skip navigation
search

MONTEJO v. LOUISIANA (No. 07-1529)
06–1807 (La.), 974 So. 2d 1238, vacated and remanded.
Syllabus

Opinion
[Scalia]
Concurrence
[Alito]
Dissent
[Stevens]
Dissent
[Breyer]
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version

Justice Scalia, Opinion of the Court

NOTICE: This opinion is subject to formal revision before publication in the preliminary print of the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D. C. 20543, of any typographical or other formal errors, in order that corrections may be made before the preliminary print goes to press.

SUPREME COURT OF THE UNITED STATES

JESSE JAY MONTEJO, PETITIONER v.LOUISIANA

on writ of certiorari to the supreme court oflouisiana


[May 26, 2009]

    Justice Scalia delivered the opinion of the Court.

    We consider in this case the scope and continued viability of the rule announced by this Court in Michigan v. Jackson, 475 U. S. 625 (1986) , forbidding police to initiate interrogation of a criminal defendant once he has requested counsel at an arraignment or similar proceeding.

I

    Petitioner Jesse Montejo was arrested on September 6, 2002, in connection with the robbery and murder of Lewis Ferrari, who had been found dead in his own home one day earlier. Suspicion quickly focused on Jerry Moore, a disgruntled former employee of Ferrari’s dry cleaning business. Police sought to question Montejo, who was a known associate of Moore.

    Montejo waived his rights under Miranda v. Arizona, 384 U. S. 436 (1966) , and was interrogated at the sheriff’s office by police detectives through the late afternoon and evening of September 6 and the early morning of September 7. During the interrogation, Montejo repeatedly changed his account of the crime, at first claiming that he had only driven Moore to the victim’s home, and ultimately admitting that he had shot and killed Ferrari in the course of a botched burglary. These police interrogations were videotaped.

    On September 10, Montejo was brought before a judge for what is known in Louisiana as a “72-hour hearing”—a preliminary hearing required under state law.1 Although the proceedings were not transcribed, the minute record indicates what transpired: “The defendant being charged with First Degree Murder, Court ordered N[o] Bond set in this matter. Further, Court ordered the Office of Indigent Defender be appointed to represent the defendant.” App. to Pet. for Cert. 63a.

    Later that same day, two police detectives visited Montejo back at the prison and requested that he accompany them on an excursion to locate the murder weapon (which Montejo had earlier indicated he had thrown into a lake). After some back-and-forth, the substance of which remains in dispute, Montejo was again read his Miranda rights and agreed to go along; during the excursion, he wrote an inculpatory letter of apology to the victim’s widow. Only upon their return did Montejo finally meet his court-appointed attorney, who was quite upset that the detectives had interrogated his client in his absence.

    At trial, the letter of apology was admitted over defense objection. The jury convicted Montejo of first-degree murder, and he was sentenced to death.

    The Louisiana Supreme Court affirmed the conviction and sentence. 06–1807 (1/16/08), 974 So. 2d 1238 (2008). As relevant here, the court rejected Montejo’s argument that under the rule of Jackson, supra,the letter should have been suppressed. 974 So. 2d, at 1261. Jackson held that “if police initiate interrogation after a defendant’s assertion, at an arraignment or similar proceeding, of his right to counsel, any waiver of the defendant’s right to counsel for that police-initiated interrogation is invalid.” 475 U. S., at 636.

    Citing a decision of the United States Court of Appeals for the Fifth Circuit, Montoya v. Collins, 955 F. 2d 279 (1992), the Louisiana Supreme Court reasoned that the prophylactic protection of Jackson is not triggered unless and until the defendant has actually requested a lawyer or has otherwise asserted his Sixth Amendment right to counsel. 974 So. 2d, at 1260–1261, and n. 68. Because Montejo simply stood mute at his 72-hour hearing while the judge ordered the appointment of counsel, he had made no such request or assertion. So the proper inquiry, the court ruled, was only whether he had knowingly, intelligently, and voluntarily waived his right to have counsel present during the interaction with the police. Id., at 1261. And because Montejo had been read his Miranda rights and agreed to waive them, the Court answered that question in the affirmative, 974 So. 2d, at 1262, and upheld the conviction.

    We granted certiorari. 554 U. S. ___ (2008).

II

    Montejo and his amici raise a number of pragmatic objections to the Louisiana Supreme Court’s interpretation of Jackson. We agree that the approach taken below would lead either to an unworkable standard, or to arbitrary and anomalous distinctions between defendants in different States. Neither would be acceptable.

    Under the rule adopted by the Louisiana Supreme Court, a criminal defendant must request counsel, or otherwise “assert” his Sixth Amendment right at the preliminary hearing, before the Jackson protections are triggered. If he does so, the police may not initiate further interrogation in the absence of counsel. But if the court on its own appoints counsel, with the defendant taking no affirmative action to invoke his right to counsel, then police are free to initiate further interrogations provided that they first obtain an otherwise valid waiver by the defendant of his right to have counsel present.

    This rule would apply well enough in States that require the indigent defendant formally to request counsel before any appointment is made, which usually occurs after the court has informed him that he will receive counsel if he asks for it. That is how the system works in Michigan, for example, Mich. Ct. Rule 6.005(A) (2009), whose scheme produced the factual background for this Court’s decision in Michigan v. Jackson. Jackson, like all other represented indigent defendants in the State, had requested counsel in accordance with the applicable state law.

    But many States follow other practices. In some two dozen, the appointment of counsel is automatic upon a finding of indigency, e.g., Kan. Stat. Ann. §22–4503(c) (2007); and in a number of others, appointment can be made either upon the defendant’s request or sua sponte by the court, e.g., Del. Code Ann., Tit. 29, §4602(a) (2003). See App. to Brief for National Legal Aid & Defender Assn. et al. as Amici Curiae 1a–21a. Nothing in our Jackson opinion indicates whether we were then aware that not all States require that a defendant affirmatively request counsel before one is appointed; and of course we had no occasion there to decide how the rule we announced would apply to these other States.

    The Louisiana Supreme Court’s answer to that unresolved question is troublesome. The central distinction it draws—between defendants who “assert” their right to counsel and those who do not—is exceedingly hazy when applied to States that appoint counsel absent request from the defendant. How to categorize a defendant who merely asks, prior to appointment, whether he will be appointed counsel? Or who inquires, after the fact, whether he has been? What treatment for one who thanks the court after the appointment is made? And if the court asks a defendant whether he would object to appointment, will a quick shake of his head count as an assertion of his right?

    To the extent that the Louisiana Supreme Court’s rule also permits a defendant to trigger Jackson through the “acceptance” of counsel, that notion is even more mysterious: How does one affirmatively accept counsel appointed by court order? An indigent defendant has no right to choose his counsel, United States v. Gonzalez-Lopez, 548 U. S. 140, 151 (2006) , so it is hard to imagine what his “acceptance” would look like, beyond the passive silence that Montejo exhibited.

    In practice, judicial application of the Louisiana rule in States that do not require a defendant to make a request for counsel could take either of two paths. Courts might ask on a case-by-case basis whether a defendant has somehow invoked his right to counsel, looking to his conduct at the preliminary hearing—his statements and gestures—and the totality of the circumstances. Or, courts might simply determine as a categorical matter that defendants in these States—over half of those in the Union—simply have no opportunity to assert their right to counsel at the hearing and are therefore out of luck.

    Neither approach is desirable. The former would be particularly impractical in light of the fact that, as amici describe, preliminary hearings are often rushed, and are frequently not recorded or transcribed. Brief for National Legal Aid & Defender Assn. et al. 25–30. The sheer volume of indigent defendants, see id., at 29, would render the monitoring of each particular defendant’s reaction to the appointment of counsel almost impossible. And sometimes the defendant is not even present. E.g., La. Code Crim. Proc. Ann., Art. 230.1(A) (West Supp. 2009) (allowing court to appoint counsel if defendant is “unable to appear”). Police who did not attend the hearing would have no way to know whether they could approach a particular defendant; and for a court to adjudicate that question ex post would be a fact-intensive and burdensome task, even if monitoring were possible and transcription available. Because “clarity of … command” and “certainty of … application” are crucial in rules that govern law enforcement, Minnick v. Mississippi, 498 U. S. 146, 151 (1990) , this would be an unfortunate way to proceed. See also Moran v. Burbine, 475 U. S. 412, 425–426 (1986) .

    The second possible course fares no better, for it would achieve clarity and certainty only at the expense of introducing arbitrary distinctions: Defendants in States that automatically appoint counsel would have no opportunity to invoke their rights and trigger Jackson, while those in other States, effectively instructed by the court to request counsel, would be lucky winners. That sort of hollow formalism is out of place in a doctrine that purports to serve as a practical safeguard for defendants’ rights.

III

    But if the Louisiana Supreme Court’s application of Jackson is unsound as a practical matter, then Montejo’s solution is untenable as a theoretical and doctrinal matter. Under his approach, once a defendant is represented by counsel, police may not initiate any further interrogation. Such a rule would be entirely untethered from the original rationale of Jackson.

A

    It is worth emphasizing first what is not in dispute or at stake here. Under our precedents, once the adversary judicial process has been initiated, the Sixth Amendment guarantees a defendant the right to have counsel present at all “critical” stages of the criminal proceedings. United States v. Wade, 388 U. S. 218, 227–228 (1967) ; Powell v. Alabama, 287 U. S. 45, 57 (1932) . Interrogation by the State is such a stage. Massiah v. United States, 377 U. S. 201, 204–205 (1964) ; see also United States v. Henry, 447 U. S. 264, 274 (1980) .

    Our precedents also place beyond doubt that the Sixth Amendment right to counsel may be waived by a defendant, so long as relinquishment of the right is voluntary, knowing, and intelligent. Patterson v. Illinois, 487 U. S. 285 , n. 4 (1988); Brewer v. Williams, 430 U. S. 387, 404 (1977) ; Johnson v. Zerbst, 304 U. S. 458, 464 (1938) . The defendant may waive the right whether or not he is already represented by counsel; the decision to waive need not itself be counseled. Michigan v. Harvey, 494 U. S. 344, 352–353 (1990) . And when a defendant is read his Miranda rights (which include the right to have counsel present during interrogation) and agrees to waive those rights, that typically does the trick, even though the Miranda rights purportedly have their source in the Fifth Amendment:

“As a general matter … an accused who is admonished with the warnings prescribed by this Court in Miranda … has been sufficiently apprised of the nature of his Sixth Amendment rights, and of the consequences of abandoning those rights, so that his waiver on this basis will be considered a knowing and intelligent one.” Patterson, supra, at 296.

    The only question raised by this case, and the only one addressed by the Jackson rule, is whether courts must presume that such a waiver is invalid under certain circumstances. 475 U. S., at 630, 633. We created such a presumption in Jackson by analogy to a similar prophylactic rule established to protect the Fifth Amendment based Miranda right to have counsel present at any custodial interrogation. Edwards v. Arizona, 451 U. S. 477 (1981) , decided that once “an accused has invoked his right to have counsel present during custodial interrogation … [he] is not subject to further interrogation by the authorities until counsel has been made available,” unless he initiates the contact. Id., at 484–485.

    The Edwards rule is “designed to prevent police from badgering a defendant into waiving his previously asserted Miranda rights,” Harvey, supra, at 350. It does this by presuming his postassertion statements to be involuntary, “even where the suspect executes a waiver and his statements would be considered voluntary under traditional standards.” McNeil v. Wisconsin, 501 U. S. 171, 177 (1991) . This prophylactic rule thus “protect[s] a suspect’s voluntary choice not to speak outside his lawyer’s presence.” Texas v. Cobb, 532 U. S. 162, 175 (2001) (Kennedy, J., concurring).

    Jackson represented a “wholesale importation of the Edwards rule into the Sixth Amendment .” Cobb, supra, at 175. The Jackson Court decided that a request for counsel at an arraignment should be treated as an invocation of the Sixth Amendment right to counsel “at every critical stage of the prosecution,” 475 U. S., at 633, despite doubt that defendants “actually inten[d] their request for counsel to encompass representation during any further questioning,” id., at 632–633, because doubts must be “resolved in favor of protecting the constitutional claim,” id., at 633. Citing Edwards, the Court held that any subsequent waiver would thus be “insufficient to justify police-initiated interrogation.” 475 U. S., at 635. In other words, we presume such waivers involuntary “based on the supposition that suspects who assert their right to counsel are unlikely to waive that right voluntarily” in subsequent interactions with police. Harvey, supra, at 350.

    The dissent presents us with a revisionist view of Jackson. The defendants’ request for counsel, it contends, was important only because it proved that counsel had been appointed. Such a non sequitur (nowhere alluded to in the case) hardly needs rebuttal. Proceeding from this fanciful premise, the dissent claims that the decision actually established “a rule designed to safeguard a defendant’s right to rely on the assistance of counsel,” post, at 6–7 (opinion of Stevens, J.), not one “designed to prevent police badgering,” post, at 7. To safeguard the right to assistance of counsel from what? From a knowing and voluntary waiver by the defendant himself? Unless the dissent seeks to prevent a defendant altogether from waiving his Sixth Amendment rights, i.e., to “imprison a man in his privileges and call it the Constitution,” Adams v. United States ex rel. McCann, 317 U. S. 269, 280 (1942) —a view with zero support in reason, history or case law—the answer must be: from police pressure, i.e., badgering. The antibadgering rationale is the only way to make sense of Jackson’s repeated citations of Edwards, and the only way to reconcile the opinion with our waiver jurisprudence.2

B

    With this understanding of what Jackson stands for and whence it came, it should be clear that Montejo’s interpretation of that decision—that no represented defendant can ever be approached by the State and asked to consent to interrogation—is off the mark. When a court appoints counsel for an indigent defendant in the absence of any request on his part, there is no basis for a presumption that any subsequent waiver of the right to counsel will be involuntary. There is no “initial election” to exercise the right, Patterson, 487 U. S., at 291, that must be preserved through a prophylactic rule against later waivers. No reason exists to assume that a defendant like Montejo, who has done nothing at all to express his intentions with respect to his Sixth Amendment rights, would not be perfectly amenable to speaking with the police without having counsel present. And no reason exists to prohibit the police from inquiring. Edwards and Jackson are meant to prevent police from badgering defendants into changing their minds about their rights, but a defendant who never asked for counsel has not yet made up his mind in the first instance.

    The dissent’s argument to the contrary rests on a flawed a fortiori: “If a defendant is entitled to protection from police-initiated interrogation under the Sixth Amendment when he merely requests a lawyer, he is even more obviously entitled to such protection when he has secured a lawyer.” Post, at 3. The question in Jackson, however, was not whether respondents were entitled to counsel (they unquestionably were), but “whether respondents validly waived their right to counsel,” 475 U. S., at 630; and even if it is reasonable to presume from a defendant’s request for counsel that any subsequent waiver of the right was coerced, no such presumption can seriously be entertained when a lawyer was merely “secured” on the defendant’s behalf, by the State itself, as a matter of course. Of course, reading the dissent’s analysis, one would have no idea that Montejo executed any waiver at all.

    In practice, Montejo’s rule would prevent police-initiated interrogation entirely once the Sixth Amendment right attaches, at least in those States that appoint counsel promptly without request from the defendant. As the dissent in Jackson pointed out, with no expressed disagreement from the majority, the opinion “most assuredly [did] not hold that the Edwards per se rule prohibiting all police-initiated interrogations applies from the moment the defendant’s Sixth Amendment right to counsel attaches, with or without a request for counsel by the defendant.” 475 U. S., at 640 (opinion of Rehnquist, J.). That would have constituted a “shockingly dramatic restructuring of the balance this Court has traditionally struck between the rights of the defendant and those of the larger society.” Ibid.

    Montejo’s rule appears to have its theoretical roots in codes of legal ethics, not the Sixth Amendment . The American Bar Association’s Model Rules of Professional Conduct (which nearly all States have adopted into law in whole or in part) mandate that “a lawyer shall not communicate about the subject of [a] representation with a party the lawyer knows to be represented by another lawyer in the matter, unless the lawyer has the consent of the other lawyer or is authorized to do so by law or a court order.” Model Rule 4.2 (2008). But the Constitution does not codify the ABA’s Model Rules, and does not make investigating police officers lawyers. Montejo’s proposed rule is both broader and narrower than the Model Rule. Broader, because Montejo would apply it to all agents of the State, including the detectives who interrogated him, while the ethical rule governs only lawyers. And narrower, because he agrees that if a defendant initiates contact with the police, they may talk freely—whereas a lawyer could be sanctioned for interviewing a represented party even if that party “initiates” the communication and consents to the interview. Model Rule 4.2, Comment 3.

    Montejo contends that our decisions support his interpretation of the Jackson rule. We think not. Many of the cases he cites concern the substantive scope of the Sixth Amendmente.g., whether a particular interaction with the State constitutes a “critical” stage at which counsel is entitled to be present—not the validity of a Sixth Amendment waiver. See Maine v. Moulton, 474 U. S. 159 (1985) ; Henry, 447 U. S. 264 ; Massiah, 377 U. S. 201 ; see also Moran, 475 U. S. 412 . Since everyone agrees that absent a valid waiver, Montejo was entitled to a lawyer during the interrogation, those cases do not advance his argument.

    Montejo also points to descriptions of the Jackson holding in two later cases. In one, we noted that “analysis of the waiver issue changes” once a defendant “obtains or even requests counsel.” Harvey, 494 U. S., at 352. But elsewhere in the same opinion, we explained that Jackson applies “after a defendant requests assistance of counsel,” 494 U. S., at 349; “when a suspect charged with a crime requests counsel outside the context of interrogation,” id., at 350; and to “suspects who assert their right to counsel,” ibid. The accuracy of the “obtains” language is thus questionable. Anyway, since Harvey held that evidence obtained in violation of the Jackson rule could be admitted to impeach the defendant’s trial testimony, 494 U. S., at 346, the Court’s varying descriptions of when the rule was violated were dicta. The dictum from the other decision, Patterson, supra, at 290, n. 3, is no more probative.3

    The upshot is that even on Jackson’s own terms, it would be completely unjustified to presume that a defendant’s consent to police-initiated interrogation was involuntary or coerced simply because he had previously been appointed a lawyer.

IV

    So on the one hand, requiring an initial “invocation” of the right to counsel in order to trigger the Jackson presumption is consistent with the theory of that decision, but (as Montejo and his amici argue, see Part II, supra) would be unworkable in more than half the States of the Union. On the other hand, eliminating the invocation requirement would render the rule easy to apply but depart fundamentally from the Jackson rationale.

    We do not think that stare decisis requires us to expand significantly the holding of a prior decision—fundamentally revising its theoretical basis in the process—in order to cure its practical deficiencies. To the contrary, the fact that a decision has proved “unworkable” is a traditional ground for overruling it. Payne v. Tennessee, 501 U. S. 808, 827 (1991) . Accordingly, we called for supplemental briefing addressed to the question whether Michigan v. Jackson should be overruled.

    Beyond workability, the relevant factors in deciding whether to adhere to the principle of stare decisis include the antiquity of the precedent, the reliance interests at stake, and of course whether the decision was well reasoned. Pearson v. Callahan, 555 U. S. ___, ___ (2009) (slip op., at 8). The first two cut in favor of abandoning Jackson: the opinion is only two decades old, and eliminating it would not upset expectations. Any criminal defendant learned enough to order his affairs based on the rule announced in Jackson would also be perfectly capable of interacting with the police on his own. Of course it is likely true that police and prosecutors have been trained to comply with Jackson, see generally Supplemental Brief for Larry D. Thompson et al. as Amici Curiae, but that is hardly a basis for retaining it as a constitutional requirement. If a State wishes to abstain from requesting interviews with represented defendants when counsel is not present, it obviously may continue to do so.4

    Which brings us to the strength of Jackson’s reasoning. When this Court creates a prophylactic rule in order to protect a constitutional right, the relevant “reasoning” is the weighing of the rule’s benefits against its costs. “The value of any prophylactic rule … must be assessed not only on the basis of what is gained, but also on the basis of what is lost.” Minnick, 498 U. S., at 161 (Scalia, J., dissenting). We think that the marginal benefits of Jackson (viz., the number of confessions obtained coercively that are suppressed by its bright-line rule and would otherwise have been admitted) are dwarfed by its substantial costs (viz., hindering “society’s compelling interest in finding, convicting, and punishing those who violate the law,” Moran, supra, at 426).

    What does the Jackson rule actually achieve by way of preventing unconstitutional conduct? Recall that the purpose of the rule is to preclude the State from badgering defendants into waiving their previously asserted rights. See Harvey, supra, at 350; see also McNeil, 501 U. S., at 177. The effect of this badgering might be to coerce a waiver, which would render the subsequent interrogation a violation of the Sixth Amendment . See Massiah, supra, at 204. Even though involuntary waivers are invalid even apart from Jackson, see Patterson, 487 U. S., at 292, n. 4, mistakes are of course possible when courts conduct case-by-case voluntariness review. A bright-line rule like that adopted in Jackson ensures that no fruits of interrogations made possible by badgering-induced involuntary waivers are ever erroneously admitted at trial.

    But without Jackson, how many would be? The answer is few if any. The principal reason is that the Court has already taken substantial other, overlapping measures toward the same end. Under Miranda’s prophylactic protection of the right against compelled self-incrimination, any suspect subject to custodial interrogation has the right to have a lawyer present if he so requests, and to be advised of that right. 384 U. S., at 474. Under Edwards’ prophylactic protection of the Miranda right, once such a defendant “has invoked his right to have counsel present,” interrogation must stop. 451 U. S., at 484. And under Minnick’s prophylactic protection of the Edwards right, no subsequent interrogation may take place until counsel is present, “whether or not the accused has consulted with his attorney.” 498 U. S., at 153.

    These three layers of prophylaxis are sufficient. Under the Miranda-Edwards-Minnick line of cases (which is not in doubt), a defendant who does not want to speak to the police without counsel present need only say as much when he is first approached and given the Miranda warnings. At that point, not only must the immediate contact end, but “badgering” by later requests is prohibited. If that regime suffices to protect the integrity of “a suspect’s voluntary choice not to speak outside his lawyer’s presence” before his arraignment, Cobb, 532 U. S., at 175 (Kennedy, J., concurring), it is hard to see why it would not also suffice to protect that same choice after arraignment, when Sixth Amendment rights have attached. And if so, then Jackson is simply superfluous.

    It is true, as Montejo points out in his supplemental brief, that the doctrine established by Miranda and Edwards is designed to protect Fifth Amendment , not Sixth Amendment , rights. But that is irrelevant. What matters is that these cases, like Jackson, protect the right to have counsel during custodial interrogation—which right happens to be guaranteed (once the adversary judicial process has begun) by two sources of law. Since the right under both sources is waived using the same procedure, Patterson, supra, at 296, doctrines ensuring voluntariness of the Fifth Amendment waiver simultaneously ensure the voluntariness of the Sixth Amendment waiver.

    Montejo also correctly observes that the Miranda-Edwards regime is narrower than Jackson in one respect: The former applies only in the context of custodial interrogation. If the defendant is not in custody then those decisions do not apply; nor do they govern other, noninterrogative types of interactions between the defendant and the State (like pretrial lineups). However, those uncovered situations are the least likely to pose a risk of coerced waivers. When a defendant is not in custody, he is in control, and need only shut his door or walk away to avoid police badgering. And noninterrogative interactions with the State do not involve the “inherently compelling pressures,” Miranda, supra, at 467, that one might reasonably fear could lead to involuntary waivers.

    Jackson was policy driven, and if that policy is being adequately served through other means, there is no reason to retain its rule. Miranda and the cases that elaborate upon it already guarantee not simply noncoercion in the traditional sense, but what Justice Harlan referred to as “voluntariness with a vengeance,” 384 U. S., at 505 (dissenting opinion). There is no need to take Jackson’sfurther step of requiring voluntariness on stilts.

    On the other side of the equation are the costs of adding the bright-line Jackson rule on top of Edwards and other extant protections. The principal cost of applying any exclusionary rule “is, of course, letting guilty and possibly dangerous criminals go free … .” Herring v. United States, 555 U. S. ___, ___ (2009) (slip op., at 6). Jackson not only “operates to invalidate a confession given by the free choice of suspects who have received proper advice of their Miranda rights but waived them nonetheless,” Cobb, supra, at 174–175 (Kennedy, J., concurring), but alsodeters law enforcement officers from even trying to obtain voluntary confessions. The “ready ability to obtain uncoerced confessions is not an evil but an unmitigated good.” McNeil, 501 U. S., at 181. Without these confessions, crimes go unsolved and criminals unpunished. These are not negligible costs, and in our view the Jackson Court gave them too short shrift.5

    Notwithstanding this calculus, Montejo and his amici urge the retention of Jackson. Their principalobjection to its elimination is that the Edwards regime which remains will not provide an administrable rule. But this Court has praised Edwards precisely because it provides “ ‘clear and unequivocal’ guidelines to the law enforcement profession,” Arizona v. Roberson, 486 U. S. 675, 682 (1988) . Our cases make clear which sorts of statements trigger its protections, see Davis v. United States, 512 U. S. 452, 459 (1994) , and once triggered, the rule operates as a bright line. Montejo expresses concern that courts will have to determine whether statements made at preliminary hearings constitute Edwards invocations—thus implicating all the practical problems of the Louisiana rule we discussed above, see Part II, supra. That concern is misguided. “We have in fact never held that a person can invoke his Miranda rights anticipatorily, in a context other than ‘custodial interrogation’… .” McNeil, supra, at 182, n. 3. What matters for Miranda and Edwards is what happens when the defendant is approached for interrogation, and (if he consents) what happens during the interrogation—not what happened at any preliminary hearing.

    In sum, when the marginal benefits of the Jackson rule are weighed against its substantial costs to the truth-seeking process and the criminal justice system, we readily conclude that the rule does not “pay its way,” United States v. Leon, 468 U. S. 897 , n. 6 (1984). Michigan v. Jackson should be and now is overruled.

V

    Although our holding means that the Louisiana Supreme Court correctly rejected Montejo’s claim under Jackson, we think that Montejo should be given an opportunity to contend that his letter of apology should still have been suppressed under the rule of Edwards. If Montejo made a clear assertion of the right to counsel when the officers approached him about accompanying them on the excursion for the murder weapon, then no interrogation should have taken place unless Montejo initiated it. Davis, supra, at 459. Even if Montejo subsequently agreed to waive his rights, that waiver would have been invalid had it followed an “unequivocal election of the right,” Cobb, 532 U. S., at 176 (Kennedy, J., concurring).

    Montejo understandably did not pursue an Edwards objection, because Jackson served as the Sixth Amendment analogy to Edwards and offered broader protections. Our decision today, overruling Jackson, changes the legal landscape and does so in part based on the protections already provided by Edwards. Thus we think that a remand is appropriate so that Montejo can pursue this alternative avenue for relief. Montejo may also seek on remand to press any claim he might have that his Sixth Amendment waiver was not knowing and voluntary, e.g., his argument that the waiver was invalid because it was based on misrepresentations by police as to whether he had been appointed a lawyer, cf. Moran, 475 U. S., at 428–429. These matters have heightened importance in light of our opinion today.

    We do not venture to resolve these issues ourselves, not only because we are a court of final review, “not of first view,” Cutter v. Wilkinson, 544 U. S. 709 , n. 7 (2005), but also because the relevant facts remain unclear. Montejo and the police gave inconsistent testimony about exactly what took place on the afternoon of September 10, 2002, and the Louisiana Supreme Court did not make an explicit credibility determination. Moreover, Montejo’s testimony came not at the suppression hearing, but rather only at trial, and we are unsure whether under state law that testimony came too late to affect the propriety of the admission of the evidence. These matters are best left for resolution on remand.

    We do reject, however, the dissent’s revisionist legal analysis of the “knowing and voluntary” issue. Post, at 10–14. In determining whether a Sixth Amendment waiver was knowing and voluntary, there is no reason categorically to distinguish an unrepresented defendant from a represented one. It is equally true for each that, as we held in Patterson,the Miranda warnings adequately inform him “of his right to have counsel present during the questioning,” and make him “aware of the consequences of a decision by him to waive his Sixth Amendment rights,” 487 U. S., at 293. Somewhat surprisingly for an opinion that extols the virtues of stare decisis, the dissent complains that our “treatment of the waiver question rests entirely on the dubious decision in Patterson,” post, at 12. The Court in Patterson did not consider the result dubious, nor does the Court today.

*  *  *

    This case is an exemplar of Justice Jackson’s oft quoted warning that this Court “is forever adding new stories to the temples of constitutional law, and the temples have a way of collapsing when one story too many is added.” Douglas v. City of Jeannette, 319 U. S. 157, 181 (1943) (opinion concurring in result). We today remove Michigan v. Jackson’s fourth story of prophylaxis.

    The judgment of the Louisiana Supreme Court is vacated, and the case is remanded for further proceedings not inconsistent with this opinion.

It is so ordered.


Notes

1 “The sheriff or law enforcement officer having custody of an arrested person shall bring him promptly, and in any case within seventy-two hours from the time of the arrest, before a judge for the purpose of appointment of counsel.” La. Code Crim. Proc. Ann., Art. 230.1(A) (West Supp. 2009).

2 The dissent responds that Jackson also ensures that the defendant’s counsel receives notice of any interrogation, post, at 6, n. 2. But notice to what end? Surely not in order to protect some constitutional right to receive counsel’s advice regarding waiver of the right to have counsel present. Contrary to the dissent’s intimations, neither the advice nor the presence of counsel is needed in order to effectuate a knowing waiver of the Sixth Amendment right. Our cases make clear that the Miranda waivers typically suffice; indeed, even an unrepresented defendant can waive his right to counsel. See supra, at 7.

3 In the cited passage, the Court noted that “[o]nce an accused has a lawyer, a distinct set of constitutional safeguards aimed at preserving the sanctity of attorney-client relationship takes effect.” Patterson, 487 U. S., at 290, n. 3. To support that proposition, the Court cited Maine v. Moulton, 474 U. S. 159 (1985) , which was not a case about waiver. The passage went on to observe that “the analysis changes markedly once an accused even requests the assistance of counsel,” 487 U. S., at 290, n. 3 (emphasis in original), this time citing Jackson. Montejo infers from the “even requests” that having counsel is more conclusive of the invalidity of uncounseled waiver than the mere requesting of counsel. But the Patterson footnote did not suggest that the analysis “changes” in both these scenarios (having a lawyer, versus requesting one) with specific reference to the validity of waivers under the Sixth Amendment . The citation of Moulton (a nonwaiver case) for the first scenario suggests just the opposite.

4 The dissent posits a different reliance interest: “the public’s interest in knowing that counsel, once secured, may be reasonably relied upon as a medium between the accused and the power of the State,” post, at 9. We suspect the public would be surprised to learn that a criminal can freely sign away his right to a lawyer, confess his crimes, and then ask the courts to assume that the confession was coerced—on the ground that he had, at some earlier point in time, made a pro forma statement requesting that counsel be appointed on his behalf.

5 The dissent claims that, in fact, few confessions have been suppressed by federal courts applying Jackson. Post, at 8. If so, that is because, as the dissent boasts, “generations of police officers have been trained to refrain from approaching represented defendants,” post, at 9, n. 4. Anyway, if the rule truly does not hinder law enforcement or make much practical difference, see post, at 7–9, and nn. 3–4, then there is no reason to be particularly exercised about its demise.