State v. Vondehn

236 P.3d 691, 348 Or. 462, 2010 Ore. LEXIS 504

• Court: Oregon Supreme Court
• Decided: July 1, 2010
• Docket: CC C040956CR; CA A128800; SC S056371
• Status: Published

Opinions

WALTERS, J.

This criminal case raises two questions of first impression regarding the right against compelled self-incrimination in Article I, section 12, of the Oregon Constitution¹ and the consequences of the failure to give Miranda warnings to a person who is in custody and subjected to custodial interrogation.

The uncontested facts establish that the police arrested defendant on a warrant, handcuffed him, and placed him in the back seat of a patrol car. A police officer then asked defendant two questions about a backpack that the officer had found in the car in which defendant had been a passenger. In response, defendant admitted that he owned the backpack and that it contained marijuana. The trial court suppressed those answers because the police had failed to administer the required Miranda warnings. The trial court did not, however, suppress the marijuana, ruling that defendant’s answer to the next question that the officer asked— whether they could search the backpack — was voluntary. Relying on defendant’s consent, the officer searched the backpack, discovered marijuana, gave the required Miranda warnings, and questioned defendant further. The officer asked defendant where he got the marijuana, how much marijuana there was, how much defendant had paid for it, and whether he was a middleman. Defendant responded to each of those questions, but, after he admitted that he was a middleman, defendant declined to provide further information and asked for an attorney. The trial court ruled that defendant’s responses to the post-Miranda questions also were voluntary and that they would be admitted.

In a stipulated facts trial, the court found defendant guilty of the crimes of delivery and possession of a controlled substance.² Defendant appealed and claimed as error the denial of his pretrial motions to suppress the marijuana and his post -Miranda statements.

The Court of Appeals observed that Miranda warnings are an essential part of the rights granted by Article I, section 12, and that the exploitation analysis articulated by this court in State v. Hall, 339 Or 7, 115 P3d 908 (2005), provided the appropriate framework to analyze the consequences of failing to give those warnings. State v. Vondehn, 219 Or App 492, 501-07, 184 P3d 567 (2008). In applying the Hall analysis, the court concluded that, because the police had obtained both the marijuana and the post -Miranda statements by exploiting defendant’s pr e-Miranda statements, they must be suppressed. Id. at 507-10.

More specifically, the court reasoned that the police had learned that the backpack belonged to defendant only through their pr e-Miranda questioning and that “[defendant's answers to [the officer’s] questions gave [the officer] the information that he needed to ask defendant for consent to search it.” Id. at 508. Therefore, the Court of Appeals identified a “but-for relationship between the unconstitutional questioning and defendant’s consent to the search.” Id. The court then decided that the state had not demonstrated that the evidence did not derive from the preceding illegality. The state had not argued to the trial court that the police inevitably would have discovered the evidence in the absence of the Miranda violation, and the police had not obtained the evidence independently of the Miranda violation. The violation provided the basis for the consent to search, and the consent led to the evidence. Moreover, there were no intervening circumstances that broke the causal chain between the Miranda violation and defendant’s consent. Because the Court of Appeals concluded that the police had obtained the marijuana by exploiting the Miranda violation, it held that the trial court had erred in denying defendant’s motion to suppress. Id.

As to the post -Miranda statements, the Court of Appeals reasoned that, “but for” the illegal questioning and search of the backpack, the officer would not have had the information on which he based his post -Miranda questions. Id. at 509-10. The record did not demonstrate that the police inevitably would have obtained the later statements, nor did the record demonstrate that the police had obtained them

independently from the earlier violations. Id. at 510. The Court of Appeals concluded that the trial court had erred in admitting both the marijuana and the post -Miranda statements and reversed defendant’s convictions. Id. at 509-10.

On review, the state acknowledges that this court has held that, when a person is in custody, the police must inform the person of his or her Miranda rights before subjecting the person to custodial interrogation, and the failure to give the required warnings necessitates the exclusion of all statements that the person makes in response to the interrogation. Applying that warning requirement and exclusionary rule to the facts of this case, the state also acknowledges that defendant was in custody and subjected to custodial interrogation when the police asked him whether he was the owner of the backpack and whether the backpack contained marijuana.³ Thus, the state concedes, the trial court properly excluded from evidence defendant’s responses to that interrogation — that he owned the backpack and that it contained marijuana.

The state contests, however, the conclusion of the Court of Appeals that the marijuana and the statements that defendant made after the police administered Miranda warnings must also be excluded. With respect to the marijuana, the state first contends that, although the text of Article I, section 12, precludes the admission of coerced testimony, it does not extend similar protection to uncompelled physical evidence. Alternatively, the state contends that the rule that requires the exclusion of statements made without the benefit of Miranda warnings is a prophylactic rule that reaches beyond the requirements of the constitution itself and that that rule should not be extended to preclude admission of physical evidence. The “mere failure,” as the state puts it, to give Miranda warnings does not constitute a constitutional violation and therefore call for a Hall exploitation analysis. With respect to the post -Miranda statements, the state contends that the sole test of their admissibility should be whether they were made voluntarily, an issue that the Court of Appeals did not reach.

This case, as framed by the state’s arguments, requires that we address the effect of the failure to give Miranda warnings in two distinct circumstances: (1) when the police commence custodial interrogation without giving required Miranda warnings and thereafter obtain incriminating physical evidence; and (2) when, after conducting an initial, unwarned custodial interrogation, the police give the required warnings and the defendant makes further incriminating statements. As to the first circumstance, we hold that when the police conduct custodial interrogation without obtaining a valid waiver of Article I, section 12, rights, they violate Article I, section 12, and the derivative physical evidence that they obtain must be suppressed. As to the second circumstance, we hold that a trial court must exclude defendant’s warned post -Miranda statements unless the state establishes that, considering the totality of the circumstances, when the police belatedly administer Miranda warnings, they effectively and accurately informed the defendant of his or her Article I, section 12, rights. We affirm the decision of the Court of Appeals in part, reverse it in part, and remand this case to the trial court for further proceedings.

I. INTERROGATION WITHOUT MIRANDA WARNINGS

The state begins its argument about the admissibility of physical evidence obtained without the benefit of Miranda warnings with the text of Article I, section 12: “[N]o person shall * * * be compelled in any criminal prosecution to testify against himself.” The state argues that that text does not prohibit the admission of physical evidence, even physical evidence that is a “fruit” of a defendant’s compelled testimony; it prohibits only compelling a person to “testify.” The state also contends that the text of Article I, section 12, does not prohibit the state from using a person’s compelled statements to investigate a crime and obtain evidence of the crime; it only creates a right not to have the statements themselves introduced “in any * * * criminal proceeding.” In support of those arguments, the state directs us to the comparatively broader wording of other state constitutions and to cases discussing the common-law privilege against self-incrimination as it was recognized at the time that Article I, section 12, was adopted.

This court first considered the text of Article I, section 12, and its history in State v. Cram, 176 Or 577, 160 P2d 283 (1945). The issue in that case was whether the testimony of a physician as to the alcohol content of a sample of the defendant’s blood, taken from him while under arrest and in custody, violated the defendant’s rights under Article I, section 12.⁴ Id. at 578-79. The court noted that, with two exceptions, all state constitutions contain provisions against “self-crimination.” Id. at 579. The wording of those provisions varies from prohibitions on “testifying” and “furnishing evidence” to “being a witness,” but the court observed that the difference in phrasing had not been considered important in construing their meaning. Id. at 579-80. The court also noted that the constitutional privilege against self-incrimination had generally been held to be declaratory of the common-law privilege and that that privilege was not limited to testimonial utterances, but extended to prevent the compelled production of documents or chattels. Id. at 581-82 (quoting 8 Wigmore on Evidence § 2263). The court also quoted Wigmore for the proposition that, when physical evidence is obtained by means other than compulsion of the defendant, it is admissible as long as admission does not depend on the defendant being called upon to make “any act or utterance of his own.” Id. at 582. Finally, the court applied those principles to conclude that the testimony of the physician about the blood sample did not violate the defendant’s Article I, section 12, rights. Id. at 593. The defendant had not been compelled to establish the authenticity, identity, or origin of the blood; those facts were proved by other witnesses. Id.

In State v. Soriano, 68 Or App 642, 646, 684 P2d 1220 (1984), aff'd and opinion adopted, 298 Or 392, 693 P2d (1984), this court again examined the history of the constitutional right against self-incrimination and again concluded that the differences in various state constitutional provisions were inconsequential: “The constitutional language varies, but courts generally treat the basic principle as the same in all the states.” 68 Or App at 646. The Soriano court also agreed that the word “testify” is not a limit on the protections that Article I, section 12, affords: “We see no reason to construe the Oregon Constitution to give protection from testifying but not from furnishing evidence.” Id. at 646-47 n 4.

In Soriano, this court held that the Oregon Constitution prohibits the state from requiring a witness to relinquish the Article I, section 12, right against self-incrimination unless it provides the witness with an alternative that affords the same protection as the constitution. Id. at 662. The issue in Soriano was whether Article I, section 12, permits the state to compel the testimony of a witness in exchange for “use” or “derivative use” immunity without also extending “transactional immunity.” Id. at 644. Use and derivative use immunity preclude the state from using compelled statements of a witness and their direct or indirect fruits, such as physical evidence discovered as a result of the statements, in a prosecution of that witness. Id. at 644 n 3. Transactional immunity precludes the state from prosecuting the witness for any offense to which the statements relate. Id. The court acknowledged that, when a witness provides compelled statements, those statements may influence a prosecution even if they are not offered in evidence or used to obtain derivative evidence. Id. at 663. For example, the statements may affect the discretionary decisions of a prosecutor to bring charges or to accept a plea bargain. Id. The court held that the state could not compel the statements of a witness without granting transactional immunity because, without protecting the witness from all evidentiary and nonevidentiary use of compelled statements, the state would not afford the witness the same protection that the constitution confers — the right to remain silent. Id. at 662.

Thus, this court has long interpreted Article I, section 12, to impose no distinction between compelled statements and physical evidence derived from such statements or between the use of compelled statements to obtain evidence and as testimony at trial. We reject the state’s argument that we should now impose those limitations on the reach of Article I, section 12. We turn to the state’s alternative argument that, even if Article I, section 12, requires that physical evidence derived from compelled statements be excluded from evidence, the same is not true for physical evidence derived from the “mere failure to provide Miranda warnings.”

For an understanding of that argument, the state directs us to the reasoning of the plurality in United States v. Patane, 542 US 630, 124 S Ct 2620, 159 L Ed 2d 667 (2004). Although the state acknowledges that we need not defer to the United States Supreme Court when we interpret the state constitutional right found in Article I, section 12, see Soriano, 68 Or App at 645-46 (giving reasons for independent interpretation of state constitutional right against self-incrimination),⁵ the state finds the logic of the lead opinion in Patane persuasive, and we would be remiss if we did not consider it. That is particularly so because it was the Supreme Court that first required, to effectuate the protections afforded by the Fifth Amendment to the United States Constitution, the warnings that this court later required to effectuate the protections afforded by Article I, section 12, known to both courts by the name of the Supreme Court’s decision, Miranda v. Arizona, 384 US 436, 86 S Ct 1602, 16 L Ed 2d 694 (1966).

In Patane, the police arrested a defendant for violating a restraining order. A detective attempted to inform the defendant of his Miranda rights, but the defendant interrupted and asserted that he knew his rights, and the officers did not attempt to complete the warning. The detective then asked the defendant about a gun. The defendant said that he did not want to discuss it because he did not want the detective to take the gun away from him. The detective persisted. The defendant told him that the gun was in a particular bedroom and gave the detective permission to retrieve the gun.

The government conceded that the defendant’s answers to the detective’s questions were inadmissible under Miranda, but argued that the failure to complete the Miranda warning did not require suppression of the gun itself. The plurality, consisting of Justice Thomas, Chief Justice Rehnquist, and Justice Scalia, took the position that, although the admission of “actually coerced statements” and physical evidence derived from such statements violates the Fifth Amendment, “a mere failure to give Miranda warnings does not, by itself, violate a suspect’s constitutional rights * * *.”⁶ 542 US at 641-44. Thus, the plurality reasoned, there was no violation of the Fifth Amendment to deter and no reason to apply the “fruit of the poisonous tree” doctrine of Wong Sun v. United States, 371 US 471, 83 S Ct 407, 9 L Ed 2d 441 (1963). Patane, 542 US at 641-42. According to the plurality, the Miranda rule is a prophylactic rule that sweeps beyond the actual protections of the Self-Incrimination Clause. Id. at 639. A court must presume that statements made without its protections are compelled, but there is no need extend that rule to exclude physical evidence obtained as a result of unwarned statements. Id. at 639-43.

Justices Kennedy and O’Connor concurred in the result. They reasoned that Miranda was based in large part on an effort to accommodate concerns about compelled testimony and other objectives of the criminal justice system. Given the important probative value of reliable physical evidence, they could not justify exclusion based on “a deterrence rationale sensitive to both law enforcement interests and a suspect’s rights during an in-custody interrogation.” Id. at 644-45.

To address the state’s argument that, in interpreting Article I, section 12, we should adopt the reasoning of the plurality opinion in Patañe, we must examine the nature and purpose of the Miranda warnings that this court requires and the reasons that statements made without the benefit of those warnings are excluded from evidence.

In State v. Magee, 304 Or 261, 266, 744 P2d 250 (1987), this court stated that Article I, section 12, “furnishes an independent basis” for requiring that police administer Miranda warnings to suspects who are in custody. The prior year, in State v. Smith, 301 Or 681, 725 P2d 894 (1986), the court had discussed the basis for the United States Supreme Court’s Miranda decision and whether similar warnings were required by the Oregon Constitution. Smith was a divided opinion that resulted in the affirmance of defendant’s conviction on the basis that the defendant was not in custody when he made the unwarned statements. When the court then decided Magee, it did not reprise the arguments that it had considered in Smith or state the rationale for its historic decision.

Since Magee, this court consistently has held that the Oregon Constitution requires suppression of statements made without the benefit of Miranda warnings. See, e.g., State v. Roble-Baker, 340 Or 631, 643-44, 136 P2d 22 (2006) (suppressing unwarned statements made during custodial interrogation); State v. Smith, 310 Or 1, 7, 791 P3d 836 (1990) (so stating). The full extent of the court’s discussion of the rationale for that rule has been to state that, when a suspect is subjected to custodial interrogation, warnings are necessary “ ‘because of the inherent level of coercion that exists in such interrogations.’ ” State v. Scott, 343 Or 195, 200, 166 P3d 528 (2007) (quoting State v. Joslin, 332 Or 373, 380, 29 P3d 1112 (2001)); see also State v. Meade, 327 Or 335, 339, 963 P2d 656 (1998). Further, in discussing the requirement that police advise suspects subjected to custodial interrogation of the right to assistance of counsel, the court has stated that that particular warning is required because “a lawyer’s presence at a custodial interrogation is one way to ensure the right to be free from compelled self-incrimination.” Meade, 327 Or at 339. A suspect must be informed if an identified lawyer has been retained or appointed and is seeking to consult with the suspect. Without that information, the suspect cannot “be said knowingly to have waived his or her right against compelled self-incrimination under Article I, section 12.” Joslin, 332 Or at 383 (citing State v. Haynes, 288 Or 59, 70, 602 P2d 272 (1979)).

In State v. Simonsen, 319 Or 510, 512, 878 P2d 409 (1994), the court followed Haynes and held that the police had violated the defendant’s Article I, section 12, rights when they interrogated him “without informing him of the fact that he had a court-appointed lawyer or the fact that the lawyer had asked to consult with defendant before further interrogation took place.” The court then stated that its rationale for suppressing the statements that the defendant had made during that interrogation was

“ ‘to preserve * * * rights to the same extent as if the government’s officers had stayed within the law. * * * In the context of a criminal prosecution, the focus then is on protecting the individual’s rights vis-a-vis the government * * *.

“ ‘This focus on individual protection under the exclusionary rule, a rule that operates to vindicate a constitutional right in the courts, supports the constitutional rule * * *. [T]he constitutionally significant fact is that the Oregon government seeks to use the evidence in an Oregon criminal prosecution. Where that is true, the Oregon constitutional protections apply.’ ”

Id. at 518-19 (quoting State v. Davis, 313 Or 246, 253-54, 834 P2d 1008 (1992)) (alterations in original; internal citations omitted). In the Davis case that the court in Simonsen quoted, the court held that evidence obtained in violation of a suspect’s rights to be free of unreasonable search and seizure under Article I, section 9, of the Oregon Constitution must be suppressed, and explained that, to give effect to constitutional rights, the government cannot “obtain a criminal conviction through the use of evidence obtained in violation of [those rights].” Davis, 313 Or at 253.

Davis in turn relied on two earlier cases — State v. Davis, 295 Or 227, 666 P2d 802 (1983), and State v. Isom, 306 Or 587, 761 P2d 524 (1988). In the former case, the court noted that, although every rule of law is intended to deter contrary conduct and is successful when it does so, deterrence does not constitute a constitutional basis for the exclusion of evidence. Davis, 295 Or at 234-35. Instead, the court stated, “ ‘[i]n demanding a trial without such evidence, the defendant invokes rights personal to himself.’ ” Id. at 235 (quoting State v. McMurphy, 291 Or 782, 785, 635 P2d 372 (1981)). In the latter case, the court ordered suppression of statements made in violation of Article I, section 12, because “the state may not prove, over objection, any crime with unconstitutionally obtained evidence.” Isom, 306 Or at 595.

Synthesizing the decisions that we have reviewed, and applying our own constitutional analysis, we now set out the basis for the requirement that police inform persons in custody and subjected to custodial interrogation of the rights afforded by Article I, section 12, and for excluding statements made without the benefit of those warnings.

Article I, section 12, affords a constitutional right to remain silent. That right is, however, subject to waiver. Because a custodial interrogation is inherently compelling, and to ensure the validity of a waiver of the right against self-incrimination, Article I, section 12, requires that the police inform a person subjected to custodial interrogation that he or she has a right to remain silent and to consult with counsel and that any statements that the person makes may be used against the person in a criminal prosecution. Article I, section 12, requires those Miranda warnings to ensure that a person’s waiver is knowing as well as voluntary. If the police conduct a custodial interrogation without first obtaining a knowing and voluntary waiver of the suspect’s rights, then they violate the suspect’s Article I, section 12, rights. To give effect to those constitutional rights, the state is precluded from using, in a criminal prosecution, statements made in response to the interrogation.

With that understanding, we return to the state’s alternative argument that we should adopt a rule that permits the admission of physical evidence that the police obtain without the benefit of Miranda warnings. In making that argument, the state concedes that, if the constitutional violation at issue results from “actual coercion,” then all evidence, including physical evidence, obtained as a result of the violation must be excluded from evidence. See Soriano, 68 Or App at 662-65 (“Prohibition of the use of the fruits of illegal police activity in court is necessary to vindicate the violated rights.”); Patane, 542 US at 644 (acknowledging that Fifth Amendment requires that physical evidence derived from actually coerced statements must be suppressed); United States v. Hubbell, 530 US 27, 37-38, 120 S Ct 2037, 147 L Ed 2d 24 (2000) (holding that Fifth Amendment requires that evidence derived from coerced statements must be suppressed). So, for example, if the police were to engage in “actual coercion” to compel the answer to the question “where is the gun located?,” both the answer, “under the bed,” and the gun found in that location would be suppressed. See State v. Miller, 300 Or 203, 709 P2d 225 (1985) (physical evidence derived from an unconstitutional custodial interrogation must be suppressed unless admissible on some independent ground).⁷

The state contends, however, that we should reach a different conclusion when the violation of Article I, section 12, is a “mere failure to provide Miranda warnings” relying on the reasons persuasive to the plurality in Patane: that such a failure does not violate a suspect’s constitutional rights and that, given the important value of reliable physical evidence, the Miranda rule should not be extended to exclude it. It is immediately obvious that the premise of the state’s argument does not hold here. It is the Oregon Constitution that requires Miranda warnings and it is the Oregon Constitution that is violated when those warnings are not given. When the police violate Article I, section 12, whether that violation consists of “actual coercion” or the failure to give the warnings necessary to a knowing and voluntary waiver, the state is precluded from using evidence derived from that violation to obtain a criminal conviction. It follows ineluctably that, when the police violate Article I, section 12, by failing to give required Miranda warnings, the state is precluded from using physical evidence that is derived from that constitutional violation to prosecute a defendant.⁸

It is now incumbent on us to apply the principles that we have enunciated to the facts of this case. As noted, defendant was in custody, in the back seat of a patrol car and handcuffed, when the police subjected him to custodial interrogation. Defendant had the right to remain silent and to advice of counsel, but the police conducted their custodial interrogation without obtaining a valid waiver of those rights. When they did so, the police violated Article I, section 12. That constitutional violation requires suppression of both the answers that defendant gave in response to, and the marijuana that the police identified and seized as a result of, that interrogation.

In response to the police officer’s first unwarned question, the officer learned that the backpack belonged to defendant. In response to the officer’s second unwarned question, the officer learned that there was marijuana in the backpack. With that relevant information, the officer then immediately requested consent to search the backpack and seized the marijuana. In this court, the state makes no argument that the request for consent to search or the seizure of the marijuana derived from some source other than defendant’s answers to those unwarned questions, nor does the state argue that, even without defendant’s responses, the police inevitably would have obtained the marijuana. Thus, in this case, we conclude that the marijuana derived from the violation of defendant’s Article I, section 12, rights, and the trial court erred in failing to exclude it from evidence. We affirm the decision of the Court of Appeals in that regard.

II. BELATED MIRANDA WARNINGS

The second question in this case is the consequence that a violation of Article I, section 12, should have when the police belatedly administer required Miranda warnings. The state argues that, after the police give the required warnings, a defendant’s statements in response to further interrogation are admissible as long as they are voluntary. Defendant argues that the constitutional violation that occurs when the police fail to give the warnings when initially required necessitates exclusion of all statements made as a result of the illegality and that that category includes statements made after belated Miranda warnings, unless the state demonstrates, under Hall, that, at the time the post -Miranda statements were made, the taint of the illegality had been attenuated.

Two Supreme Court cases illustrate the range of factual circumstances that present that issue. In Oregon v. Elstad, 470 US 298, 105 S Ct 1285, 84 L Ed 2d 222 (1985), the police contacted a young suspect at his home to arrest fiim for burglary. Before the arrest, one officer spoke with the mother, while the other officer mentioned to the suspect that he felt that the suspect had been involved in the burglary. The suspect admitted that he had. The officers then placed the suspect under arrest and took him to the police station where they administered Miranda warnings and, when the suspect waived his rights, interrogated him. The defendant was charged with burglary and sought to suppress his post-Miranda statements as the illegal “fruit” of the Miranda violation.

The Court rejected defendant’s argument. It ruled that the admissibility of any such statement turns “solely on whether it is knowingly and voluntarily made.” Id. at 309. In considering whether the defendant’s post -Miranda statements met those standards, the Court discounted whatever psychological impact that his unwarned statements could have had on his later waiver of his Fifth Amendment rights and found the warned statements admissible because the connection between the defendant’s prior admission and his ultimate decision to cooperate was “speculative and attenuated at best.” Id. at 312-14.

In 2004, the Supreme Court again explored the issue of belated Miranda warnings in Missouri v. Seibert, 542 US 600, 124 S Ct 2601, 159 L Ed 2d 643 (2004). In that case, the officer had made a “conscious decision” to withhold Miranda warnings. The officer conducted the interrogation in the station house, and the interrogation was “systematic, exhaustive, and managed with psychological skill.” Id. at 616. The defendant made incriminating statements in response to the interrogation. Then, after a pause of only 15 to 20 minutes, the same officer, in the same place, recited the Miranda warnings but did not advise the defendant that her prior statements could not be used. In fact, the officer referenced the defendant’s earlier statements and used them to convince her to repeat her earlier confesson. Justice Souter wrote for four members of the court and concluded that all of the defendant’s post -Miranda statements must be suppressed, not as the “fruit” of a prior unlawful act, but because the state had not established that the belated Miranda warnings were “effective.” Id. at 604 (plurality opinion).

The plurality explained its reasoning as follows:

“Just as ‘no talismanic incantation [is] required to satisfy [Miranda’s] strictures,’ California v. Prysock, 453 US 355, 359[, 101 S Ct 2806, 69 L Ed 2d 696] (1981) (per curiam), it would be absurd to think that mere recitation of the litany suffices to satisfy Miranda in every conceivable circumstance. ‘The inquiry is simply whether the warnings reasonably “conve[y] to [a suspect] his rights as required by Miranda.” ’ Duckworth v. Eagan, 492 US 195, 203[, 109 S Ct 2875, 106 L Ed 2d 166] (1989) (quoting Prysock, supra, at 361). The threshold issue when interrogators question first and warn later is thus whether it would be reasonable to find that in these circumstances the warnings could function ‘effectively’ as Miranda requires. Could the warnings effectively advise the suspect that he had a real choice about giving an admissible statement at that juncture? Could they reasonably convey that he could choose to stop talking even if he had talked earlier? For unless the warnings could place a suspect who has just been interrogated in a position to make such an informed choice, there is no practical justification for accepting the formal warnings as compliance with Miranda, or for treating the second stage of interrogation as distinct from the first, unwarned and inadmissible segment.”

Id. at 611-12 (alterations in original; footnote omitted).

The Seibert plurality identified, in the contrast between Elstad and Seibert, a “series of relevant facts that bear on whether Miranda warnings delivered midstream could be effective enough to accomplish their object[.]” Seibert, 542 US at 615. Those facts include: (1) the completeness and detail of the questions and answers in the first round of interrogation, (2) the overlapping content of the two statements, (3) the timing and setting of the first and the second rounds of interrogation, (4) the continuity of police personnel, (5) the degree to which the interrogator’s questions treated the second round as continuous with the first, and (6) whether the police cautioned that the earlier unwarned statement could not be used in any subsequent prosecution. Id. at 615-16.

Justice Kennedy concurred in the judgment, but his approach to the issue was different. Justice Kennedy opined that the question-first technique “creates too high a risk that postwarning statements will be obtained when a suspect was deprived of knowledge essential to his ability to understand the nature of his rights and the consequences of abandoning them.” Id. at 621 (internal quotation marks and citation omitted). He concluded that the post-warning statements should be excluded only when that technique is used deliberately. Id. at 622. When the police purposely engage in a two-step interrogation, Justice Kennedy opined, “postwarning statements that are related to the substance of prewarning statements must be excluded absent specific, curative steps.” Id. at 621. Justice Kennedy explained that curative measures “should be designed to ensure that a reasonable person in the suspect’s situation would understand the import and effect of the Miranda warning and of the Miranda waiver[ ]” and could include, for example, a substantial break in time and circumstances or an additional warning that the prior unwarned statements would likely be inadmissible. Id. at 622.

Justice O’Connor, the author of Elstad, penned the dissent for the four remaining members of the court. She agreed with the plurality that the test of admissibility should be an objective one and not determined, as Justice Kennedy urged, by whether the police acted intentionally:

“A suspect who experienced exactly the same interrogation as Seibert, save for a difference in the undivulged, subjective intent of the interrogating officer when he failed to give Miranda warnings, would not experience the interrogation any differently. Whether intentional or inadvertent, the state of mind of the police is irrelevant to the question of the intelligence and voluntariness of respondent’s election to abandon his rights.”

542 US at 624-25 (internal quotation marks and brackets omitted). For Justice O’Connor, the correct issue was whether, under all the circumstances, the post-warning statements were voluntary. Id. at 628.

Because we analyze the issue under the Oregon, and not the United States, Constitution, we are not bound by the Court’s decisions in either Elstad or Seibert. Nevertheless, we find the reasoning in those cases helpful because they focus, as we must, on the source and purpose for the Miranda requirement and the exclusion of evidence obtained when Miranda warnings are not given as required.

As our prior discussion indicates, the Oregon Constitution does not require the exclusion of evidence obtained in the absence of Miranda warnings to deter illegal police conduct. The Oregon Constitution requires Miranda warnings to ensure that a waiver of the rights conferred by Article I, section 12, is knowing as well as voluntary. When the police fail to give the required warnings, a suspect’s responses to their unwarned questions must be excluded from evidence. When the police then correct course and give the required warnings, the relevant inquiry must be whether the belated warnings are effective and accomplish the purpose for which they are intended. The fact that the police initially violate a defendant’s constitutional rights by failing to give the warnings necessary to a valid waiver does not preclude a defendant from later validly waiving those rights. If the state establishes that the police accurately and effectively, although belatedly, gave the suspect the information necessary to a valid waiver of the right against self-incrimination, then, under the Oregon Constitution, a suspect’s subsequent voluntary statements will be admissible.⁹ In arriving at that conclusion, we adopt the reasoning and the analysis of the Siebert plurality as our own.

When the police give Miranda warnings at the time that they are first required, the constitution does not demand that the state establish that the warnings were effective. The state need only establish that the police recited the warnings completely and coherently.¹⁰ The problem that Seibert demonstrates, however, is that when the police question first and warn later, their exhibition and exercise of authority and violation of the defendant’s constitutional rights may communicate to a defendant, as the Court believed they did in that case, that, before the defendant will be released, he or she must answer the questions asked. In that circumstance, the police not only fail to provide the defendant with the information necessary to a valid waiver — that the defendant has a right to remain silent and to confer with an attorney — the police also convey a contrary message. In that situation, when the police later administer Miranda warnings, we cannot assume that the mere recitation of Miranda warnings is sufficient to serve the intended informative function.

That being said, we note that Seibert is at one end of the range of the factual circumstances that present the issue that we address. Elstad is at the other. Not every instance in which the police question first and warn later communicates a mixed message. Whether and to what extent police officers who fail to administer Miranda warnings before beginning custodial interrogation obfuscate or contradict the information that Miranda warnings are intended to convey and whether and to what extent those officers later correct that misinformation are issues that trial courts must confront and determine. In doing so, courts should consider all relevant circumstances, including those facts to which the plurality in Seibert pointed — the completeness and detail of the questions and answers in the first round of interrogation, the overlapping content of the statements given by the suspect, the timing and setting of the first and the second interrogation sessions, the continuity of police personnel, the degree to which the interrogator’s questions treated the second round of interrogation as continuous with the first, and whether the police cautioned that the earlier unwarned statement could not be used in any subsequent prosecution.¹¹

In this case, the Court of Appeals used many, if not all, of those facts in its “fruit of the poisonous tree,” or “exploitation,” analysis. The difference between that court’s analysis and our own is not in the facts considered but in the stick by which we require that they be measured. In our analysis, the test of the efficacy of the belated warnings is an objective one. A court considers the factual circumstances to determine the accuracy and effectiveness of the information that the police convey; a court does not use those circumstances to attempt to determine the psychological effect that the particular police course of conduct had on the particular defendant or whether the initial failure to warn caused the particular defendant to make the post -Miranda statements.

Although neither party in this case advocates for the view articulated by Justice Kennedy in Seibert — that the admissibility of statements that a defendant makes after Miranda warnings are belatedly given should be determined by whether the police intentionally engaged in a two-step interrogation process — we think it helpful to state explicitly that we reject that approach. Our focus is not on the subjective intent of the police but on the objective message that the police actually convey by the techniques that they use and the warnings that they give. That does not mean, however, that the deliberateness with which the police act is entirely irrelevant. In Isom, for instance, the defendant established that the police had purposely disregarded his request for counsel, and this court did not hesitate to hold that the statements that he made thereafter were obtained in violation of Article I, section 12. 306 Or at 595. If the police purposely obscure the legal and practical significance of a belated Miranda admonition, as they did in Seibert, it will not be difficult for a court to determine under the objective test that we describe today that the police did not accurately and effectively deliver the information necessary to a valid waiver of Article I, section 12, rights.

Finally, it remains for us to decide whether the trial court in this case erred in admitting defendant’s post-Miranda statements. To do so we must review the facts in more detail than initially stated.

This case arose from a routine roadside stop of a car to investigate a traffic violation and possible driving under the influence of intoxicants. Defendant was a passenger in the car, which a friend was driving. Officer Stoneberg made the stop; a second officer, Officer Espelien, arrived shortly afterward. Stoneberg approached the driver’s side of the car and, as he did, smelled a strong odor of fresh marijuana. Espelien approached the passenger’s side and initially waited near the rear corner of the car while Stoneberg talked to the driver. Espelien, likewise, smelled a very strong odor of fresh marijuana. Both officers thought that the smell was strongest towards the trunk area of the car. Espelien approached defendant (the passenger) to ask him for identification. Defendant initially lied about who he was but then gave, Espelien the correct information. Espelien discovered an outstanding warrant for defendant. Stoneberg therefore arrested defendant, handcuffed him, and had him sit in the back of the patrol car while he completed his investigation. Stoneberg did not give Miranda warnings to defendant at that point because he ordinarily does not conduct any kind of investigation or ask questions of a person arrested on an outstanding warrant.

Stoneberg continued his investigation of the driver. He asked her if there were any illegal drugs or weapons in the car. She said no. Stoneberg then asked her for, and the driver gave, consent to search the car. In the course of the search, Stoneberg opened the car trunk, at which point the smell of fresh marijuana became even stronger. The only thing in the trunk was a backpack. According to Stoneberg, the smell from the backpack “is what really hit [him]” and, upon lifting it, he could tell that the backpack was not empty. At that point, given the smell and the weight of the backpack, Stoneberg believed that there was a substantial quantity of marijuana in it. Stoneberg asked the driver if the backpack was hers and if he could open it, and she told Stoneberg that it did not belong to her. He asked her who owned it, and she said she did not know.

Stoneberg then walked to the patrol car with the backpack, opened the back door, and asked defendant if the backpack belonged to him. Defendant said yes. Stoneberg asked defendant if there was marijuana inside, and defendant said yes. Stoneberg asked if he could search the backpack, and defendant told Stoneberg he could. Stoneberg then searched the backpack while standing by the patrol car, in defendant’s presence, and found two folded-down grocery bags, each containing fresh marijuana.

After searching the backpack, Stoneberg walked over to Espelien and consulted with him. After about five minutes, Stoneberg went back to the patrol car to ask defendant more questions about the marijuana. At that point, he gave defendant Miranda warnings. He asked defendant if he understood his rights, and defendant said he did. Defendant waived his rights and agreed to answer Stoneberg’s questions. Stoneberg asked defendant several questions about the marijuana. In particular, Stoneberg asked defendant where he got the marijuana; defendant explained that he had gotten it in Tualatin the day before. Stoneberg asked how much marijuana was in each bag; defendant told him that each bag contained about a quarter of a pound of marijuana. Stoneberg asked how much the marijuana cost; defendant said that he had paid a total of $2,200 for the marijuana. Stoneberg asked defendant if he was a middleman. Defendant answered yes. Eventually, Stoneberg told defendant that he would like to know more about his middleman role. In response, defendant told Stoneberg that he wanted a lawyer. Neither Stoneberg nor Espelien asked defendant any further questions. The post -Miranda questions took place intermittently over a 15- to 20-minute time period.

Both officers described their contact with defendant as conversational in tone. Espelien further described defendant, once defendant gave his correct identification to Espelien, as seeming apologetic and even ashamed of the circumstances. Defendant remained “very cooperative” and willingly answered questions until, when Stoneberg asked for more information about his role as a middleman, he invoked his right to counsel.

We now consider all those facts to determine whether the Miranda warnings, when given, accurately and effectively conveyed the information necessary to a knowing and voluntary waiver of the right against self-incrimination. We first observe that there was a marked difference in the questioning before and after Stoneberg administered the Miranda warnings. The unwarned questions were routine in nature and consumed less than a minute of time.¹² The second warned questions were significantly more detailed and probing. This was not a situation, like that in Seibert, in which the police conducted extensive questioning and elicited significant detailed facts in the first interrogation session and then repeated that questioning post -Miranda.

Second, although Stoneberg posed the second set of questions shortly after the first set, there was a break in the questioning. Given that the first set of questions consumed less than a minute, the five-minute break between questions, followed by the Miranda warning, was an objective indication that the situation had changed and was governed by new rules. Again, this was not a circumstance, like that in Seibert, in which both interrogation sessions were protracted and conducted at the police station. In such a circumstance, a short break between interrogation sessions would not be of the same import.

Third, although Stoneberg did not caution defendant that his earlier unwarned statements could not be used in any subsequent proceeding, neither did he point out to defendant, as did the officer in Seibert, that defendant had already made incriminating disclosures that provided all of the information that Stoneberg needed. And in fact, at least some of the information that Stoneberg had obtained — that the backpack contained marijuana — was obviously discernable from another source — the odor of the backpack itself. When an officer does caution a defendant that the unwarned statements that the defendant made may not be admissible, that caution may militate (indeed, often will) in favor of finding that the officer’s belated Miranda warnings were effective, but such a caution is not necessary to that result.

Fourth, although defendant was under arrest and handcuffed when Stoneberg questioned him and was thus in inherently compelling circumstances, he was not subjected to additional coercion. The police conducted their questioning in a conversational tone and it was of short duration.

The facts that we have outlined lead us to conclude that the belated Miranda warnings that Stoneberg gave accurately and effectively communicated that defendant had, from that time forth, a right to remain silent. We therefore agree with the trial court, although for different reasons, that defendant’s post -Miranda statements should have been admitted.

The decision of the Court of Appeals is affirmed in part and reversed in part. The judgment of the circuit court is reversed, and the case is remanded to the circuit court for further proceedings.

1 Article I, section 12, provides, in part, that “no person shall * * * be compelled in any criminal prosecution to testify against himself.”

2 Defendant was convicted under former ORS 475.992 (2003), renumbered as ORS 475.840 (2005).

3 As a result of the state’s acknowledgement, we do not consider whether the two questions that the police asked in this case constituted “interrogation.” See State v. Scott, 343 Or 195, 203, 166 P3d 528 (2007) (defining “interrogation” as “conduct that the police should know [is] reasonably likely to elicit an incriminating response”). When we refer to “custodial interrogation” elsewhere in this opinion, we mean interrogation that meets the Scott definition.

4 The defendant had not contested the act of obtaining the blood as an unconstitutional search or seizure. This case does not present that issue or the issue discussed by the dissent in State v. Fish, 321 Or 48, 64-71, 893 P2d 1023 (1995) (Gillette, J., concurring in part and dissenting in part) — whether physical evidence concerning a person’s identity, appearance, or physical condition implicates Article I, section 12.

5 In Soriano, the court noted that the common-law rule that no person is bound to accuse himself first received constitutional status in Virginia in 1776 and that it was not added to the United States Constitution until 1791. 68 Or App at 646. The court explained that “[w]hen the United States Supreme Court first acted it simply adopted one line of state decisions and rejected another.” Id.

6 The plurality acknowledged that the Court had held in Dickerson v. United States, 530 US 428, 444, 120 S Ct 2336, 147 L Ed 2d 405 (2000), that Miranda announced a constitutional rule that could not be altered by Congress, Patane, 542 US at 636, but cited Chavez v. Martinez, 538 US 760, 772-89, 123 S Ct 1994, 155 L Ed 2d 984 (2003) (plurality opinion) for the proposition that a failure to give those constitutionally required warnings does not violate the constitution. Patane, 542 US at 641. One of the problems with the plurality’s approach is that Chavez was a case brought under 42 USC section 1983. The plaintiff in that case had not been criminally prosecuted and, therefore, no evidence, direct or derivative, had ever been used against him. However, we need not pick and choose here: a plurality opinion, particularly a plurality opinion addressing a federal constitutional issue that is not before us, is not controlling, and it therefore stands only for whatever persuasive power its logic carries.

7 In Miller, the police found physical evidence when the defendant handed them the keys to a hotel room. Before the defendant did so, he had told the police that he did not want to talk to them without a lawyer. The police had disregarded that statement and continued their questioning. It was then that the defendant handed over the keys. The court found that the defendant had asserted, and not waived, his right to counsel under the Fifth Amendment. Id. at 224.

8 We note that the result we reach is consistent with that of other state courts that have decided the same issue under their state constitutions. See, e.g., State v. Peterson, 181 Vt 436, 923 A2d 585 (2007) (physical evidence obtained in violation of Miranda rights must be excluded at trial as “fruit of poisonous tree”); Commonwealth v. Martin, 444 Mass 213, 827 NE2d 198 (2005) (physical evidence derived from unwarned statements presumptively excludable from evidence as “fruit” of improper failure to provide such warnings); State v. Knapp, 285 Wis 2d 86, 700 NW2d 899 (2005) (physical evidence obtained as direct result of intentional Miranda violation excluded as “fruit of poisonous tree”).

9 Voluntariness of course is always a requirement for admission of a defendant’s incriminating statements. Even warned statements may be inadmissible if they are not otherwise voluntary. See State v. Montez, 309 Or 564, 572, 789 P2d 1352 (1990) (assessing voluntariness of statements made after warnings were administered and defendant made “equivocal” remarks regarding request to consult with attorney).

10 We acknowledge, of course, that a defendant is entitled to demonstrate (whether by defendant’s own testimony or otherwise) that the defendant’s waiver was not knowing. So, for example, a defendant may demonstrate that he or she did not understand the warnings due to cognitive or linguistic limitations.

11 We note that other state courts applying their state constitutions also have used multi-factor tests to determine the admissibility of post-Miranda statements. See State v. O’Neill, 193 NJ 148, 180-81, 936 A2d 438 (2007); People v. Paulman, 5 NY3d 122, 130-31, 833 NE2d 239 (2005); and State v. Northern, 262 SW3d 741, 763-64 (Tenn 2008).

12 Given the brevity of the initial three questions that Stoneberg asked, and the one word answers that defendant gave in reply, the total amount of time involved for all three questions would have been a matter of perhaps only 30 seconds or less.