Inside the Interrogation Room, 86 J. Crim. L. & Criminology 266, 286-87 (1996). It would make sense that if suspects’ belief in their ability to convince interrogators of their innocence motivates most waivers, then suspects who have learned that they are unable to persuade interrogators of their innocence are less likely to waive. My own experience was that suspects had a wide variety of reasons for wanting to talk with interrogators, but a belief in their ability to talk their way out of trouble was certainly one of the most common motives that I perceived.[/tippy] Perhaps it is even true that most persons in custody feel intimidated and are reluctant to invoke their rights. [tippy title=”91″ header=”off”]See, e.g., Stephen J. Schulhofer, Confessions and the Court, 79 Mich. L. Rev. 865, 880-82 (1981) (reviewing Yale Kamisar, Police Interrogation and Confessions: Essays in Law and Policy (1980)).[/tippy] But once a suspect has waived his right to be free from compelled self-incrimination, an insistence on further safeguards is vulnerable to Justice Scalia’s charge that under the Fifth Amendment, the Court may not properly concern itself with “[p]reventing foolish (rather than compelled) confessions.” [tippy title=”92″ header=”off”]Dickerson v. United States, 530 U.S. 428, 449 (2000) (Scalia, J., dissenting).[/tippy]
In particular, when a suspect waives his rights under Miranda, the suspect has knowingly and intentionally decided to make his own assessment of the risks and benefits of submitting to custodial interrogation without expert advice. A decision to proceed with interrogation without counsel under such circumstances may be foolish, but it nevertheless satisfies the settled standards for a valid waiver. [tippy title=”93″ header=”off”]This point also demonstrates the non-prophylactic character of the secondary protection offered by the rule of Edwards v. Arizona, 451 U.S. 477 (1981), that when a suspect invokes the right to counsel recognized in Miranda, the authorities may not subsequently communicate with the suspect unless communication is initiated by the suspect. Id. at 484-85. Once an accused has indicated that he doubts his own ability to make a waiver decision in the absence of expert assistance, the presumption against waiver amply supports the conclusion that the suspect should not be pressed to make such a decision. The suspect’s invocation is itself powerful evidence that the suspect cannot make a knowing and intelligent waiver decision in the absence of counsel. As the Court has put it, “the accused having expressed his own view that he is not competent to deal with the authorities without legal advice, a later decision at the authorities’ insistence to make a statement without counsel’s presence may properly be viewed with skepticism.” Arizona v. Roberson, 486 U.S. 675, 681 (1988) (quoting Michigan v. Mosley, 423 U.S. 96, 110 n.2 (1975) (White, J., concurring in the result)). When a suspect invokes only the right to silence but expresses no desire for expert assistance in assessing his options, there is no similar basis for doubting the suspect’s knowing and intelligent desire to make a waiver decision without advice, and hence a different standard is in order. See Roberson, 486 U.S. at 683; Mosley, 423 U.S. at 109-10 (White, J., concurring in the result).[/tippy] As we have seen, the traditional standard for waiver merely requires an intentional relinquishment of a known right. Thus, a valid waiver does not require that the defendant be able to correctly assess his interests or free himself from the pressures created by a potential prosecution or conviction; it only requires that a defendant make an intentional decision to surrender a right of which he is aware. [tippy title=”94″ header=”off”]See, e.g., Iowa v. Tovar, 541 U.S. 77, 90-92 (2004); Patterson v. Illinois, 487 U.S. 285, 292-93 (1988); Fare v. Michael C., 442 U.S. 707, 724-25 (1979).[/tippy] A valid waiver requires that the defendant knows he has an identified right and intends to waive it; but the defendant need “not know the specific detailed consequences of invoking it.” [tippy title=”95″ header=”off”]United States v. Ruiz, 536 U.S. 622, 629 (2002).[/tippy] Indeed, the Court has “never read the Constitution to require that the police supply a suspect with a flow of information to help him calibrate his self-interest in deciding whether to speak or stand by his rights.” [tippy title=”96″ header=”off”]Moran v. Burbine, 475 U.S. 412, 422 (1986). For example:
The Constitution does not require that a criminal suspect know and understand every possible consequence of a waiver of the Fifth Amendment privilege. The Fifth Amendment’s guarantee is both simpler and more fundamental: A defendant may not be compelled to be a witness against himself in any respect. The Miranda warnings protect this privilege by ensuring that a suspect knows that he may choose not to talk to law enforcement officers, to talk only with counsel present, or to discontinue talking at any time. The Miranda warnings ensure that a waiver of these rights is knowing and intelligent by requiring that the suspect be fully advised of this constitutional privilege, including the critical advice that whatever he chooses to say may be used as evidence against him.
Colorado v. Spring, 479 U.S. 564, 574 (1987) (citations omitted).[/tippy]
Thus, the Miranda warnings rest on a traditional conception of waiver. It may be that most suspects make bad decisions about whether to submit to custodial interrogation, but settled waiver law instructs us that this provides no basis for invalidating a waiver. [tippy title=”97″ header=”off”]See supra text accompanying notes 74-78.[/tippy] Indeed, under the guilty-plea cases, even if the accused and his counsel misapprehend the strength of the prosecution’s case or the availability of defenses, a guilty plea is still considered a valid waiver. [tippy title=”98″ header=”off”]See, e.g., Ruiz, 536 U.S. at 630; United States v. Broce, 488 U.S. 563, 571-72 (1989); McMann v. Richardson, 397 U.S. 759, 769-70 (1970); Brady v. United States, 397 U.S. 742, 756-57 (1970). Thus, although Robert Mosteller has argued that any type of what could be characterized as official deception should be forbidden prior to the time that a waiver is obtained (see Robert P. Mosteller, Police Deception Before Miranda Warnings: The Case for a Per Se Prohibition Against an Entirely Unjustified Practice at the Most Critical Moment, 39 Texas Tech. L. Rev. 1239, 1263-72 (2007)), this approach is difficult to reconcile with traditional principles of waiver. Waiver law asks whether the suspect knew of the rights at issue and intended to surrender them and not whether he was able to accurately assess his interests at the time of the waiver. Deception as to, say, the strength of the evidence against the suspect may lead him to overvalue the benefits of submitting to interrogation, but this type of error has never been thought to impair a waiver. To the contrary, it has long been settled that “the Constitution …. permits a court to accept a guilty plea, with its accompanying waiver of various constitutional rights,despite various forms of misapprehension under which a defendant might labor.” Ruiz, 536 U.S. at 630. Applying these principles, the Court has held that a Miranda waiver is valid even though, prior to obtaining the waiver, federal agents failed to disclose to a suspect arrested on a weapons charge that the true object of the investigation was an unsolved murder in which the arrestee was a suspect. See Colorado v. Spring, 479 U.S. 564, 573-79 (1987). While there was no affirmative misrepresentation in that case, it is difficult to understand why deception by omission has any different consequences for the knowing, intelligent, or voluntary character of a waiver than deception through affirmative misrepresentation..[/tippy] A Miranda waiver is certainly no less valid if the suspect somehow misapprehends his own best interests. To be sure, defendants undertaking plea bargaining usually have counsel at their side, while suspects undergoing custodial interrogation usually do not. [tippy title=”99″ header=”off”]Brady, 397 U.S. at 753-54.[/tippy] Still, Miranda grants suspects subject to custodial interrogation a right to counsel, and we have seen that waiver of this right under Miranda comports with traditional waiver principles. [tippy title=”100″ header=”off”]See supra text accompanying notes 74-96.[/tippy] Nor is there any plausible basis to treat the right to counsel as nonwaivable; indeed, it is now settled that even the Sixth Amendment’s right to counsel can be extrajudicially waived by an uncounseled defendant as long as the waiver has been preceded by Miranda warnings. [tippy title=”101″ header=”off”]Patterson v. Illinois, 487 U.S. 285, 292-97 (1988).[/tippy]
One can disagree with the traditional rules for waiver, but Miranda is faithful to them, and one cannot criticize Miranda on this basis without developing a new and as-yet unprecedented conception of waiver of constitutional rights. [tippy title=”102″ header=”off”]Professor Godsey proposes that in addition to traditional Miranda warnings, suspects be further advised that silence cannot be used against them, that they have a right to stop questioning at any time, and that warnings be readministered during lengthy interrogations. See Godsey, supra note 32, at 810, 813-15. See also Mark Berger, Compromise and Continuity: Miranda Waivers, Confession Admissibility, and the Retention of Interrogation Protections, 49 U. Pitt. L. Rev. 1007, 1054-61 (1988) (contending that waivers should be readministered during lengthy interrogations or when the topic of interrogation shifts). At least some aspects of this proposal are hard to support under current law. As for an admonition that a suspect’s silence cannot be used against him, it is hard to identify this as a right that a suspect must knowingly and intelligently waive when he receives warnings. When Miranda warnings are administered, a suspect is not being asked to waive a right he then enjoys, i.e., not to have silence used against him, because pre-Miranda silence can be used to infer a suspect’s guilt. See Fletcher v. Weir, 455 U.S. 603 (1982) (per curiam); Jenkins v. Anderson, 447 U.S. 231 (1980). There is instead only a right not to have post-Miranda silence used as evidence of guilt. See Doyle v. Ohio, 426 U.S. 610 (1976). It may be that knowledge of the post-Miranda right is useful to a suspect when deciding whether to assert his right to remain silent, but as we have seen, the Supreme Court has never required that a suspect, when deciding whether to invoke a constitutional right, must “know the specific detailed consequences of invoking it.” Ruiz, 536 U.S. at 629. Nor has the Court “require[d] that the police supply a suspect with a flow of information to help him calibrate his self-interest in deciding whether to speak or stand by his rights.” Moran v. Burbine, 475 U.S. 412, 422 (1986). This same point suggests that waiver law does not require that a suspect be permitted to reassess the tactical value of waiver whenever the subject of interrogation shifts or during lengthy interrogations. Moreover, a right to have warnings readministered is difficult to support under traditional waiver law, which, as we have seen, requires only that the decision to waive be knowing and intelligent, not that the suspect be encouraged to revisit that decision periodically. See Wyrick v. Fields, 459 U.S. 42 (1982) (per curiam). Nevertheless, in Miranda the Court stressed that warnings must be an effective means of “apprising accused persons of their right of silence and in assuring a continuous opportunity to exercise it….” 384 U.S. at 467. That might support an additional admonition that a suspect can invoke his right to remain silent at any time, or a need for repeated warnings during lengthy interrogations, although it is worth noting that in Miranda itself the Court did not think that its reasoning required the additional warnings advocated by Professor Godsey. Accordingly, if there is a justification for doctrinal reform in order to support these proposals, it is presumably a prophylactic one. Professor Godsey, however, adduces little empirical support for such reform in the interests of prophylaxis. Although Professor Godsey claims that suspects frequently fail to invoke their Miranda rights because they believe that an assertion of rights will be taken as an admission of guilt, see Godsey, supra note 32, at 793-96, the supporting evidence he cites reveals that the only empirical evidence derived from actual interrogations consists of the interview of a single suspect. See Ofshe & Leo, supra note 34, at 1002. We have seen, however, that other empirical evidence suggests that suspects choose to submit to custodial interrogation because they believe that they can persuade interrogators of their innocence, see supra note 90, and not because they fear that their silence will be used against them, believe that they lack the right to stop questioning, or forget their rights during the course of lengthy interrogation. Moreover, Professor Godsey’s proposal might actually encourage suspects to waive their rights imprudently in the belief that they will be free to retract the waiver. In the absence of empirical study, it is hard to know whether this additional warning would make suspects’ decisions more or less considered. Eugene Milhizer goes even further, arguing that in addition to advising a suspect of his right to stop questioning at any time, the warnings should be reformulated to advise the suspects of the potential benefits of truthful cooperation. See Eugene R. Milhizer, Rethinking Police Interrogation: Encouraging Reliable Confessions While Respecting Suspects’ Dignity, 41 Val. U. L. Rev. 1, 99-107 (2006). We have seen, however, that requiring that a suspect not only know his rights but receive helpful information in order to assess his own interests would require a rather dramatic reformulation of the law of waiver. In any event, if there is an argument for additional prophylactic protection of a suspect’s ability to engage in careful balancing of the tactical considerations relating to interrogation, his ability to invoke his right to have counsel present during interrogation would seemingly be the most efficacious means of protecting that interest.[/tippy] After all, the Fifth Amendment protects suspects against compulsion, not improvidence. If a suspect is willing to submit to interrogation under compulsive circumstances, it is unclear at best why the Fifth Amendment should be thought to forbid him from doing so merely because the waiver is likely to be improvident. Whatever one might be protecting by insisting on some new and more demanding test for waiver, it certainly is not the right to be free from compelled self-incrimination. Thus, when the Court explains that it has “never read the Constitution to require that the police supply a suspect with a flow of information to help him calibrate his self-interest in deciding whether to speak or stand by his rights,” [tippy title=”103″ header=”off”]Burbine, 475 U.S. at 422.[/tippy] it is far from apparent what is wrong with that conception of waiver, at least for purposes of protecting Fifth Amendment rights. [tippy title=”104″ header=”off”]For helpful elaboration on this point, see Berger, supra note 102, at 1042-51. I consider the case for greater regulation of interrogation under the Due Process Clause in Part II infra.[/tippy]
Thus, the non-prophylactic account of Miranda that I have offered involves rather settled Fifth Amendment and waiver law. As I have demonstrated, Miranda rests on a conception of compulsion traceable to Bram, and an equally settled conception of waiver. [tippy title=”105″ header=”off”]A non-prophylactic conception of Miranda would not even necessarily require alteration of the scope of the Miranda exclusionary rule, at least when it comes to impeachment evidence. When an unwarned statement is sufficiently reliable to have probative value as impeachment, it is far from clear that the government has “compelled” a defendant to become a witness against himself when the statement becomes admissible only because the defendant has elected to testify inconsistently with the statement. Indeed, before the Court began referring to Miranda as a prophylactic rule, it held that unwarned but reliable statements were admissible for impeachment purposes. See Harris v. New York, 401 U.S. 222 (1971). For a lengthier argument in favor of this conclusion, see Donald Dripps, Is the Miranda Caselaw Really Inconsistent? A Proposed Fifth Amendment Synthesis, 17 Const. Comment. 19, 27-43 (2000). The argument for derivative use of unwarned statements in the prosecution’s case-in-chief, however, is considerably stronger. See, e.g., Yale Kamisar, Response: On the “Fruits” of Miranda Violations, Coerced Confessions, and Compelled Testimony, 93 Mich. L. Rev. 929 (1995). Indeed, on the originalist view that the term “witness” includes anyone who provides the prosecution with physical or documentary evidence, even when no testimonial use is made of the act of production and only the physical or documentary evidence itself is introduced at trial, see United States v. Hubbell, 530 U.S. 27, 49-56 (2000) (Thomas, J., concurring), the Fifth Amendment itself would prohibit the introduction of the nontestimonial fruits of any compelled provision of evidence to the prosecution.[/tippy] Treating Miranda as a form of prophylactic constitutional law, in contrast, is a far more radical approach than that outlined here, and far less faithful to Miranda itself. [tippy title=”106″ header=”off”]In Miranda itself, the Court acknowledged that it “might not find the defendants’ statements [in the four consolidated cases at bar] to have been involuntary in traditional terms.” 384 U.S. at 457. This should not be taken as an acknowledgement that the Court was departing from traditional Fifth Amendment standards. The voluntariness test is anchored in the Due Process Clause, which provides protection against the admission into evidence of an involuntary confession separate and apart from the Fifth Amendment right against compelled self-incrimination. See Withrow v. Williams, 507 U.S. 680, 688-89, 693-94 (1993); Miller v. Fenton, 474 U.S. 104, 109-10 (1985). But by the time of Miranda, as we have seen, the Fifth Amendment had become applicable to state prosecutions. Thus, “involuntary in traditional terms” refers to the due process test.[/tippy] It is remarkable that courts and commentators alike have so readily accepted the reinterpretation of Miranda as prophylactic that began in Tucker–in other words, that custodial interrogation does not inherently involve compulsion–without any apparent recognition of the tension between such an assumption and both Bram and the concept of “compulsion” under the Fifth Amendment.
II. The Constitution Is Not Perfect: The Flawed Case for Additional Constitutional Regulation of Interrogation
The available empirical evidence on the implementation of Miranda, [tippy title=”107″ header=”off”]In invoking empirical evidence, I do not mean to enter the debate over the effects of Miranda on rates of confession and crime-solving. For a useful summary of the evidence on that point, see George C. Thomas III & Richard A. Leo, The Effects of Miranda v. Arizona: “Embedded” in our National Culture?, in 29 Crime and Justice: A Review of Research 203, 232-45 (Michael Tonry ed., 2002). In my view, this is one of those debates that turns to a critical degree on whether Miranda is prophylactic. If it is, then consideration of its costs and benefits is amply warranted, as Professor Strauss has argued, see Strauss, supra note 18, at 967; if not, then even if Miranda has adversely affected law enforcement, that is a price that the Constitution requires be paid.[/tippy] although limited, indicates that suspects subjected to custodial interrogation invoke their right to halt the interrogation about twenty percent of the time, with the vast majority of invocations occurring at the point at which warnings are administered. [tippy title=”108″ header=”off”]Cassell & Hayman, supra note 59, at 859-60; Leo, supra note 32, at 653. There were a number of studies in the immediate wake of Miranda that also reflected about a twenty percent invocation rate. See Paul G. Cassell, Miranda’s Social Costs: An Empirical Reassessment, 90 Nw. U. L. Rev. 387, 495-96 (1996). There was a puzzling lack of further empirical study of Miranda until the mid-1990s. Cassell & Hayman, supra note 59, at 843-49. As Professor Schulhofer notes, there is cause for concern about the reliability of the immediate post-Miranda studies since they cover a period in which Miranda was still novel and the police had not yet adapted their tactics to its commands. See Stephen J. Schulhofer, Miranda’s Practical Effect: Substantial Benefits and Vanishingly Small Social Costs, 90 Nw. U. L. Rev. 500, 506-10 (1996). Nevertheless, the consistency of the invocation rate in the post-Miranda and later studies suggests that at least on this point, the earlier studies may be reliable indications of Miranda’s current effects. The most recent study of juveniles also found about a twenty percent invocation rate, see Barry C. Feld, Juveniles’ Competence to Exercise Miranda Rights: An Empirical Study of Policy and Practice, 91 Minn. L. Rev. 26, 82 (2006), although an earlier study of juveniles had found that they invoke their rights at a rate of less than ten percent. See Thomas Grisso, Juveniles’ Waiver of Rights: Legal and Psychological Competence 35-36 (1981).[/tippy] One of the most prominent members of the Miranda-is-a-failure school, William Stuntz, thinks that this evidence means that Miranda fails to effectively regulate interrogation because the rate of invocation is so low, and because invocations are so concentrated at the time warnings are given, that Miranda provides too little regulatory bite over the course of the subsequent interrogation. [tippy title=”109″ header=”off”]See Stuntz, supra note 30, at 986-92.[/tippy] He contends that Miranda’s approach is to combat police overreaching by relying on suspects to invoke their rights, but the data, he argues, demonstrates that “suspects cannot do the kind of sorting that Miranda doctrine calls for.” [tippy title=”110″ header=”off”]Id. at 991.[/tippy]
Professor Stuntz’s critique is puzzling. As we have seen, the only “sorting” that Miranda expects of suspects is that they will knowingly and intelligently decide whether to waive their Fifth Amendment rights, and under settled waiver principles, suspects facing interrogation are perfectly competent to engage in such “sorting.” In any event, it is unclear what kind of abuses Professor Stuntz thinks are going on during post-warning custodial interrogation that cry out for additional regulation. The Fifth Amendment, as we have also seen, is satisfied by a valid waiver under Miranda. To be sure, a suspect who has waived Miranda rights has not also waived his analytically distinct due process right to exclude from evidence an involuntary confession. [tippy title=”111″ header=”off”]See, e.g., Withrow v. Williams, 507 U.S. 680, 688-89, 693-94 (1993); Miller v. Fenton, 474 U.S. 104, 109-10 (1985); Mincey v. Arizona, 437 U.S. 385, 397-98 (1978).[/tippy] But Richard Leo’s study, which Professor Stuntz himself labels as “the most thorough to date,” [tippy title=”112″ header=”off”]Stuntz, supra note 30, at 990.[/tippy] establishes not only that levels of compliance with Miranda are high, [tippy title=”113″ header=”off”]See Leo, supra note 90, at 275-76. Accord Cassell & Hayman, supra note 59, at 888-90; Feld, supra note 108, at 71; George C. Thomas III, Stories about Miranda, 102 Mich. L. Rev. 1959, 1975 (2004). There is, however, some evidence that some police departments have engaged in unwarned questioning as a matter of policy, providing warnings only after incriminating statements are obtained and then inducing the suspects to repeat those statements in compliance with Miranda. See Weisselberg, supra note 16, at 136-39. It is difficult to characterize this as police misconduct, since this tactic was seemingly sanctioned in Oregon v. Elstad, 470 U.S. 298, 314 (1985); indeed, the author of the Elstad opinion (and three other Justices of the Supreme Court) thought that such an approach had been condoned by that decision. See Missouri v. Seibert, 542 U.S. 600, 627-29 (2004) (O’Connor, J., dissenting). In any event, this practice is likely to come to a halt in light of the conclusion of a majority of the Court in Seibert that deliberate use of two-stage questioning with warnings given only after an incriminating statement is made will lead to suppression. See id. at 614 (plurality opinion); id. at 620-22 (Kennedy, J., concurring in the judgment). See also Steven D. Clymer, Are Police Free to Disregard Miranda?, 112 Yale L.J. 447, 545-47 (2002).[/tippy] but also that the incidence of coercion during custodial interrogation is low, only about two percent. [tippy title=”114″ header=”off”]See Leo, supra note 90, at 282-84. The other leading study in recent years of Miranda’s implementation reached a similar conclusion. See Cassell & Hayman, supra note 59, at 888-94, 920. Although the interrogators in the Leo study knew they were being observed, Leo convincingly explains why that ultimately did not skew the data. See Leo, supra note 90, at 270-72. Barry Feld’s recent study of juvenile interrogations also found no evidence of coercion. See Feld, supra note 108, at 70-90. Professor Leo’s work is full of anecdotes describing what he regards as police overreaching, although the anecdotes are unaccompanied by any statistics establishing that these practices occur with frequency. See, e.g., Leo & White, supra note 45, at 431-50; Ofshe & Leo, supra note 34, at 1001-14.[/tippy] Even a study of judicial opinions available on Westlaw involving challenges to confessions–hardly a random sample–found no reason to doubt Professor Leo’s findings on coercion. [tippy title=”115″ header=”off”]See Thomas, supra note 113, at 1962, 1980-95.[/tippy]
My own experience with custodial interrogation is consistent with this data, but does not lead me to Professor Stuntz’s conclusion. My rough guess is that perhaps twenty percent of the custodial interrogations in which I participated as a prosecutor produced invocations, although my memory is vague; it could have been as high as one-third. But I very clearly remember that in every case, invocations occurred when the warnings were administered. In my experience, that pattern reflects the potency of the warnings–as I still vividly recall, the moment at which I administered warnings and solicited a waiver was the point in the process where I always felt a striking loss of control. At that point in the process, I could not advocate for cooperation; I had to stop and let the suspect ponder both the warnings and my request for a waiver. Small wonder that invocations cluster at that point. If suspects attached little meaning or significance to Miranda warnings, or if interrogators were somehow able to deemphasize them, invocations would occur in a more random pattern throughout the course of questioning.
But put all this aside. Even crediting Professor Stuntz’s fears, Miranda is not properly labeled a failure. Miranda states only a single objective–to achieve compliance with the Fifth Amendment by ensuring that custodial interrogations occur only upon a valid waiver of the Fifth Amendment right to be free from compelled self-incrimination. [tippy title=”116″ header=”off”]See 384 U.S. at 467, 478-79.[/tippy] No other objective is properly open to a court under the Fifth Amendment, which forbids, as we have seen, only compelled confessions, and not improvident ones. Nothing in Professor Stuntz’s account explains why Miranda fails on this score. As I demonstrate above, Miranda produces valid waivers under long-settled conceptions of waiver. The waivers may often be foolish, but they are made with full awareness of the rights foregone.
There is, accordingly, no case to be made against Miranda on its own terms. At best, Professor Stuntz’s critique supports an argument that there is some sort of police abuse during post-warning interrogation that requires additional regulation, but such an argument is not based on the inherent compulsion in custodial interrogation. Demanding better post-waiver regulation of interrogation under the Due Process Clause is easier said than done. Consider videotaping interrogations, perhaps the most common recommendation offered by the Miranda-is-a-failure camp. [tippy title=”117″ header=”off”]See supra note 32.[/tippy] There is little doctrinal support for constitutionally mandated videotaping; the Court has held that the Due Process Clause does not require that the prosecution create or preserve evidence merely because it is potentially exculpatory. [tippy title=”118″ header=”off”]See Arizona v. Youngblood, 488 U.S. 51, 56-59 (1988); California v. Trombetta, 467 U.S. 479, 485-90 (1984).[/tippy] As the Court has written, it is “unwilling[ ] to . . . impos[e] on the police an undifferentiated and absolute duty to retain and to preserve all material that might be of conceivable evidentiary significance” absent a showing that “the police themselves by their conduct indicate that the evidence could form a basis for exonerating the defendant.” [tippy title=”119″ header=”off”]Youngblood, 488 U.S. at 58.[/tippy] Thus, at least when a law enforcement agency’s standard policy involves no taping of interrogation, there is little doctrinal basis for a constitutional attack.
Perhaps doctrinal innovation would be warranted if videotaping yielded clear benefits, but it is unclear what videotaping is expected to accomplish. We have seen that there is little empirical evidence that noncompliance with Miranda or coercion during interrogation is common. [tippy title=”120″ header=”off”]See supra text accompanying note 113.[/tippy] There is also little evidence that credibility disputes are common in litigation about custodial interrogation; George Thomas’s review of custodial interrogation cases available on Westlaw disclosed that only four percent turned on the credibility of the participants. [tippy title=”121″ header=”off”]See Thomas, supra note 113, at 1982-83.[/tippy] Thus, there is little reason to believe that videotaping is likely to improve Miranda compliance or enhance the reliability of factfinding–there is simply not much empirical evidence that either is a significant problem. [tippy title=”122″ header=”off”]An additional argument made in support of videotaping is that it will improve assessments of the voluntariness of confessions because it will enable the trier of fact to determine whether the suspect volunteered critical facts about the crime that were not publicly known or whether those facts were in fact supplied by interrogators. Leo, et al., supra note 31, at 511. But again, absent a rule forbidding the police from utilizing such facts during interrogation, it is unclear that evidence of the use of such tactics will have much of an effect. In violent crimes, for example, I would sometimes confront a suspect with particularly grisly aspects of the physical evidence in an effort to shake the suspect and suggest that laboratory analysis would ultimately link him to the crime. In my experience, juries are frequently sympathetic to this type of explanation for what might initially seem an overly aggressive interrogation tactic.[/tippy] Nor is videotaping likely to deter abuse even if it is infrequent–a truly sadistic officer will indulge himself in a location without cameras, and would be prepared to lie about what he has done after the fact.
Thus, we cannot expect videotaping to curb what are already deemed abuses under current law, which, in any event, have not been shown to be common. Moreover, in the absence of substantive regulation of interrogation techniques, it is unclear how videotaping is expected to lead to more effective regulation of interrogation. In Frazier v. Cupp, [tippy title=”123″ header=”off”]394 U.S. 731 (1969).[/tippy] for example, the interrogation was taped, [tippy title=”124″ header=”off”]Id. at 737.[/tippy] and yet that did not stop the police from deceiving the suspect by falsely claiming that his accomplice had confessed. [tippy title=”125″ header=”off”]Id. at 737-39. Similarly, although Professor Feld advocates videotaping interrogations, the videotapes that he observed did not prevent what he regarded as improper, if subtle, efforts to influence juveniles to waive their rights similar to those observed by Professor Leo. See Feld, supra note 108, at 28, 90-99. The same was true of Professor Leo’s study, in which he was actually present during interrogation, or observed videotaped interrogations, and nevertheless observed a variety of tactics that he viewed as unduly coercive. See Leo, supra note 90, at 269-72, 292-303.[/tippy] Thus, while a common claim of the Miranda-is-a-failure camp is that police deception during interrogation should be curbed, [tippy title=”126″ header=”off”]See supra note 35.[/tippy] videotaping will not lead to that objective. Given the political pressure to apprehend lawbreakers, the police can be expected to employ every lawful means for obtaining confessions, and therefore videotaping will not likely deter tactics that the courts condone. [tippy title=”127″ header=”off”]Although videotaping advocates do not make the point, perhaps they believe that police will engage in self-regulation if they are concerned that juries will react adversely to what they perceive to be overly aggressive interrogation techniques. No empirical evidence has yet surfaced of such an effect, however, in jurisdictions that perform videotaping. As a prosecutor, I would have advised investigators to continue to utilize all lawful interrogation techniques during videotaped confessions, while seeking appropriate instructions informing the jury that the police used only lawful interrogation techniques during the videotaped session. It is therefore unclear at best that in such circumstances videotaping would have any predictable and significant regulatory bite, especially given the reality that juries are likely to approve of the use of such tactics when they are told they are lawful means to solve serious crimes.[/tippy] One may dislike the kind of tactics recommended by the manuals that encourage a suspect to believe that cooperation is in his interest, but absent substantive reform limiting the kinds of appeals that interrogators may make–before or after soliciting waivers–videotaping alone will not change interrogators’ behavior.
The same problem infects proposals to require judges to undertake some sort of stricter review of the reliability and voluntariness of confessions. [tippy title=”128″ header=”off”]See supra note 34.[/tippy] The advocates of this approach frequently invoke the Central Park Jogger case to support their proposals for independent judicial review of the reliability of confessions. [tippy title=”129″ header=”off”]Davies, supra note 34, at 230-52; Leo, et al., supra note 31, at 479-87.[/tippy] In that case, three of the five defendants were arrested near the scene of a brutal assault and rape shortly after it occurred, at a time at which roving bands of youths had committed multiple assaults in the park; five suspects then confessed to participating in the assault while denying personal involvement in the rape; each confession provided details about the involvement of the other defendants; and three of the confessions were videotaped in the presence of the suspects’ parents. [tippy title=”130″ header=”off”]See Timothy Sullivan, Unequal Verdicts: The Central Park Jogger Trials 23-47, 84-85 (1992); Davies, supra note 34, at 215-16; Leo et al., supra note 31, at 479-82.[/tippy] Nevertheless, the confessions were false; DNA evidence, tested years later, identified as the rapist another individual not linked to any of the defendants. [tippy title=”131″ header=”off”]See Davies, supra note 34, at 220-22; Leo et al., supra note 31, at 482-84.[/tippy]
The Central Park Jogger case is an uncertain poster child for more muscular judicial review of confessions. To be sure, there were inconsistencies in the Central Park Jogger confessions as each offender accused others of playing a principal role, [tippy title=”132″ header=”off”]See Davies, supra note 34, at 244; Leo et al., supra note 31, at 536-37.[/tippy] but in cases involving joint action this is commonplace; in their confessions, offenders frequently endeavor to minimize their own role. [tippy title=”133″ header=”off”]See, e.g., Lilly v. Virginia, 527 U.S. 116, 138-39 (1999); Lee v. Illinois, 476 U.S. 530, 544-45 (1986).[/tippy] Indeed, the advocates of these reforms make no claim that confessions accompanied by these kinds of inconsistencies are usually unreliable; and they make no effort as well to identify anything approximating an error rate for confessions of this type. In the Central Park Jogger case, for example, given that three defendants had been arrested near the scene of the crime, and that all five had been placed there by other witnesses, perhaps only with the benefit of hindsight can one confidently say that a judge should have deemed the confessions insufficiently reliable to be put before a jury.
Thus, the proposals for more searching reliability review come with all the hazards of regulation by anecdote; it is unclear whether the Central Park Jogger case reflects a systemic problem or is a bizarre outlier–perhaps borne of the fact that the defendants may well have participated in other assaults in the park that night and therefore were willing to admit a role in an assault without understanding their vicarious liability for the rape itself. For her part, one of the principal advocates of searching judicial scrutiny of confessions, Sharon Davies, does not propose a per se rule that interlocking but inconsistent confessions of persons found near the scene of a crime are never admissible absent some additional corroboration; she instead proposes a non-exclusive list of at least ten factors. [tippy title=”134″ header=”off”]See Davies, supra note 34, at 242-43. Professor Stuntz, for his part, is unable to identify any set of criteria to govern judicial review of interrogation techniques. See Stuntz, supra note 30, at 995-98. Professor Kamisar has similarly argued that custodial interrogation ought to take place in the presence of a judicial officer with the power to regulate the process, but is similarly unable to specify the criteria that should govern such regulation. See Kamisar, supra note 34.[/tippy] Ten-factor tests are unlikely, however, to produce predictable outcomes. Professor Davies makes little effort to defend her ten-factor test as an effective means of regulation in itself; instead she speculates that judges will do a better job than juries because their training and experience gives them greater expertise in evaluating the reliability of confessions. [tippy title=”135″ header=”off”]See Davies, supra note 34, at 250-52.[/tippy] She identifies no empirical evidence to support this claim, however, and with reason. For more than four decades, the Due Process Clause has been understood to require judges to make an independent finding of voluntariness before permitting a confession to be presented to a jury. [tippy title=”136″ header=”off”]See Jackson v. Denno, 378 U.S. 368, 387 (1964).[/tippy] If judicial training and experience enabled judges to identify confessions that are the result of undue police pressure, then we should have expected that judges would have learned long ago how to smoke out confessions that were the likely result of police pressure under the rubric of voluntariness. After all, the requirement that judges make an independent finding of voluntariness permits as searching a review as is necessary to assure the court that a confession has not been induced by police overreaching or an overborne will. [tippy title=”137″ header=”off”]While the voluntariness test, as framed by the Court, requires an inquiry into police overreaching rather then reliability per se, see Colorado v. Connelly, 479 U.S. 157, 167 (1986), the voluntariness test is broad enough to reach an unwarranted effort by the police to induce a suspect willing only to admit to a relatively minor crime to link himself to a far more serious one. As the Court has put it: “the admissibility of a confession turns as much on whether the techniques for extracting the statements, as applied to this suspect, are compatible with a system that presumes innocence and assures that a conviction will not be secured by inquisitorial means as on whether the defendant’s will was in fact overborne.” Miller v. Fenton, 474 U.S. 104, 116 (1985).[/tippy] The fact that judges seem not to have developed what Professor Davies regards as reliable methods for screening out police-induced false confessions suggests that judicial expertise is not likely to solve the problem with which she is concerned.
Professor Leo, in addition to advocating mandatory videotaping if feasible (although the Central Park Jogger case would seem to argue against the efficacy of videotaping), would require a court to weigh three factors: whether the confession contains nonpublic information that can be independently verified; whether the confession led the authorities to discover new evidence; and whether the suspect’s confession is consistent with the objective evidence. [tippy title=”138″ header=”off”]See Leo et al., supra note 31, at 531. George Thomas has advanced proposal along somewhat the same lines to exclude confessions that are a product of interrogation techniques that involve both high pressure and large incentives to confess. See George C. Thomas III, Regulating Police Deception During Interrogation, 39 Texas Tech. L. Rev. 1293, 1298-1317 (2007). This proposal, like that of Professor Leo, seems to involve at least as much judicial discretion as Professor Davies’ approach, and it is far from clear that its actual operation would differ in any meaningful respect from the traditional voluntariness inquiry.[/tippy] Thus, like Professor Davies, Professor Leo advocates a discretionary test, rather than per se rules, again with good reason. As the late Welsh White observed, it makes little sense to exclude all confessions that are not corroborated by nonpublic information–that approach would make admissibility turn on frequently fortuitous circumstances that determine whether there is nonpublic information about the circumstances underlying a particular crime that can be used to independently corroborate a confession. [tippy title=”139″ header=”off”]See Welsh S. White, What Is an Involuntary Confession Now?, 50 Rutgers L. Rev. 2001, 2024-28 (1998).[/tippy] Indeed, Professor White may have been understating the problem. In my experience, it was difficult to get even highly motivated cooperating defendants to remember the details of crimes they had committed. Given the level at which most offenders operate, my guess is that Professor Leo is asking for an unrealistic degree of corroboration–certainly he provides no empirical evidence that the level of corroboration that he would require is usually present in truthful confessions. In any event, his balancing test, preserving as it does the ample discretion that inheres in all balancing tests, offers little guarantee that judicial screening of confessions will improve. The traditional voluntariness test requires judges to consider all of the factors identified by Professor Leo when assessing the admissibility of a confession, and yet by his own account, it has failed to weed out coerced confessions.
Thus, it is far from apparent that reliability review would be more effective than the voluntariness review that predominated prior to Miranda, and which most commentators have condemned as having failed to produce consistent and effective regulation. [tippy title=”140″ header=”off”]See, e.g., Mark Berger, Taking the Fifth: The Supreme Court and the Privilege against Self-Incrimination 104-12 (1980); Kamisar, A Dissent, supra note 51, at 69-76; White, supra note 30, at 39-48; Godsey, supra note 35, at 468-71; Ogletree, supra note 33, at 1833-35; Penney, supra note 65, at 337-62; Schulhofer, supra note 91, at 869-72; Seidman, supra note 30, at 727-36; Weisselberg, supra note 16, at 113-16. Even Professor Stuntz, another important advocate of greater judicial regulation of confessions, has acknowledged, “once one assumes that some pressure is acceptable, it is very hard to define how much pressure is too much.” Stuntz, supra note 30, at 980. Professor Godsey attempts to solve this problem by arguing that interrogators should be forbidden to threaten to impose what would be objectively characterized as a penalty on a suspect during interrogation to punish silence or encourage a confession, while acknowledging that a threat to seek the suspect’s conviction or an offer of leniency if the suspect cooperates should not be viewed as a penalty. See Godsey, supra note 35, at 515-38. It is unclear that this proposal is any different than existing law; indeed, Professor Godsey makes no effort to demonstrate that courts currently admit confessions obtained through what he regards as improper penalties. There is reason to be skeptical on this point. See, e.g., Lynumn v. Illinois, 372 U.S. 528 (1963) (inducing confession by threats that defendant would lose custody of her children); Spano v. New York, 360 U.S. 315 (1959) (inducing confession by threat that suspect’s friend woujld lose his job if suspect failed to cooperate). See also Paul Marcus, It’s Not Just about Miranda: Determining the Voluntariness of Confessions in Criminal Prosecutions, 40 Val. U. L. Rev. 601, 619-21 (2006) (reviewing cases). Moreover, as we will see below, the empirical work performed to date has not disclosed the use of such punitive tactics as a significant cause of false confessions. In any event, Professor Godsey’s concession that a confession made in the hope of leniency does not invalidate a resulting confession confirms the propriety of the most commonly used interrogation techniques.[/tippy] It is equally unclear that the judiciary could speak with sufficient uniformity on interrogation tactics to constitute a reliable regulator. Given the malleability of the voluntariness/reliability tests, it is unclear that a more muscular form of judicial review of confessions would be consistent or principled. [tippy title=”141″ header=”off”]For example, despite supposed judicial expertise in sentencing, the empirical evidence on pre-guidelines sentencing demonstrated the existence of substantial inter-judge sentencing disparities that were, among other things, the impetus for state and federal sentencing guidelines. See, e.g., Blakeley v. Washington, 542 U.S. 296, 315-20 (2004) (O’Connor, J., dissenting); id. at 332 (Breyer, J., dissenting); Theresa W. Karle & Thomas Sager, Are the Federal Sentencing Guidelines Meeting Congressional Goals?: An Empirical and Case Law Analysis, 40 Emory L.J. 393, 395 (1991); Norval Morris, Towards Principled Sentencing, 37 Md. L. Rev. 267, 272-74 (1977); David B. Mustard, Racial, Ethnic, and Gender Disparities in Sentencing: Evidence from the U.S. Federal Courts, 44 J.L. & Econ. 285, 311-12 (2001); Ilene H. Nagel, Foreword: Structuring Sentencing Discretion: The New Federal Sentencing Guidelines, 80 J. Crim. L & Criminology 883, 895-97 (1990); Kate Stith & Steve Y. Koh, The Politics of Sentencing Reform: The Legislative History of the Federal Sentencing Guidelines, 28 Wake Forest L. Rev. 223, 227-28 (1993). There is little reason not to expect similar disparities in judicial attitudes toward interrogation techniques under a regime of searching but ad hoc judicial review.[/tippy]
To be sure, judicial regulation would have relatively predictable regulatory consequences if it were based upon clear rules. The best example of such an approach comes from those who advocate the abolition of custodial interrogation by the police. [tippy title=”142″ header=”off”]See supra note 36 and accompanying text.[/tippy] The proponents of this approach argue that valid waivers cannot be given by suspects who are subject to the pressures of custodial interrogation. [tippy title=”143″ header=”off”]Irene and Yale Rosenberg, for example, argue that even a counseled waiver is necessarily infected by the pressures of custodial interrogation. See Rosenberg & Rosenberg, supra note 36, at 107-14. Professor Kamisar argues that the required warnings ought to be given by or in the presence of a judicial officer. See Kamisar, supra note 34.[/tippy] The advocates of abolition of custodial interrogation, however, make little effort to explain how their position can be squared with longstanding principles of waiver. As we have seen, the settled rule is that a suspect can validly waive constitutional rights when he thinks it is in his interest to do so despite the pressures created by a pending investigation or prosecution, and regardless of whether the suspect has correctly assessed his own interests. [tippy title=”144″ header=”off”]See supra text accompanying notes 74-102. To be fair, Charles Ogletree has maked an attempt along these lines. Writing before the Supreme Court established that the Sixth Amendment right to counsel can be waived extrajudicially in Patterson v. Illinois, 487 U.S. 285, 292-93 (1988), he acknowledged that the right to counsel can be validly waived before a judicial officer, but argued that “[a] suspect does not have the same degree of protection in the stationhouse when the warnings are given by a police officer intent on interrogating the suspect….” Ogletree, supra note 33, at 1844 n.97. We have seen, however, that the test for waiver asks only if the defendant knows he has a right to counsel and intentionally surrenders it; waiver law has never asked whether the defendant knows enough about the value of counsel to correctly assess his own interests. See, e.g., Iowa v. Tovar, 541 U.S. 77, 87-92 (2004). And, as we have also seen, the Miranda warnings ensure that the suspect knows that he has a right to have counsel present during questioning. Thus, the structure of waiver law seems far more consistent with Patterson than with Professor Ogletree’s position, and the propriety of extrajudicial waivers of the Fifth Amendment right to counsel recognized in Miranda should follow a fortiori from Patterson.[/tippy] Of course, one could simply announce that as a prophylactic matter, the Miranda right to counsel cannot be waived during custodial interrogation, but such a rule would be quite a radical innovation in the law of waiver. [tippy title=”145″ header=”off”]An additional objection can be made to the proposal that is sometimes advanced that would conjoin judicially supervised interrogation, whether in counsel’s presence or not, with a warning that the suspect’s refusal to answer questions could be used as evidence of his guilt at trial. See, e.g., Amar, supra note 49, at 70-88; Kamisar, supra note 34, at 84, 93-94; Alschuler, supra note 51, at 2667-72; see also Russell D. Covey, Interrogation Warrants, 26 Cardozo L. Rev. 1867, 1904-09, 1926-32 (2005) (advocating for compulsory police interrogation authorized by a warrant). These proposals, however, do not properly accommodate the text of the Fifth Amendment. A guilty defendant subject to such a procedure, for example, has no choice but to incriminate himself, either by confessing, providing an exculpatory account that could later be disproved and therefore used as an incriminating false statement of exculpation (as in Bram), or remaining silent, which would itself be treated as evidence of guilt. It appears that compulsory pretrial examination of an accused was common in most states at the time of the framing, but there is little evidence of any prevalent legal understanding that squared this practice with the text of the Fifth Amendment. See Moglen, supra note 52, at 1123-29. In any event, the practices of the states at the time of the framing are of little relevance since the Fifth Amendment was not made applicable to the states until Malloy v. Hogan, 378 U.S. 1 (1964). Rather than basing his argument on an originalist understanding of the Fifth Amendment, Professor Alschuler asserts that “[a] suspect’s answers to orderly questioning in a safeguarded courtroom environment should not be regarded as the product of compulsion,” Alschuler, supra note 51, at 2670, but it is difficult to take this argument seriously. Under such a system, a suspect would be compelled to attend the judicial interrogation under threat of sanction. That is compulsion no less than is requiring a suspect to take an oath under threat of contempt. As we have seen, the threat of criminal sanctions if a suspect does not submit to interrogation is the proper place to draw the line between persuasion and compulsion. And while this objection is particularly acute for suspects who are in fact guilty, the text of the Fifth Amendment does not limit its protections to the innocent.[/tippy]
Perhaps doctrinal innovation would be appropriate based on a sufficient showing that the current approach has led to widespread abuse. Abuse, however, cannot be defined as evidence that police succeed in obtaining waivers in a substantial majority of cases. As we have seen, nothing in the Fifth Amendment forbids a waiver of the right to be free from compelled self-incrimination. A showing that current waiver law failed to prevent what must be compelled or involuntary confessions, however, might well justify doctrinal reform. Such a case for innovation, however, is necessarily an empirical one, and only a few of the Miranda critics attempt to make it. [tippy title=”146″ header=”off”]This is not to say that the critics have been completely unable to find empirical evidence of interrogation techniques inconsistent with Miranda. For example, there is anecdotal evidence that police sometimes tell suspects that the only way they can learn about the charges against them or obtain some form of leniency is to waive their Miranda rights. Leo & White, supra note 45, at 440. This evidence shows a misdescription of the suspect’s rights and the consequences of invoking them given a suspect’s right to notice of charges and his ability to engage in plea bargaining even after asserting his Miranda rights. Accordingly, under traditional waiver principles, these tactics will fail to produce valid waivers because they mislead the suspect about the nature of his rights and the consequences of invoking them. Indeed, under current law, confessions obtained through misrepresentations of this type are usually suppressed. Marcus, supra note 140, at 615-18.[/tippy]
One Miranda critic who did take up this cudgel was Professor White. Based on a review of confessions in cases in which the suspect was ultimately exonerated, he argued for a prohibition on techniques that have produced significant numbers of false confessions: lengthy interrogations, interrogation of vulnerable suspects such as minors or the mentally disabled, and interrogations involving threats, deception or promises. [tippy title=”147″ header=”off”]White, supra note 30, at 196-215. See also, e.g., Alschuler, supra note 32, at 971-74 (false statements about the evidence against the suspect); Gohara, supra note 35, at 834-40 (deception); Young, supra note 35, at 456-75 (same).[/tippy] Similarly, Professor Leo has used empirical evidence of false confessions to support his proposal that confessions must be corroborated by independent evidence to guard against what he regards as an unacceptable risk of a false confession. [tippy title=”148″ header=”off”]Leo et al., supra note 31, at 512-19, 525-35.[/tippy] This approach, like a flat ban on custodial interrogation, is unquestionably prophylactic. Professor White, for example, never argued that lengthy interrogations, interrogation of vulnerable suspects, or interrogations involving threats and deception always produce involuntary or unreliable confessions; instead, his claim was that the risk of an involuntary, unreliable, or otherwise suspect confession is so high in such circumstances that confessions obtained through these tactics should be barred from evidence by the Due Process Clause. [tippy title=”149″ header=”off”]White, supra note 30, at 196-215; White, supra note 139, at 2042-58.[/tippy] Similarly, although the Leo proposal is vague about the doctrinal basis for its approach, it, too, is rooted in concerns about the supposed prevalence of unreliable confessions, and appears to be premised on the Due Process Clause. [tippy title=”150″ header=”off”]See Leo et al., supra note 31, at 493-522; Ofshe & Leo, supra note 34, at 1115-22.[/tippy]
The Due Process Clause is indeed the best doctrinal support for these proposals. Suspects who have waived their Miranda rights have already made a decision to subject themselves to compelled self-incrimination by agreeing to answer the questions of their captors, and in any event, the advocates of these reforms do not claim that the confessions that they would exclude necessarily involve compulsion within the meaning of the Fifth Amendment. [tippy title=”151″ header=”off”]For present purposes, I put aside the question whether some suspects, because of their age or mental condition, should be deemed incapable of validly waiving their Miranda rights. There will undoubtedly be circumstances in which youth or mental disability might prevent an individual from supplying a valid waiver, but there will also be many circumstances in which minors or those under a mental disability will be able to make a deliberate and voluntary choice to waive their Miranda rights, which is all that standard waiver doctrine requires. See, e.g., Colorado v. Connelly, 479 U.S. 157, 169-70 (1986) (holding that voluntariness of a waiver “has always depended on the absence of police overreaching, not on ‘free choice’ in any broader sense of the word”). It is worth noting, however, that one recent study concluded that the ability of juveniles older than fifteen to understand their Miranda rights while under interrogation was on par with that of adults. Feld, supra note 108, at 90-92. The author warned, however, that juveniles may be more vulnerable to police influence. Id. at 98-100. Another recent study provides far greater reason to doubt that mentally retarded subjects are capable of giving valid waivers. See Morgan Cloud et al., Words without Meaning: The Constitution, Confessions, and Mentally Retarded Suspects, 69 U. Chi. L. Rev. 495 (2002).[/tippy] To be sure, Bram supplies at least some support for using the Fifth Amendment to regulate threats or promises; in Bram, the Court wrote that “a confession, in order to be admissible, must be free and voluntary: that is, must not be extracted by any sort of threats or violence, nor obtained by any direct or implied promises, however slight, nor by the exertion of any improper influence.” [tippy title=”152″ header=”off”]Bram v. United States, 168 U.S. 532, 542-43 (1897) (quoting Sir Wm. Oldnall Russell, A Treatise on Crimes and Misdemeanors 478 (Horace Smith & A.P. Perceval Keep eds., 6th ed. 1896)). See also Hirsch, supra note 35, at 54-59.[/tippy] This passage, however, did not consider whether an admissible confession might result if a confession is preceded by a knowing and intelligent waiver of Fifth Amendment rights. Moreover, the Court has more recently stated that “this passage [in] Bram . . . does not state the standard for determining the voluntariness of a confession.” [tippy title=”153″ header=”off”]Arizona v. Fulminante, 499 U.S. 279, 285 (1991).[/tippy] The Court was correct; as we have seen, under settled principles of waiver, the fact that a suspect is facing the threat of enhanced sanctions if he asserts his constitutional rights does not invalidate a waiver of those rights, as long as the suspect understands his rights and elects to surrender them in the hope of some reciprocal advantage. [tippy title=”154″ header=”off”]See supra text accompanying notes 74-102.[/tippy]
Thus, we are left with the Due Process Clause as the basis for additional judicial regulation of interrogation. Assessing the case for due process regulation, however, involves a necessarily empirical inquiry. Suppose, for example, that the type of threat that is condoned by the plea bargaining cases–the threat to take the suspect to trial on the most serious possible charges and then seek the harshest possible sentence–was likely to produce only accurate confessions because only guilty suspects were likely to yield to such a threat. [tippy title=”155″ header=”off”]The defenders of plea bargaining often take essentially this view of the plea bargaining process, which, we have seen, also involves threats of punishment as a means to garner waivers of constitutional rights. See generally, e.g., Frank H. Easterbrook, Plea Bargaining as Compromise, 101 Yale L.J. 1969 (1992); Scott W. Howe, The Value of Plea Bargaining, 58 Okla. L. Rev. 599 (2005); Robert E. Scott & William J. Stuntz, Plea Bargaining as Contract, 101 Yale L.J. 1909 (1992).[/tippy] Or suppose that the type of deception condoned by Frazier v. Cupp–false claims that the authorities have highly incriminating evidence against the suspect–was also likely to produce reliable confessions because only the guilty were likely to yield to such claims. [tippy title=”156″ header=”off”]Some courts have condoned such deception because they believe that suspects who are confident of their innocence will not be influenced by this form of deception. See, e.g., State v. Kelekolio, 849 P.2d 58, 71-74 (Haw. 1993); Sheriff, Washoe County v. Bessey, 914 P.2d 618 (Nev. 1996).[/tippy] As it happens, there is some empirical evidence that pressure or deception is useful to obtaining confessions, and is more likely to produce truthful than false confessions. [tippy title=”157″ header=”off”]See Christopher Slobogin, Lying and Confessing, 39 Texas Tech, L. Rev. 1275, 1280-84 (2007). One particularly striking study involved interrogation of participants about whether they had improperly received assistance during a decision making exercise, utilizing for some participants a interrogation tactic involving an offer of leniency if they confessed and, for others, the use of tactics designed to minimize the seriousness of the offense, The study indicated that these pressure tactics, at least when used separately, increased the rate of true confessions far more than the rate of false confessions[/tippy] It is surely difficult to construct an argument that due process forbids interrogation techniques that are likely to produce accurate confessions by suspects who have knowingly and intelligently agreed to submit to interrogation despite an awareness of their Miranda rights–indeed, the advocates of additional due process regulation do not make such an argument. [tippy title=”158″ header=”off”]I put aside confessions obtained by false promises of leniency or other benefits. Although the Supreme Court has never squarely decided whether a confession induced by a promise of leniency must be suppressed if the promise is not honored, it has held that a guilty plea cannot stand when induced by an unfulfilled promise. See Santobello v. New York, 404 U.S. 257, 261-63 (1971). It would seem to follow that a promise that induces a confession must be honored. See Welsh S. White, Confessions Induced by Broken Government Promises, 43 Duke L.J. 947 (1994). For a review of the pertinent case law, which makes inadmissibility turn on whether the defendant confessed in reliance on an unfulfilled promise, see Marcus, supra note 140, at 621-24. To be sure, a skillful interrogator will be able to raise a suspect’s hopes for leniency without making a promise–I did just that on many occasions.[/tippy] To be sure, some commentators are troubled by any form of official deception, [tippy title=”159″ header=”off”]See, e.g., Margaret L. Paris, Lying to Ourselves, 76 Or. L. Rev. 817 (1997); Margaret L. Paris, Trust, Lies, and Interrogation, 3 Va. J. Soc. Pol’y & L. 3 (1995); Young, supra note 35, at 468-71. Deborah Young, without citing empirical evidence to support her supposition, adds a utilitarian claim by arguing that police deceit is counterproductive because it will produce distrust that will ultimately reduce civilian cooperation with the authorities. See id. at 457-60. Those who claim that the law enforcement community does not understand where its own interests lie should bear the burden of adducing some empirical evidence to that effect.[/tippy] but adopting that view would mark a radical change in our constitutional tradition which, for example, has long tolerated police undercover work despite the necessary deceit that it usually entails. [tippy title=”160″ header=”off”]See, e.g., United States v. White, 401 U.S. 745 (1971); Hoffa v. United States, 385 U.S. 293 (1966); Lewis v. United States, 385 U.S. 206 (1966); On Lee v. United States, 343 U.S. 747 (1952); Andrews v. United States, 162 U.S. 420 (1896); Grimm v. United States, 156 U.S. 604 (1895).[/tippy] Moreover, it is a respectable moral position to permit official deceit in the interest of a greater social good, such as the detection or prevention of crime, at least when there is not an unreasonable likelihood of convicting the innocent and when those who employ these tactics are subject to political accountability. [tippy title=”161″ header=”off”]For an argument along these lines, see Christopher Slobogin, Deceit, Pretext, and Trickery: Investigative Lies by the Police, 76 Or. L. Rev. 775 (1997).[/tippy] Thus, for the most part, the advocates of additional due process regulation of interrogation tactics stake their position on an empirical claim that these tactics endanger the innocent. [tippy title=”162″ header=”off”]See, e.g., White, supra note 30, at 196-215; Alschuler, supra note 32, at 967-78; Gohara, supra note 35, at 816-34.[/tippy]
The problem with the empirical case for greater due process regulation, however, is that we have no idea what rate of false confessions is produced by the tactics that the critics have targeted. For example, there is no data that provides even a rough guess about how likely the tactics that Professor White has identified produce false confessions, much less wrongful convictions, [tippy title=”163″ header=”off”]See, e.g., Paul G. Cassell, Protecting the Innocent from False Confessions and Lost Confessions–and from Miranda, 88 J. Crim. L. & Criminology 497, 503-38 (1998); Laurie Magid, Deceptive Police Interrogation Practices: How Far Is Too Far?, 99 Mich. L. Rev. 1168, 1188-97 (2001).[/tippy] as Professor White ultimately acknowledged. [tippy title=”164″ header=”off”]See White, supra note 30, at 1224-29. Instead, Professor White argued that the data suggests that about one-tenth of wrongful convictions involve false confessions in potentially capital cases, while admitting that the percentage of confession-induced wrongful convictions is likely lower in non-capital cases. See id. at 1228-29. This tells us next to nothing about the rate at which the interrogation tactics of which he complains induce wrongful confessions, and even less about how many truthful confessions would be lost under the reforms that he advocates.[/tippy] Similarly, Professor Leo’s approach, forbidding what are thought to be inadequately corroborated confessions, makes no effort to identify an error rate for such confessions. [tippy title=”165″ header=”off”]See Leo et al., supra note 31, at 512-20.[/tippy] It may be that a good many confessions with no more corroboration than was present in the Central Park Jogger case, for example, are entirely accurate. Nor have the advocates of the Leo approach explained why judges are more likely than juries to be able to identify confessions that are likely to be false. No empirical evidence supports such a claim, and, as we have seen, the judicial track record under the current voluntariness test provides little cause for optimism.
Maybe even more important, we do not even know if the tactics identified by the critics produce disproportionate numbers of false confessions. Perhaps they do not. When I engaged in interrogation as a prosecutor, for example, I usually engaged in some degree of puffing about the strength of the case against the suspect–something that might well be branded deception. I also regularly engaged in what could be characterized as threats–at least the kind of threat to seek the maximum punishment generally condoned by the plea bargaining cases. [tippy title=”166″ header=”off”]To be fair, I should acknowledge that as I gained experience, I tended to place these threats into a type of “I’d really like to help you if you’ll let me” context, not because I had particular scruples about threats, but because I found that a congenial ambience made for more effective interrogation.[/tippy] And, in every multiple confession case that I handled, the suspects initially contradicted each other, placing greater culpability on each other.
Thus, it would not surprise me if the vast majority of custodial interrogations involve the features condemned by critics. If so, the fact that a study of false confessions will frequently disclose the use of the interrogation tactics identified by Professor White, or what Professor Leo would regard as insufficiently corroborated confessions, provides no basis to conclude that these features increase the likelihood that a confession is false. At best, it is probably reasonable to presume that more aggressive interrogation techniques will produce a higher rate of confessions than more passive approaches, but it is entirely unclear that the rate of false confessions will also increase through more aggressive techniques. [tippy title=”167″ header=”off”]To be sure, one can build an anecdotal case that interrogators sometimes persuade a suspect that his position is so hopeless that he has no realistic choice but to confess, see, e.g., Saul M. Kassin, On the Psychology of Confessions: Does Innocence Put Innocents At Risk?, 60 Am. Psychologist 215, 220-22 (2005); Ofshe & Leo, supra note 34, at 1004-114, but this says nothing about the rate at which the very same tactics induce a guilty suspect to provide an accurate confession.[/tippy]
Although the critics make no effort to identify an error rate for the confessions that they would exclude from evidence, perhaps their surveys of false confession cases justifies an assumption that there is some nontrivial error rate associated with the interrogation tactics that they target for elimination. But “[t]here is always in litigation a margin of error . . . .” [tippy title=”168″ header=”off”]Speiser v. Randall, 357 U.S. 513, 525 (1958).[/tippy] I never prosecuted an individual about whose guilt I had a reasonable doubt, but if it surfaced today that some of the convictions I obtained were inaccurate, I would not be shocked. Even under a reasonable-doubt standard, factfinding is necessarily a probabilistic business, and most kinds of proof inject a risk of error. For example, the available empirical evidence demonstrates a risk of error in the use of eyewitness testimony, [tippy title=”169″ header=”off”]See, e.g., Brian L. Cutler & Steven D. Penrod, Mistaken Identification: The Eyewitness, Psychology, and the Law 10-13 (1995); Gross et al., supra note 31, at 542-44; Gary L. Wells et al., From the Lab to the Police Station: A Successful Application of Eyewitness Research, 55 Am. Psychologist 581 (2000).[/tippy] accomplice testimony, [tippy title=”170″ header=”off”]See, e.g., Alexandra Natapoff, Beyond Unreliable: How Snitches Contribute to Wrongful Convictions, 37 Golden Gate U. L. Rev. 107 (2006).[/tippy] and even fingerprint evidence. [tippy title=”171″ header=”off”]See, e.g., Simon A. Cole, More than Zero: Accounting for Error in Latent Fingerprint Identification, 95 J. Crim. L. & Criminology 985 (2005); Simon A. Cole, The Prevalence and Potential Causes of Wrongful Conviction by Fingerprint Evidence, 37 Golden Gate U. L. Rev. 39 (2006).[/tippy]
In fact, false confessions may not be the leading cause of erroneous convictions. Surveys of documented exonerations consistently find that eyewitness testimony is the leading cause of false convictions. [tippy title=”172″ header=”off”]See, e.g., Gross et al., supra note 31, at 542-46; Adrian T. Grounds, Understanding the Effects of Wrongful Imprisonment, in Crime and Justice: A Review of Research 1, 10-11 (Michael Tonry ed., 2005); Kevin Jon Heller, The Cognitive Psychology of Circumstantial Evidence, 105 Mich. L. Rev. 241, 253-56 (2006). This estimate should be viewed with caution, however, since documented exonerations are not likely to be a random sample of all wrongful convictions, but instead are heavily skewed toward sex crimes and other types of offenses in which DNA or other conclusive physical evidence can establish factual innocence, or serious crimes subject to intensive investigation. See, e.g., Gross, et al., supra note 31, at 529-40.[/tippy] It would seem to follow (once any need to identify an actual error rate has been jettisoned) that if due process mandates regulation of interrogation practices because of the risk of error, due process must compel restrictions on the use of eyewitness testimony, and other types of evidence or investigative tactics that produce what one could plausibly surmise to be an unacceptable error rate. [tippy title=”173″ header=”off”]Due process already requires the exclusion of eyewitness identifications that are thought to be the product of unduly suggestive identification procedures absent sufficient indicia of reliability. Manson v. Brathwaite, 432 U.S. 98, 109-14 (1977). But since this rule appears not to have prevented some significant number of false convictions, the logic of the due process position would seem to be that it must be supplemented by some additional prophylactic safeguards.[/tippy] Moreover, even if courts could somehow divine error rates, how are they to decide what constitutes an unacceptable risk of error? Three percent? Ten percent? We are not told, but presumably the critics would require restrictions on investigative tactics at something well below fifty percent. That would mean that far more reliable evidence of guilt would be excluded than would false evidence of guilt. And what about the large numbers of guilty offenders who will go unpunished if courts brand as impermissible investigative tactics that are far more likely to produce accurate than false convictions but that nevertheless produce error rates that are thought to be unacceptable? The Miranda-is-a-failure scholarship evinces no particular concern about this problem, although surely there is reason to believe that conviction rates will be reduced if eyewitness identifications, accomplice testimony, aggressive interrogation techniques, or other tactics thought to produce unacceptable error rates are sharply circumscribed, if not prohibited altogether. [tippy title=”174″ header=”off”]The risk that reforms will go wrong is far from hypothetical. Although a large majority of experts in the field of identification testimony have relied on laboratory data to urge that witnesses view potential offenders sequentially rather than in simultaneous lineups in procedures administered by a “blind” official who does not know who has been identified as a suspect by investigators, see Dawn McQuiston-Surrett, Roy S. Malpass & Colin G. Tredoux, Sequential vs. Simultaneous Lineups: A Review of Methods, Data, and Theory, 12 Psychol. Pub. Pol’y & L. 137, 137-38 (2006), the limited empirical data gathered in the field to date shows that the rate at which the suspect is identified goes down while the rate at which an innocent “filler” is identified goes up. See id. at 161-62. Thus, it appears that this technique may increase the risk that an innocent suspect will be identified.[/tippy]
All of this should suggest that due process regulation of interrogation and other investigative techniques based on a presumed risk of error is deeply problematic. Nor is there any ready answer to the argument that “due process,” in this context, requires something more than deference to the political process. [tippy title=”175″ header=”off”]When it comes to what are considered “legislative” rules applied to large numbers of cases, the legislative process is ordinarily thought to supply all the process that is constitutionally “due.” See, e.g., Atkins v. Parker, 472 U.S. 115, 129-30 (1985); Logan v. Zimmerman Brush Co., 455 U.S. 422, 432-33 (1982); Bi-Metallic Investment Co. v. State Bd. of Equalization, 239 U.S. 441, 445 (1915).[/tippy] It is hard to be unsympathetic to the problem of false convictions, but for that very reason, it is far from clear that the political process will fail to respond to the problem, at least when reliable data becomes available demonstrating how the error rate in the criminal process can be efficaciously reduced without unacceptable limitations on the ability to convict the guilty.
Due process is not unconcerned with the risk of error in the criminal justice system. Due process is understood to require the prosecution to prove a criminal defendant’s guilt beyond reasonable doubt for just this reason. [tippy title=”176″ header=”off”]See, e.g., Cage v. Louisiana, 498 U.S. 39, 39-40 (1990) (per curiam); Jackson v. Virginia, 443 U.S. 307, 315 (1979); Addington v. Texas, 441 U.S. 418, 423-24 (1979); In re Winship, 397 U.S. 358, 370-72 (1970) (Harlan, J., concurring).[/tippy] The advocates of due process regulation of interrogation (and other investigative) techniques, however, seek additional protection, based on evidence that these techniques produce some nontrivial (although as yet unascertained) error rate. As a doctrinal matter, the absence of any historical support for prophylactic due process regulation of interrogation techniques based on a presumed risk of error might itself doom the case for new regulation. [tippy title=”177″ header=”off”]See Medina v. California, 505 U.S. 437, 442-46 (1992). See generally Jerold H. Israel, Free-Standing Due Process and Criminal Procedure: The Supreme Court’s Search for Interpretative Guidelines, 45 St. Louis U. L.J. 303 (2001). To be fair, there is at least a bit of precedent for ahistorical due process regulation in order to reduce the risk of error in the criminal process. The Court departed from history when, in Brady v. Maryland, 373 U.S. 83 (1963), it held that prosecutors must disclose exculpatory information to the defense, and subsequently broadened that duty to require prosecutors to identify and disclose exculpatory information in the hands of the police and other investigators, see Kyles v. Whitley, 514 U.S. 419, 437 (1995), even though historically prosecutors had never been placed under any type of duty of disclosure. Michael Moore, Criminal Discovery, 19 Hastings L.J. 865, 865-67, 893-99 (1968); see generally John H. Langbein, The Origins of Adversary Criminal Trial 283-343 (2003).[/tippy] Even putting that problem aside, however, no one could tenably read the Due Process Clause as a prohibition of error in the criminal justice system. In any system administered by humans, there will be error. Surely “due process” accommodates that much reality.
Conclusion
The advocates of due process prohibition of interrogation techniques thought to pose unacceptable risks of error are what Henry Monaghan once called constitutional “perfectionists.” [tippy title=”178″ header=”off”]Henry P. Monaghan, Our Perfect Constitution, 56 N.Y.U. L. Rev. 353, 358 (1981).[/tippy] Although he applied the label to the 1970s-vintage movement among legal scholars to read the Constitution as mandating whatever they believed to be a more just system of governance, [tippy title=”179″ header=”off”]See id. at 355-60.[/tippy] the same phenomenon is today perhaps even more prevalent among scholars of constitutional criminal procedure. Led by Professor Stuntz, they read the Constitution to command whatever they believe to be the optimal system of convicting the guilty and protecting the innocent. [tippy title=”180″ header=”off”]Professor Stuntz’s criticism of Miranda, for example, is that it fails to achieve what he regards as distributive justice by maximizing the number of suspects who submit to interrogation while minimizing what he regards as abusive questioning. See Stuntz, supra note 30, at 992-98. It is far from clear how this objective comports with the text of the Fifth Amendment, but that does not seem to be the point of the exercise.[/tippy]
The Constitution’s objectives, however, are more modest. Our Constitution does not contain a “no-conviction-of-the-innocent” clause, presumably not because anyone wants to convict the innocent, but because such an objective is unattainable. Perhaps the empirical evidence may one day be available that will enable us to reduce the risk of error in the criminal process without placing unacceptable constraints on our ability to convict the guilty. That day, however, has not yet arrived. And, although scholars can occupy the ivory towers of constitutional perfectionism with impunity, those in the trenches of law enforcement are expected to deal with the grimmest of daily realities. In that world, perfectionism is beyond reach. Getting a valid Miranda waiver ought to be good enough.