496 U.S. 292
SCOTUS
1990

Lead Opinion

Justice Kennedy

delivered the opinion of the Court.

An undercover government agent was placed in the cell of respondent Perkins, who was incarcerated on charges unrelated to the subject of the agent’s investigation. Respondent made statements that implicated him in the crime that the agent sought to solve. Respondent claims that the statements should be inadmissible because he had not been given Miranda warnings by the agent. We hold that the statements are admissible. Miranda warnings are not required when the suspect is unaware that he is speaking to a law enforcement officer and gives a voluntary statement.

I

In November 1984, Richard Stephenson was murdered in a suburb of East St. Louis, Illinois. The murder remained unsolved until March 1986, when one Donald Charlton told police that he had learned about a homicide from a fellow inmate at the Graham Correctional Facility, where Charlton had been serving a sentence for burglary. The fellow inmate was Lloyd Perkins, who is the respondent here. Charlton told police that, while at Graham, he had befriended respondent, who told him in detail about a murder that respondent had committed in East St. Louis. On hearing Charlton’s account, the police recognized details of the Stephenson murder that were not well known, and so they treated Charlton’s story as a credible one.

By the time the police heard Charlton’s account, respondent had been released from Graham, but police traced him to a jail in Montgomery County, Illinois, where he was being held pending trial on a charge of aggravated battery, unrelated to the Stephenson murder. The police wanted to investigate further respondent’s connection to the Stephenson murder, but feared that the use of an eavesdropping device would prove impracticable and unsafe. They decided instead to place an undercover agent in the cellblock with respondent and Charlton. The plan was for Charlton and un dercover agent John Parisi to pose as escapees from a work release program who had been arrested in the course of a burglary. Parisi and Charlton were instructed to engage respondent in casual conversation and report anything he said about the Stephenson murder.

Parisi, using the alias “Vito Bianco,” and Charlton, both clothed in jail garb, were placed in the cellblock with respondent at the Montgomery County jail. The cellblock consisted of 12 separate cells that opened onto a common room. Respondent greeted Charlton who, after a brief conversation with respondent, introduced Parisi by his alias. Parisi told respondent that he “wasn’t going to do any more time” and suggested that the three of them escape. Respondent replied that the Montgomery County jail was “rinky-dink” and that they could “break out.” The trio met in respondent’s cell later that evening, after the other inmates were asleep, to refine their plan. Respondent said that his girlfriend could smuggle in a pistol. Charlton said: “Hey, Pm not a murderer, Pm a burglar. That’s your guys’ profession.” After telling Charlton that he would be responsible for any murder that occurred, Parisi asked respondent if he had ever “done” anybody. Respondent said that he had and proceeded to describe at length the events of the Stephenson murder. Parisi and respondent then engaged in some casual conversation before respondent went to sleep. Parisi did not give respondent Miranda warnings before the conversations.

Respondent was charged with the Stephenson murder. Before trial, he moved to suppress the statements made to Parisi in the jail. The trial court granted the motion to suppress, and the State appealed. The Appellate Court of Illinois affirmed, 176 Ill. App. 3d 443, 531 N. E. 2d 141 (1988), holding that Miranda v. Arizona, 384 U. S. 436 (1966), prohibits all undercover contacts with incarcerated suspects that are reasonably likely to elicit an incriminating response.

We granted certiorari, 493 U. S. 808 (1989), to decide whether an undercover law enforcement officer must give Miranda warnings to an incarcerated suspect before asking him questions that may elicit an incriminating response. We now reverse.

II

In Miranda v. Arizona, supra, the Court held that the Fifth Amendment privilege against self-incrimination prohibits admitting statements given by a suspect during “custodial interrogation” without a prior warning. Custodial interrogation means “questioning initiated by law enforcement officers after a person has been taken into custody . . . .” Id., at 444. The warning mandated by Miranda was meant to preserve the privilege during “incommunicado interrogation of individuals in a police-dominated atmosphere.” Id., at 445. That atmosphere is said to generate “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.” Id., at 467. “Fidelity to the doctrine announced in Miranda requires that it be enforced strictly, but only in those types of situations in which the concerns that powered the decision are implicated.” Berkemer v. McCarty, 468 U. S. 420, 437 (1984).

Conversations between suspects and undercover agents do not implicate the concerns underlying Miranda. The essential ingredients of a “police-dominated atmosphere” and compulsion are not present when an incarcerated person speaks freely to someone whom he believes to be a fellow inmate. Coercion is determined from the perspective of the suspect. Rhode Island v. Innis, 446 U. S. 291, 301 (1980); Berkemer v. McCarty, supra, at 442. When a suspect considers himself in the company of cellmates and not officers, the coercive atmosphere is lacking. Miranda, 384 U. S., at 449 (“[T]he ‘principal psychological factor contributing to a successful interrogation is privacy—being alone with the person under interrogation’”); id., at 445. There is no empirical basis for the assumption that a suspect speaking to those whom he assumes are not officers will feel compelled to speak by the fear of reprisal for remaining silent or in the hope of more lenient treatment should he confess.

It is the premise of Miranda that the danger of coercion results from the interaction of custody and official interrogation. We reject the argument that Miranda warnings are required whenever a suspect is in custody in a technical sense and converses with someone who happens to be a government agent. Questioning by captors, who appear to control the suspect’s fate, may create mutually reinforcing pressures that the Court has assumed will weaken the suspect’s will, but where a suspect does not know that he is conversing with a government agent, these pressures do not exist. The state court here mistakenly assumed that because the suspect was in custody, no undercover questioning could take place. When the suspect has no reason to think that the listeners have official power over him, it should not be assumed that his words are motivated by the reaction he expects from his listeners. “[W]hen the agent carries neither badge nor gun and wears not ‘police blue,’ but the same prison gray” as the suspect, there is no “interplay between police interrogation and police custody.” Kamisar, Brewer v. Williams, Massiah and Miranda: What is “Interrogation”? When Does it Matter?, 67 Geo. L. J. 1, 67, 63 (1978).

Miranda forbids coercion, not mere strategic deception by taking advantage of a suspect’s misplaced trust in one he supposes to be a fellow prisoner. As we recognized in Miranda: “Confessions remain a proper element in law enforcement. Any statement given freely and voluntarily without any compelling influences is, of course, admissible in evidence.” 384 U. S., at 478. Ploys to mislead a suspect or lull him into a false sense of security that do not rise to the level of compulsion or coercion to speak are not within Miranda’s concerns. Cf. Oregon v. Mathiason, 429 U. S. 492, 495-496 (1977) (per curiam); Moran v. Burbine, 475 U. S. 412 (1986) (where police fail to inform suspect of attorney’s efforts to reach him, neither Miranda nor the Fifth Amendment requires suppression of prearraignment confession after voluntary waiver).

Miranda was not meant to protect suspects from boasting about their criminal activities in front of persons whom they believe to be their cellmates. This case is illustrative. Respondent had no reason to feel that undercover agent Parisi had any legal authority to force him to answer questions or that Parisi could affect respondent’s future treatment. Respondent viewed the cellmate-agent as an equal and showed no hint of being intimidated by the atmosphere of the jail. In recounting the details of the Stephenson murder, respondent was motivated solely by the desire to impress his fellow inmates. He spoke at his own peril.

The tactic employed here to elicit a voluntary confession from a suspect does not violate the Self-Incrimination Clause. We held in Hoffa v. United States, 385 U. S. 293 (1966), that placing an undercover agent near a suspect in order to gather incriminating information was permissible under the Fifth Amendment. In Hoffa, while petitioner Hoffa was on trial, he met often with one Partin, who, unbeknownst to Hoffa, was cooperating with law enforcement officials. Partin reported to officials that Hoffa had divulged his attempts to bribe jury members. We approved using Hoffa’s statements at his subsequent trial for jury tampering, on the rationale that “no claim ha[d] been or could [have been] made that [Hoffa’s] incriminating statements were the product of any sort of coercion, legal or factual.” Id., at 304. In addition, we found that the fact that Partin had fooled Hoffa into thinking that Partin was a sympathetic colleague did not affect the volun-tariness of the statements. Ibid. Cf. Oregon v. Mathiason, supra, at 495-496 (officer’s falsely telling suspect that suspect’s fingerprints had been found at crime scene did not render interview “custodial” under Miranda); Frazier v. Cupp, 394 U. S. 731, 739 (1969); Procunier v. Atchley, 400 U. S. 446, 453-454 (1971). The only difference between this case and Hoffa is that the suspect here was incarcerated, but detention, whether or not for the crime in question, does not warrant a presumption that the use of an undercover agent to speak with an incarcerated suspect makes any confession thus obtained involuntary.

Our decision in Mathis v. United States, 391 U. S. 1 (1968), is distinguishable. In Mathis, an inmate in a state prison was interviewed by an Internal Revenue Service agent about possible tax violations. No Miranda warning was given before questioning. The Court held that the suspect’s incriminating statements were not admissible at his subsequent trial on tax fraud charges. The suspect in Mathis was aware that the agent was a Government official, investigating the possibility of noncompliance with the tax laws. The case before us now is different. Where the suspect does not know that he is speaking to a government agent there is no reason to assume the possibility that the suspect might feel coerced. (The bare fact of custody may not in every instance require a warning even when the suspect is aware that he is speaking to an official, but we do not have occasion to explore that issue here.)

This Court’s Sixth Amendment decisions in Massiah v. United States, 377 U. S. 201 (1964), United States v. Henry, 447 U. S. 264 (1980), and Maine v. Moulton, 474 U. S. 159 (1985), also do not avail respondent. We held in those cases that the government may not use an undercover agent to circumvent the Sixth Amendment right to counsel once a suspect has been charged with the crime. After charges have been filed, the Sixth Amendment prevents the government from interfering with the accused’s right to counsel. Moulton, supra, at 176. In the instant case no charges had been filed on the subject of the interrogation, and our Sixth Amendment precedents are not applicable.

Respondent can seek no help from his argument that a bright-line rule for the application of Miranda is desirable. Law enforcement officers will have little difficulty putting into practice our holding that undercover agents need not

give Miranda warnings to incarcerated suspects. The use of undercover agents is a recognized law enforcement technique, often employed in the prison context to detect violence against correctional officials or inmates, as well as for the purposes served here. The interests protected by Miranda are not implicated in these cases, and the warnings are not required to safeguard the constitutional rights of inmates who make voluntary statements to undercover agents.

We hold that an undercover law enforcement officer posing as a fellow inmate need not give Miranda warnings to an incarcerated suspect before asking questions that may elicit an incriminating response. The statements at issue in this case were voluntary, and there is no federal obstacle to their admissibility at trial. We now reverse and remand for proceedings not inconsistent with our opinion.

It is so ordered.





Concurrence Opinion

Justice Brennan,

concurring in the judgment.

The Court holds that Miranda v. Arizona, 384 U. S. 436 (1966), does not require suppression of a statement made by an incarcerated suspect to an undercover agent. Although I do not subscribe to the majority’s characterization of Miranda in its entirety, I do agree that when a suspect does not know that his questioner is a police agent, such questioning does not amount to “interrogation” in an “inherently coercive” environment so as to require application of Miranda. Since the only issue raised at this stage of the litigation is the applicability of Miranda,* I concur in the judgment of the Court.

This is not to say that I believe the Constitution condones the method by which the police extracted the confession in this case. To the contrary, the deception and manipulation practiced on respondent raise a substantial claim that the confession was obtained in violation of the Due Process Clause. As we recently stated in Miller v. Fenton, 474 U. S. 104, 109-110 (1985):

“This Court has long held that certain interrogation techniques, either in isolation or as applied to the unique characteristics of a particular suspect, are so offensive to a civilized system of justice that they must be condemned under the Due Process Clause of the Fourteenth Amendment. . . . Although these decisions framed the legal inquiry in a variety of different ways, usually through the ‘convenient shorthand’ of asking whether the confession was ‘involuntary,’ Blackburn v. Alabama, 361 U. S. 199, 207 (1960), the Court’s analysis has consistently been animated by the view that ‘ours is an accusatorial and not an inquisitorial system,’ Rogers v. Richmond, 365 U. S. 534, 541 (1961), and that, accordingly, tactics for eliciting inculpatory statements must fall within the broad constitutional boundaries imposed by the Fourteenth Amendment’s guarantee of fundamental fairness.”

That the right is derived from the Due Process Clause “is significant because it reflects the Court’s consistently held view that 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.” Id., at 116. See Spano v. New York, 360 U. S. 315, 320-321 (1959) (“The abhorrence of society to the use of involuntary confessions does not turn alone on their inherent untrustworthiness. It also turns on the deep-rooted feeling that the police must obey the law while enforcing the law; that in the end life and liberty can be as much endangered from illegal methods used to convict those thought to be criminals as from the actual criminals themselves”); see also Degraffenreid v. McKellar, 494 U. S. 1071, 1072-1074 (1990) (Marshall, J., joined by Brennan, J., dissenting from denial of certiorari).

The method used to elicit the confession in this case deserves close scrutiny. The police devised a ruse to lure respondent into incriminating himself when he was in jail on an unrelated charge. A police agent, posing as a fellow inmate and proposing a sham escape plot, tricked respondent into confessing that he had once committed a murder, as a way of proving that he would be willing to do so again should the need arise during the escape. The testimony of the undercover officer and a police informant at the suppression hearing reveal the deliberate manner in which the two elicited incriminating statements from respondent. See App. 43-53 and 66-73. We have recognized that “the mere fact of custody imposes pressures on the accused; confinement may bring into play subtle influences that will make him particularly susceptible to the ploys of undercover Government agents.” United States v. Henry, 447 U. S. 264, 274 (1980). As Justice Marshall points out, the pressures of custody make a suspect more likely to confide in others and to engage in “jailhouse bravado.” See post, at 307-308. The State is in a unique position to exploit this vulnerability because it has virtually complete control over the suspect’s environment. Thus, the State can ensure that a suspect is barraged with questions from an undercover agent until the suspect confesses. Cf. Mincey v. Arizona, 437 U. S. 385, 399 (1978); Ashcraft v. Tennessee, 322 U. S. 143, 153-155 (1944). The testimony in this case suggests the State did just that.

The deliberate use of deception and manipulation by the police appears to be incompatible “with a system that presumes innocence and assures that a conviction will not be secured by inquisitorial means,” Miller, supra, at 116, and raises serious concerns that respondent’s will was overborne. It is open to the lower court on remand to determine whether, under the totality of the circumstances, respondent’s confession was elicited in a manner that violated the Due Process Clause. That the confession was not elicited through means of physical torture, see Brown v. Mississippi, 297 U. S. 278 (1936) or overt psychological pressure, see Payne v. Arkansas, 356 U. S. 560, 566 (1958), does not end the inquiry. “[A]s law enforcement officers become more responsible, and the methods used to extract confessions more sophisticated, [a court’s] duty to enforce federal constitutional protections does not cease. It only becomes more difficult because of the more delicate judgments to be made.” Spano, supra, at 321.

Notes

As the case comes to us, it involves only the question whether Miranda applies to the questioning of an incarcerated suspect by an undercover agent. Nothing in the Court’s opinion suggests that, had respondent previously invoked his Fifth Amendment right to counsel or right to silence, his statements would be admissible. If respondent had invoked either right, the inquiry would focus on whether he subsequently waived the particular right. See Edwards v. Arizona, 451 U. S. 477 (1981); Michigan v. Mosley 423 U. S. 96, 104 (1975). As the Court made clear in Moran v. Burbine, 475 U. S. 412, 421 (1986), the waiver of Miranda rights “must [be] voluntary in the sense that it [must be] the product of a free and deliberate choice rather than intimidation, coercion or deception." (Emphasis added.) Since respondent was in custody on an unrelated charge when he was questioned, he may be able to challenge the admission of these statements if he previously had invoked his Miranda rights with respect to that charge. See Arizona v. Roberson, 486 U. S. 675 (1988); Mosley, supra, at 104. Similarly, if respondent had been formally charged on the unrelated charge and had invoked his Sixth Amendment right to counsel, he may have a Sixth Amendment challenge to the admissibility of these statements. See Michigan v. Jackson, 475 U. S. 625, 629-636 (1986). Cf. Roberson, supra, at 683-685.



Dissenting Opinion

Justice Marshall,

dissenting.

This Court clearly and simply stated its holding in Miranda v. Arizona, 384 U. S. 436 (1966): “[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.” Id., at 444. The conditions that require the police to apprise a defendant of his constitutional rights—custodial interrogation conducted by an agent of the police—were present in this case. Because Lloyd Perkins received no Miranda warnings before he was subjected to custodial interrogation, his confession was not admissible.

The Court reaches the contrary conclusion by fashioning an exception to the Miranda rule that applies whenever “an undercover law enforcement officer posing as a fellow inmate . . . ask[s] questions that may elicit an incriminating response” from an incarcerated suspect. Ante, at 300. This exception is inconsistent with the rationale supporting Miranda and allows police officers intentionally to take advantage of suspects unaware of their constitutional rights. I therefore dissent.

The Court does not dispute that the police officer here conducted a custodial interrogation of a criminal suspect. Perkins was incarcerated in county jail during the questioning at issue here; under these circumstances, he was in custody as that term is defined in Miranda. 384 U. S., at 444; Mathis v. United States, 391 U. S. 1, 4-5 (1968) (holding that defendant incarcerated on charges different from the crime about which he is questioned was in custody for purposes of Miranda). The United States argues that Perkins was not in custody for purpose of Miranda because he was familiar with the custodial environment as a result of being in jail for two days and previously spending time in prison. Brief for United States as Amicus Curiae 11. Perkins’ familiarity with confinement, however, does not transform his incarceration into some sort of noncustodial arrangement. Cf. Orozco v. Texas, 394 U. S. 324 (1969) (holding that suspect who had been arrested in his home and then questioned in his bedroom was in custody, notwithstanding his familiarity with the surroundings).

While Perkins was confined, an undercover police officer, with the help of a police informant, questioned him about a serious crime. Although the Court does not dispute that Perkins was interrogated, it downplays the nature of the 35-minute questioning by disingenuously referring to it as a “conversation.” Ante, at 295, 296. The officer’s narration of the “conversation” at Perkins’ suppression hearing, however, reveals that it clearly was an interrogation.

“[Agent:] You ever do anyone?
“[Perkins:] Yeah, once in East St. Louis, in a rich white neighborhood.
“Informant: I didn’t know they had any rich white neighborhoods in East St. Louis.
“Perkins: It wasn’t in East St. Louis, it was by a race track in Fairview Heights. . . .
“[Agent]: You did a guy in Fairview Heights?
“Perkins: Yeah in a rich white section where most of the houses look the same.
“[Informant]: If all the houses look the same, how did you know you had the right house?
“Perkins: Me and two guys cased the house for about a week. I knew exactly which house, the second house on the left from the corner.
“[Agent]: How long ago did this happen?
“Perkins: Approximately about two years ago. I got paid $5,000 for that job.
“[Agent]: How did it go down?
“Perkins: I walked up [to] this guy[’s] house with a sawed-off under my trench coat.
“[Agent]: What type gun[?]
“Perkins: A .12 gauge Remmington [sic] Automatic Model 1100 sawed-off.” App. 49-50.

The police officer continued the inquiry, asking a series of questions designed to elicit specific information about the victim, the crime scene, the weapon, Perkins’ motive, and his actions during and after the shooting. Id., at 50-52. This interaction was not a “conversation”; Perkins, the officer, and the informant were not equal participants in a free-ranging discussion, with each man offering his views on different topics. Rather, it was an interrogation: Perkins was subjected to express questioning likely to evoke an incriminating re sponse. Rhode Island v. Innis, 446 U. S. 291, 300-301 (1980).

Because Perkins was interrogated by police while he was in custody, Miranda required that the officer inform him of his rights. In rejecting that conclusion, the Court finds that “conversations” between undercover agents and suspects are devoid of the coercion inherent in station house interrogations conducted by law enforcement officials who openly represent the State. Ante, at 296. Miranda was not, however, concerned solely with police coercion. It dealt with any police tactics that may operate to compel a suspect in custody to make incriminating statements without full awareness of his constitutional rights. See Miranda, supra, at 468 (referring to “inherent pressures of the interrogation atmosphere”); Estelle v. Smith, 451 U. S. 454, 467 (1981) (“The purpose of [the Miranda] admonitions is to combat what the Court saw as ‘inherently compelling pressures’ at work on the person and to provide him with an awareness of the Fifth Amendment privilege and the consequences of forgoing it”) (quoting Miranda, 384 U. S., at 467). Thus, when a law enforcement agent structures a custodial interrogation so that a suspect feels compelled to reveal incriminating information, he must inform the suspect of his constitutional rights and give him an opportunity to decide whether or not to talk.

The compulsion proscribed by Miranda includes deception by the police. See Miranda, supra, at 453 (indicting police tactics “to induce a confession out of trickery,” such as using fictitious witnesses or false accusations); Berkemer v. McCarty, 468 U. S. 420, 433 (1984) (“The purposes of the safeguards prescribed by Miranda are to ensure that the police do not coerce or trick captive suspects into confessing”) (emphasis deleted and added). Cf. Moran v. Burbine, 475 U. S. 412, 421 (1986) (“[T]he relinquishment of the right [protected by the Miranda warnings] must have been voluntary in the sense that it was the product of a free and deliberate choice rather than intimidation, coercion, or deception”) (em phasis added). Although the Court did not find trickery by itself sufficient to constitute compulsion in Hoffa v. United States, 385 U. S. 293 (1966), the defendant in that case was not in custody. Perkins, however, was interrogated while incarcerated. As the Court has acknowledged in the Sixth Amendment context: “[T]he mere fact of custody imposes pressures on the accused; confinement may bring into play subtle influences that will make him particularly susceptible to the ploys of undercover Government agents.” United States v. Henry, 447 U. S. 264, 274 (1980). See also Massiah v. United States, 377 U. S. 201, 206 (1964) (holding, in the context of the Sixth Amendment, that defendant’s constitutional privilege against self-incrimination was “more seriously imposed upon . . . because he did not even know that he was under interrogation by a government agent”) (citation and internal quotation marks omitted).

Custody works to the State’s advantage in obtaining incriminating information. The psychological pressures inherent in confinement increase the suspect’s anxiety, making him likely to seek relief by talking with others. Dix, Undercover Investigations and Police Rulemaking, 53 Texas L. Rev. 203, 230 (1975). See also Gibbs, The First Cut is the Deepest: Psychological Breakdown and Survival in the Detention Setting, in The Pains of Imprisonment 97, 107 (R. Johnson & H. Toch eds. 1982); Hagel-Seymour, Environmental Sanctuaries for Susceptible Prisoners, in The Pains of Imprisonment, supra, at 267, 279; Chicago Tribune, Apr. 15, 1990, p. D3 (prosecutors have found that prisoners often talk freely with fellow inmates). The inmate is thus more susceptible to efforts by undercover agents to elicit information from him. Similarly, where the suspect is incarcerated, the constant threat of physical danger peculiar to the prison environment may make him demonstrate his toughness to other inmates by recounting or inventing past violent acts. “Because the suspect’s ability to select people with whom he can confide is completely within their control, the police have a unique opportunity to exploit the suspect’s vulnerability. In short, the police can insure that if the pressures of confinement lead the suspect to confide in anyone, it will be a police agent.” (Footnote omitted.) White, Police Trickery in Inducing Confessions, 127 U. Pa. L. Rev. 581, 605 (1979). In this case, the police deceptively took advantage of Perkins’ psychological vulnerability by including him in a sham escape plot, a situation in which he would feel compelled to demonstrate his willingness to shoot a prison guard by revealing his past involvement in a murder. See App. 49 (agent stressed that a killing might be necessary in the escape and then asked Perkins if he had ever murdered someone).

Thus, the pressures unique to custody allow the police to use deceptive interrogation tactics to compel a suspect to make an incriminating statement. The compulsion is not eliminated by the suspect’s ignorance of his interrogator’s true identity. The Court therefore need not inquire past the bare facts of custody and interrogation to determine whether Miranda warnings are required.

The Court’s adoption of an exception to the Miranda doctrine is incompatible with the principle, consistently applied by this Court, that the doctrine should remain simple and clear. See, e. g., Miranda, supra, at 441-442 (noting that one reason certiorari was granted was “to give concrete constitutional guidelines for law enforcement agencies and courts to follow”); McCarty, supra, at 430 (noting that one of “the principal advantages of the [Miranda] doctrine . . . is the clarity of that rule”); Arizona v. Roberson, 486 U. S. 675, 680 (1988) (same). See also New York v. Quarles, 467 U. S. 649, 657-658 (1984) (recognizing need for clarity in Miranda doctrine and finding that narrow “public safety” exception would not significantly lessen clarity and would be easy for police to apply). We explained the benefits of a brightline rule in Fare v. Michael C., 442 U. S. 707 (1979): “Miranda’s holding has the virtue of informing police and prosecutors with specificity as to what they may do in conducting custo dial interrogation, and of informing courts under what circumstances statements obtained during such interrogation are not admissible.” Id., at 718.

The Court’s holding today complicates a previously clear and straightforward doctrine. The Court opines that “[l]aw enforcement officers will have little difficulty putting into practice our holding that undercover agents need not give Miranda warnings to incarcerated suspects.” Ante, at 299-300. Perhaps this prediction is true with respect to fact patterns virtually identical to the one before the Court today. But the outer boundaries of the exception created by the Court are by no means clear. Would Miranda be violated, for instance, if an undercover police officer beat a confession out of a suspect, but the suspect thought the officer was another prisoner who wanted the information for his own purposes?

Even if Miranda, as interpreted by the Court, would not permit such obviously compelled confessions, the ramifications of today’s opinion are still disturbing. The exception carved out of the Miranda doctrine today may well result in a proliferation of departmental policies to encourage police officers to conduct interrogations of confined suspects through undercover agents, thereby circumventing the need to administer Miranda warnings. Indeed, if Miranda now requires a police officer to issue warnings only in those situations in which the suspect might feel compelled “to speak by the fear of reprisal for remaining silent or in the hope of more lenient treatment should he confess,” ante, at 296-297, presumably it allows custodial interrogation by an undercover officer posing as a member of the clergy or a suspect’s defense attorney. Although such abhorrent tricks would play on a suspect’s need to confide in a trusted adviser, neither would cause the suspect to “think that the listeners have official power over him,” ante, at 297. The Court’s adoption of the “undercover agent” exception to the Miranda rule thus is necessarily also the adoption of a substantial loophole in our jurisprudence protecting suspects’ Fifth Amendment rights.

I dissent.