State v. Delong

• Court: Oregon Supreme Court
• Decided: June 18, 2015
• Docket: S062176
• Status: Published

Opinion

No. 23 June 18, 2015 365

IN THE SUPREME COURT OF THE STATE OF OREGON

STATE OF OREGON, Petitioner on Review, v. WILLIAM RICK DELONG, Respondent on Review. (CC 09CR1050FE; CA A146907; SC S062176)

En Banc On review from the Court of Appeals.* Argued and submitted October 8, 2014, at Bend Senior High School, Bend, Oregon. Michael A. Casper, Deputy Solicitor General, Salem, argued the cause and filed the brief for petitioner on review. With him on the brief were Ellen F. Rosenblum, Attorney General, and Anna M. Joyce, Solicitor General. Daniel C. Bennett, Deputy Public Defender, Salem, argued the cause and filed the brief for respondent on review. With him on the brief was Peter Gartlan, Chief Defender, Office of Public Defense Services. Shauna M. Curphy, Portland, filed the brief for amici curiae Oregon Justice Resource Center, Albina Ministerial Alliance Coalition for Justice and Police Reform, The Portland Chapter of the National Lawyers Guild, Inc., and the American Civil Liberties Union Foundation of Oregon, Inc. With her on the brief were Sara F. Werboff and Jordan R. Silk. KISTLER, J. The decision of the Court of Appeals is reversed, and the case is remanded to the Court of Appeals for further pro- ceedings consistent with this decision. ______________ * Appeal from Douglas County Circuit Court, Joan G. Seitz, Judge. 260 Or App 718, 320 P3d 653 (2014). 366 State v. Delong

Brewer, J., concurred and filed an opinion. Walters, J., dissented and filed an opinion in which Baldwin, J., joined. Baldwin, J., dissented and filed an opinion in which Walters, J., joined.

Case Summary: During a traffic stop, a deputy sheriff placed handcuffed defendant, placed him in a police car, and asked him if “there was anything that we should be concerned about” in his car without first giving defendant a Miranda warning. Defendant responded, “ ‘No,’ and that if [the deputies] wanted to search the vehicle [they] could.” The search resulted in the discovery of methamphet- amine and drug paraphernalia and, after the deputies read defendant a Miranda warning, defendant gave incriminating statements. Defendant moved to sup- press the physical evidence during the search and the statements that he later made on the ground that that evidence was the product of the earlier Miranda violation. The trial court denied the motion. The Court of Appeals reversed. Held: (1) voluntary consent to search can attenuate the taint of failing to give a Miranda warning in violation of Article I, section 12, of the Oregon Constitution; (2) in determining whether an invitation to search breaks the causal connection between a Miranda violation and subsequently obtained physical evidence, the nature of the illegality, the character of the consent, and the causal relationship between the two will bear on whether the physical evidence is attenuated from the violation. The decision of the Court of Appeals is reversed, and the case is remanded to the Court of Appeals for further proceedings consistent with this decision. Cite as 357 Or 365 (2015) 367

KISTLER, J. During a traffic stop, a deputy sheriff placed defen- dant in custody and then asked him, without first advis- ing him of his Miranda rights, “if there was anything we should be concerned about” in his car. Defendant “told [the deputy] ‘no,’ and that if we wanted to search the vehicle, we could.” On appeal, the state conceded that the deputy vio- lated Article I, section 12, of the Oregon Constitution when he asked defendant that question without first advising him of his Miranda rights. The state argued, however, that the physical evidence that the deputies later found in defendant’s car did not “derive from” the Miranda violation. The Court of Appeals disagreed. State v. Delong, 260 Or App 718, 320 P3d 653 (2014). Relying on State v. Vondehn, 348 Or 462, 236 P3d 691 (2010), the Court of Appeals reasoned that both defendant’s offer and the resulting evidence derived from the violation. Having allowed the state’s petition for review, we reverse the Court of Appeals decision and remand this case to the Court of Appeals. Sergeant Robeson worked for the Douglas County Sheriff’s office.1 One evening, while Robeson was on patrol, defendant’s car pulled out in front of Robeson. Apparently noticing Robeson’s marked patrol car behind him, defendant “immediately pulled off” into a store parking lot. Robeson continued driving, went around the corner, and pulled over to the side of the road to see if defendant would resume driving once Robeson passed by. “[A] few seconds later,” defendant drove past Robeson. In doing so, defendant con- firmed Robeson’s suspicion that he had been trying to avoid Robeson, and he also gave Robeson the opportunity to see that he was not wearing a seat belt. Robeson stopped defendant for that traffic violation. See ORS 811.210 (requiring that drivers wear seat belts). He approached defendant’s car and asked him for his driver’s license, registration, and proof of insurance. Defendant gave Robeson his name but could not produce a driver’s license or other picture identification. Driving without a license is a traffic offense; however, it is a defense to that charge 1 We take the facts from the hearing on defendant’s suppression motion and state them consistently with the trial court’s ruling. 368 State v. Delong

that the driver in fact had a valid license. See ORS 807.570. Robeson sought to determine defendant’s identity so that he could see if defendant in fact had a valid license. Robeson also wanted to identify defendant to see if there were a rea- son why defendant apparently had sought to avoid him; spe- cifically, Robeson wanted to see if there were an outstanding warrant for defendant’s arrest. There was a passenger in defendant’s car, and Robeson removed defendant from his car, frisked and hand- cuffed him, and put him in the backseat of the patrol car before asking him some background questions to verify his identity.2 At that point, Robeson had not advised defendant of his Miranda rights. After asking some questions regard- ing defendant’s identity, Robeson asked defendant “if there was anything we should be concerned about” in his car.3 In response to that question, defendant “told [Robeson] ‘no,’ and that if we wanted to search the vehicle, we could.”4 Robeson relayed that response to another deputy, who had arrived at the traffic stop. The second deputy searched defendant’s car and found what appeared to be marijuana residue in an ashtray underneath the driver’s seat. He then opened a canvas fanny pack that was inside the car, where he found methamphetamine and drug paraphernalia. At that point, the second deputy advised defendant of his Miranda rights. Defendant stated that he understood his rights and then

2 Robeson testified at the suppression hearing that he separated defendant from the passenger so that she could not conform her answers to defendant’s. At the suppression hearing, defendant did not challenge Robeson’s decision to hand- cuff him. Perhaps for that reason, neither the state nor defendant asked Robeson about the circumstances that led him to do so. 3 Justice Baldwin’s dissent states that “Robeson questioned [defendant] about illegal activity unrelated to the stop without first warning him that he had a right to remain silent.” 357 Or at 398-99 (Baldwin, J., dissenting). Justice Walters’ dissent contains a similar statement. To the extent that the dissents suggest that Robeson asked defendant something other than (1) questions about defendant’s identity and (2) “if there was anything [the deputies] should be con- cerned about” in defendant’s car, that suggestion does not appear consistent with the record. 4 Defendant, for his part, denied that he “volunteer[ed]” that Robeson could search his car. However, he agreed that he consented to a search of his car. Defendant testified that Robeson asked him,”[I]f he could search—if I minded if he searched the vehicle.” Defendant testified that he “told [Robeson] I don’t care but I got a whole bunch of stuff in the trunk of the car. You know, ‘I’d like you to put it back when you’re done.’ ” Cite as 357 Or 365 (2015) 369

acknowledged that the methamphetamine and drug para- phernalia were his. Before trial, defendant moved to suppress both the physical evidence found during the search of his car and the statements that he made afterwards on the ground that the deputies had unlawfully extended the stop. At the hearing on that motion, defendant raised another ground for suppress- ing that evidence. He argued that, when Robeson asked him if there were anything he should be concerned about in the car, Robeson violated his state and federal Miranda rights.5 In the trial court, the state responded that Robeson’s question did not constitute interrogation. In the state’s view, that question was no different from the background ques- tions regarding identity that had preceded it. The trial court denied defendant’s suppression motion. It ruled that the dep- uties had not unlawfully extended the stop, and it agreed with the state that Miranda warnings were not required, apparently on the ground that Robeson’s question had not constituted interrogation. The trial court accordingly denied defendant’s suppression motion and ruled that the physical evidence found in defendant’s car and the warned state- ments that he made to the second deputy were admissible at his trial. Considering that and other evidence, the jury found defendant guilty of possessing methamphetamine. On appeal, defendant challenged the trial court’s ruling on his suppression motion. He argued that, once Robeson placed him in the back of his patrol car and hand- cuffed him, Article I, section 12, required Robeson to advise him of his Miranda rights before asking him whether there was anything in his car that should concern the deputies.6 The state, in response, conceded that Robeson had violated 5 Defendant did not argue in the trial court that the previous questions that Robeson asked regarding defendant’s identity constituted interrogation or that, in asking those questions, Robeson had violated his Miranda rights. Cf. State v. Cunningham, 179 Or App 498, 501, 40 P3d 535 (2002) (explaining that the federal definition of interrogation, which this court adopted for the purposes of Article I, section 12, contains an exception for questions “normally attendant to arrest and custody”). 6 Defendant also argued that, even if he invited the deputies to search his car, the scope of his consent did not extend to opening the fanny pack in his car, where the deputies found methamphetamine and drug paraphernalia. The Court of Appeals did not reach that issue, and the parties have not briefed it on review. 370 State v. Delong

Article I, section 12, of the Oregon Constitution when he asked defendant that question without first advising him of his Miranda rights. The state argued, however, that, because defendant’s invitation to search his car attenuated the taint of the Miranda violation, the physical evidence that the depu- ties discovered in the car was not the product of the violation. The Court of Appeals held that Article I, section 12, required Robeson to advise defendant of his Miranda rights before asking him if there was anything in the car that should concern the deputies. Delong, 260 Or App at 724. It also held that the physical evidence the deputies found in defendant’s car “derived from” that violation under this court’s decision in Vondehn. Id. at 726-27. The court reasoned that Robeson “exploited, or took advantage of, the Article I, section 12, violation to obtain [defendant’s] consent; he offered consent during a custodial interrogation while denying any wrong- doing.” Id. at 727. The court accordingly held that the trial court should have suppressed the physical evidence discov- ered in defendant’s car and the statements that defendant made after receiving Miranda warnings. We allowed the state’s petition for review to consider whether, under Article I, section 12, the physical evidence that the deputies discovered in defendant’s car “derived from” the earlier Miranda violation. See Vondehn, 348 Or at 476 (stating that standard). On that issue, defendant argues that, because his invitation to search his car was the foreseeable result of the deputy’s unwarned question, the evidence that the deputies found in his car derived from that Miranda violation and should be suppressed. The state responds that, because Miranda is a judge-made rule and not a constitutional right, we should suppress only the evidence that resulted directly from the Miranda violation. In the state’s view, we should not suppress the evidence that resulted directly from a Miranda violation and the “fruit of the poisonous tree,” as we ordi- narily do for state constitutional violations.7 Alternatively,

7 The fruit of the poisonous tree has been defined as “challenged evidence [that] is ‘secondary’ or ‘derivative’ in character,” as when “a confession is obtained after an illegal arrest, physical evidence is located after an illegally obtained confession, or an in-court identification is made following an illegally conducted pretrial identification.” Wayne R. LaFave, Jerold H. Israel, Nancy J. King, Orin S. Kerr, 3 Criminal Procedure § 9.3(a) (3d ed 2007 and supp 2014). Cite as 357 Or 365 (2015) 371

the state argues that, even if we suppress both the direct evidence resulting from the Miranda violation and the “fruit of the poisonous tree,” our Article I, section 9, cases demon- strate that the evidence that the deputies found in defendant’s car was not the fruit of the poisonous tree and thus did not derive from the Miranda violation. Our decision in Vondehn provides the starting point for our analysis. Accordingly, we first describe that decision. We then explain that the specific holding in Vondehn does not control the resolution of this case. We also explain that, even if we apply the remedial standard that we ordinarily apply to Article I, section 9, violations, defendant’s invitation to search his car attenuated the taint from the Miranda viola- tion. Finally, we address defendant’s argument and the dis- sents’ view that, even if defendant’s invitation to search his car would be sufficient to attenuate the taint of an Article I, section 9, violation, his invitation was not sufficient to atten- uate the taint of an Article I, section 12, violation. We begin with our decision in Vondehn. In that case, the officers asked the defendant whether he owned a back- pack found in a stopped car, whether it contained marijuana, and whether they could search it. Vondehn, 348 Or at 484. The defendant answered “yes” to each of those questions. Id. Pursuant to the defendant’s consent, the officers searched his backpack, found marijuana, advised the defendant of his Miranda rights, and then asked him additional questions about the marijuana that they had found. Id. at 484-85. On review, this court held that, under Article I, section 12, the officers should have advised the defendant of his Miranda rights before asking him whether he owned the backpack. Id. at 476. It held that the answers that the defendant gave before being advised of his Miranda rights and the mari- juana that the officers found in his backpack should have been suppressed. Id. at 476-77. It also held, however, that the answers that the defendant gave after being advised of his Miranda rights should not be suppressed. Id. at 486. Much of this court’s opinion in Vondehn focused on the state’s argument that we should interpret Article I, sec- tion 12, the same way that the plurality in United States v. Patane, 542 US 630, 124 S Ct 2620, 159 L Ed 2d 667 372 State v. Delong

(2004), would have interpreted the Fifth Amendment. See Vondehn, 348 Or at 470 (noting the state’s reliance on the plurality opinion in Patane).8 This court did not accept that argument. Relying on our cases interpreting Article I, sec- tion 12, the court explained that “the Oregon Constitution requires Miranda warnings” and, as a result, the failure to give those warnings, when required, is itself a constitutional violation that requires a remedy. Id. at 475-76. The court also rejected the state’s argument that vindicating a defen- dant’s Article I, section 12, Miranda rights requires only that his or her unwarned statements be suppressed. Id. It held that Article I, section 12, also precludes the state from using “physical evidence that is derived from [a Miranda] violation to prosecute a defendant.” Id. Having reached that conclusion, the court turned to the question whether the marijuana that the officers had found in the defendant’s backpack “derived from” the Miranda violation in that case. On that question, the court reasoned: “In this court, the state makes no argument that the request for consent to search or the seizure of the mari- juana 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.” Id. at 476-77. The court’s conclusion that the marijuana derived from the Miranda violation in Vondehn thus appears to have turned primarily on the absence of any argument to

8 The plurality opinion in Patane would have held that, under the Fifth Amendment, the “mere failure to give Miranda warnings does not, by itself, vio- late a suspect’s constitutional rights or even the Miranda rule.” 542 US at 641 (opinion of Thomas, J.). In the plurality’s view, “[p]otential violations occur, if at all, only upon the admission of unwarned statements into evidence at trial.” Id. The plurality reasoned that, because the mere failure to advise a suspect of his or her Miranda rights does not violate either the Fifth Amendment or the Miranda rule, there is no reason to apply the “fruit of the poisonous tree” doctrine to that failure. Id. at 643-44. It followed, the plurality reasoned, that excluding state- ments obtained in violation of Miranda from a defendant’s criminal trial is the only remedy that the Fifth Amendment requires. Id. at 644. Cite as 357 Or 365 (2015) 373

the contrary. See id.; see also id. at 490 (Linder, J., concur- ring) (reaching a similar conclusion). The court then turned to the question whether the statements that the defendant made after receiving Miranda warnings should be suppressed. In analyzing that state constitutional issue, the court employed a multifactor test that it drew from Missouri v. Seibert, 542 US 600, 124 S Ct 2601, 159 L Ed 2d 643 (2004), and Oregon v. Elstad, 470 US 298, 105 S Ct 1285, 84 L Ed 2d 222 (1985). Vondehn, 348 Or at 480-81. The court explained that, in considering those factors, it was not seeking to determine whether the defen- dant’s warned statements were the fruit of the poisonous tree. Id. at 482.9 Rather, it was seeking to determine whether the belated Miranda warnings were effective in ensuring that the defendant’s decision to waive his right against self- incrimination was knowing and voluntary. Id. Considering those factors, the court concluded that the belated Miranda warnings in that case had been effective. Id. at 486. The statements that the court held admissible in Vondehn followed belated Miranda warnings, and the test that the court articulated (whether the belated warn- ings were effective) applies in that circumstance. When no belated Miranda warnings have been given, the question whether the taint flowing from a Miranda violation has been attenuated will vary depending on the totality of the cir- cumstances. See State v. Jarnagin, 351 Or 703, 716, 277 P3d 535 (2012). In deciding whether the taint has been attenu- ated, the court has considered, among other things: “the nature of the violation, the amount of time between the violation and any later statements, whether the suspect remained in custody before making any later statements, subsequent events that may have dissipated the taint of the earlier violation, and the use that the state has made of the unwarned statements.” Id. 9 The court explained that “a court does not use those circumstances to attempt to determine the psy- chological effect that the particular police course of conduct had on the par- ticular defendant or whether the initial failure to warn caused the particular defendant to make the post-Miranda statements.” Vondehn, 348 Or at 482. Rather, the test was an objective one. Id. 374 State v. Delong

With that background in mind, we turn to this case. At first blush, the holding in Vondehn regarding the search of the defendant’s backpack in that case would seem to con- trol the resolution of this case. As in Vondehn, the deputies in this case did not advise defendant of his Miranda rights before obtaining his consent to search his car. This case dif- fers from Vondehn, however, in at least two respects. As dis- cussed above, this court did not have occasion in Vondehn to explore, at any length, whether the marijuana found in the defendant’s backpack in that case derived from the pre- ceding Miranda violation, in large part because the state had not argued that it did not. Here, the state has argued that the physical evidence that the deputies found did not derive from the Miranda violation. Moreover, as is often true in cases such as this, the issue in this case arises in a different factual posture from the issue in Vondehn. As discussed below, Robeson’s unwarned question in this case was open-ended; defendant’s direct response to the question was exculpatory; and he invited the deputies to search his car without an express request for consent. We accordingly cannot say that the specific holding in Vondehn controls our resolution of this case, and we look instead to the factors identified in Jarnagin to determine whether the physical evidence that the deputies found in defendant’s car was the product of the preceding Miranda violation. On that issue, we note that the amount of time that passed between the Miranda violation and the discov- ery of the physical evidence was brief. Additionally, defen- dant remained in custody during the encounter. In those respects, this case is similar to Vondehn. As noted above, however, this case differs from Vondehn in other respects, and we focus initially on the primary factual difference, defendant’s invitation to search his car. See Jarnagin, 351 Or at 716 (explaining that, in deciding attenuation, we con- sider, among other things, “subsequent events that may have dissipated the taint of the earlier violation”). When Sergeant Robeson asked defendant “if there was anything we should be concerned about” in his car, defendant “told [him] ‘no,’ and that if [the officers] wanted to search the vehicle [they] could.” Defendant’s answer Cite as 357 Or 365 (2015) 375

divides into two parts: (1) a statement that nothing in his car should concern the officers and (2) an invitation to the officers to search his car if they wanted to do so. The sec- ond part of defendant’s answer can be viewed in one of two ways: either as a volunteered response that was admissible in defendant’s criminal trial or, even if defendant’s response were not admissible in his criminal trial, as evidence of attenuation that was relevant to his motion to suppress and admissible in the hearing on that motion. Cf. State v. Wright, 315 Or 124, 131, 843 P2d 436 (1992) (explaining that the fact that evidence is inadmissible under the evidence code at trial does not mean that it is inadmissible in deciding a motion to suppress). We begin with the first way of looking at defen- dant’s response. The trial court found that defendant had “volunteered” the invitation to search. The concept of a vol- unteered statement has a unique place in Miranda jurispru- dence. In announcing the requirement that officers advise custodial suspects of their rights before questioning them, the United States Supreme Court was careful to recognize that that requirement does not preclude the admission of a defendant’s volunteered statements in his or her criminal trial. Miranda v. Arizona, 384 US 436, 478, 86 S Ct 1602, 16 L Ed 2d 694 (1966) (“Volunteered statements of any kind are not barred by the Fifth Amendment and their admissi- bility is not affected by our holding today.”). The volunteered statements that the Court dis- cussed in Miranda were statements that a suspect made in custody without any questioning by the police. See id.; Wayne R. LaFave, Jerold H. Israel, Nancy J. King, and Oris S. Kerr, 2 Criminal Procedure § 6.7(d) (3d ed 2007 and 2014 supp) (discussing volunteered statements). Other courts have recognized that the concept also applies to nonrespon- sive statements that a suspect makes during custodial ques- tioning. See LaFave et al, 2 Criminal Procedure § 6.7(d) (dis- cussing cases). Specifically, those courts have held that, to the extent that a defendant’s answer is not responsive to the officer’s question, then the answer is a volunteered state- ment, as the Court used that term in Miranda, and admis- sible in the defendant’s criminal trial. See id. 376 State v. Delong

In this case, defendant’s invitation to search his car was nonresponsive in one sense. Robeson did not ask if he could search defendant’s car. He asked if there were anything in the car that he should be concerned about. Defendant’s answer went beyond what Robeson had asked and included an offer for the officers to search his car “if [they] wanted to.” As defendant argues, however, Robeson’s question can be viewed as prompting the second part of defendant’s answer. See State v. Unger, 356 Or 59, 79, 333 P3d 1009 (2014) (explaining that asking a defendant whether he had any drugs or guns in his apartment could be viewed as prompting the defendant’s invitation to “go ahead and look”). Viewed in that manner, the second part of defen- dant’s answer may not be sufficiently nonresponsive to come within the concept of a “volunteered” statement that the Court identified in Miranda and thus may not be admissible as evidence in defendant’s criminal trial.10 Even if we assume that defendant’s response to Robeson’s question was not “volunteered,” as the Court used that term in Miranda, defendant’s response still reflects a volitional act on his part and, as such, implicates another strand of our case law. In considering a related issue, this court has held that similar offers are sufficient to atten- uate the taint of an Article I, section 9, violation. State v. Rodriguez, 317 Or 27, 854 P2d 399 (1993); State v. Kennedy, 290 Or 493, 624 P2d 99 (1981).11 The point of those cases is not that the statement itself (the defendant’s offer) was

10 Justice Baldwin’s dissent devotes some time to explaining that defendant’s invitation to search his car was not volunteered. We note that Robeson testified that defendant’s invitation was “volunteered,” and the trial court expressly cred- ited Robeson’s testimony on that point. As a factual matter, describing defen- dant’s invitation as “volunteered” seems accurate. Of course, the legal effect of that invitation is a separate question. And, as we explain above, we assume that defendant’s invitation was not “volunteered” in the sense that the Court used that term in Miranda. It follows, we think, that our difference with the dissent on this point is not substantial. 11 The issue in those Article I, section 9, cases was whether a defendant’s invitation to search his or her effects attenuated the taint of an unlawful seizure. The issue in this Article I, section 12, case is whether defendant’s invitation to search attenuated the effects of an unwarned question that followed a lawful cus- todial seizure. Both types of cases involve situations that can influence a defen- dant’s ability to make an independent decision. And, as the concurrence explains, the factors that we have considered in deciding attenuation in both situations are virtually the same. Cite as 357 Or 365 (2015) 377

admissible in the defendant’s criminal trial. Rather, the point is that the offer was sufficient to attenuate the taint of the preceding constitutional violation, making subsequently discovered evidence admissible. This court’s decision in Kennedy illustrates that line of cases. In Kennedy, the police approached the defen- dant as he was leaving the Portland airport. 290 Or at 495. Acting on information that the defendant fit a “drug smug- gler’s profile,” the officers asked to talk to him. Id. When the defendant asked one of the officers why they wanted to talk, the officer explained that he “had information that led [him] to believe that [defendant] may be carrying narcotics on his person or in his luggage.” Id. at 496 (brackets in original). “Defendant denied that he was carrying narcotics and said, ‘Would you like to search my luggage?’ ” Id. The court noted that, when the defendant made that offer, the officer “had not made any request for consent to search [the] defendant or his luggage.” Id. During the ensuing search, the officers found and seized a vial with cocaine residue on it. Id. In deciding whether that evidence was the product of an unlawful stop, this court assumed that the officers had stopped the defendant and that they lacked reasonable sus- picion to do so. Id. at 499. The court also recognized that the stop was the “but for” cause of the officers’ discovery of the evidence. Id. at 500-01. However, relying on, among other things, the “defendant’s offer to let [the officer] search his luggage without a prior request for consent,” the court con- cluded that the discovery of the evidence was sufficiently attenuated from any illegality to say that it did not derive from it. Id. at 504-06 (explaining that the defendant’s unso- licited offer to search was the “[m]ost importan[t]” consider- ation in reaching its conclusion). The decision in Rodriguez is to the same effect.12 See also State v. Crandall, 340 Or 12 In Rodriguez, state and federal officers arrested the defendant at his home. 317 Or at 29. The officers advised the defendant of his Miranda rights. Id. at 30. In response to the question, “Do you have any drugs or guns in the house,” the defendant replied, “No, go ahead and look.” Id. The officers did so and found a gun, which the defendant later sought to suppress as the product of an unlawful arrest. Similarly to Kennedy, the court accepted the state’s concession that the arrest violated Article I, section 9. Id. at 37. The court also recognized that the arrest was the “but for” cause of defendant’s statement, “Go ahead and look.” Id. at 39-40. The court concluded, however, that defendant’s unsolicited offer 378 State v. Delong

645, 136 P3d 30 (2006) (defendant’s unilateral act in hid- ing drugs under a parked car after officers unlawfully had stopped and directed him to come over and talk to them attenuated the taint of the unlawful stop). In Unger, this court explained that Kennedy and Rodriguez stand for the proposition that, “in some situations, a defendant’s voluntary consent itself may be sufficient to demonstrate that the unlawful conduct did not affect or had only a tenuous connection to the evidence produced.” 356 Or at 77-78. Unger thus reaffirmed that, when a defendant’s consent to a search either was not affected by or was only tenuously connected to a prior illegality, the defendant’s vol- untary consent can be sufficient to break the causal chain. The court identified three factors that bear on when volun- tary consent will attenuate a prior illegality: “That legal determination—whether, in the circumstances of a particular case, consent has so attenuated the con- nection between the prior illegal conduct and the evidence obtained in the consent search—requires a court to con- sider the illegal conduct that comprised the stop or search, the character of the consent, and the causal relationship between the two.” Id. at 78.13 Regarding the first factor (the nature of the ille- gal conduct), the state has conceded that Robeson violated defendant’s Article I, section 12, rights when he asked him if there were anything in his car that should concern the deputies. That violation can hardly be characterized as egregious, however. This was not the sort of prolonged sta- tionhouse questioning that concerned the United States Supreme Court in Miranda. See 384 US at 448-55 (describ- ing interrogation techniques designed to break down a sus- pect’s will). Robeson did not engage “in repeated efforts to wear down [defendant’s] resistance.” See State v. Foster, 288 Or 649, 655-56, 607 P2d 173 (1980) (considering that situa- tion) (internal quotation marks omitted); State v. Mendacino,

to search his house sufficiently attenuated the taint of the unlawful arrest and upheld the admission of the evidence found in the ensuing search. Id. at 41-42. 13 The factors that the court identified in Unger are a subset of the factors that the court identified in Vondehn and Jarnagin. Cite as 357 Or 365 (2015) 379

288 Or 231, 238, 603 P2d 1376 (1980) (same). The trial court found that only two to three minutes elapsed between the time that Robeson stopped defendant and defendant’s invita- tion to search his car. And other than the initial background questions he asked, Robeson asked defendant only one ques- tion: “[I]f there was anything we should be concerned about” in defendant’s car. Defendant has not argued that his response to that question was “actually coerced” within the meaning of Article I, section 12, nor could he reasonably do so. Similarly, he has not argued that Robeson deliberately sought to violate his state constitutional rights. Although Robeson’s question went too far, his question was not dissimilar from a question that the Court of Appeals had held falls within an exception to the state constitutional Miranda requirement. See State v. Cunningham, 179 Or App 498, 504, 40 P3d 535 (2002) (asking defendant “whether he had anything that was sharp or would hurt him” before the officer performed a lawful pat- down search did not constitute “interrogation” as defined by the United States Supreme Court and adopted by the Oregon Supreme Court).14 The violation was not egregious. The second factor we consider is the character of defendant’s consent. As noted, defendant invited the depu- ties to search his car if they wanted to do so. Defendant’s invitation to search his car in this case is virtually identi- cal to the invitations in Kennedy and Rodriquez, which this court held attenuated the taint of the unlawful seizures in those cases. It may be that Robeson’s question in this case prompted defendant’s invitation to search his car, as defen- dant argues, but the question that Robeson posed was more open-ended and thus more benign than the officers’ state- ments in Kennedy and Rodriguez. In Kennedy, the officer told the defendant that they had reason to believe that he 14 We express no opinion on whether the Court of Appeals was correct in Cunningham in applying that exception to the definition of “interrogation.” We note the Court of Appeals decision only to observe that Robeson’s question was similar to one that the Court of Appeals had approved. Robeson’s question differed in two respects, however. Unlike the question in Cunningham, which focused on harm to the officer as he carried out procedures designed to effectuate a lawful seizure, Robeson’s question was not limited to officer safety, and it asked about items in the car even though defendant was handcuffed and seated in the deputy’s patrol car. 380 State v. Delong

had drugs on his person or in his luggage. 290 Or at 496. In Rodriguez, the officer asked if the defendant had any drugs or guns in his apartment. Rodriguez, 317 Or at 30. In this case, Robeson asked only whether there was anything that the deputies should be concerned about in defendant’s car. Finally, we consider the causal connection between the violation and defendant’s invitation. This is not a case in which Robeson’s unwarned questioning left “ ‘little, if any- thing, of incriminating potential * * * unsaid.’ ” See Jarnagin, 351 Or at 722 (quoting Seibert, 542 US at 616-17 (plurality opinion)). Rather, defendant told the deputies that he did not have anything of concern in his car before extending an invitation to them to search his car if they wanted to. Not only did the deputies not trade on the first part of defen- dant’s response, but there was nothing on which to trade. Nothing that defendant said in the first part of his response to Robeson’s unwarned question impaired defendant’s abil- ity to make an independent decision to invite the deputies to search his car if they wanted to do so. In that respect, this case is no different from Kennedy and Rodriguez. Under the analysis in Kennedy and Rodriguez, defendant’s invitation to search his car attenuated the taint flowing from Robeson’s unwarned question. The evidence that the deputy found in defendant’s car did not derive from the preceding Miranda violation.15 Defendant and the two dissenting opinions take a different position. They reason that, even if the invitations to search in Kennedy and Rodriguez were sufficient to atten- uate the taint of the Article I, section 9, violations in those cases, defendant’s invitation in this case is not sufficient to

15 Justice Walters’ dissent compares this case to Jarnagin, where the offi- cers obtained from defendant, during multiple extended interviews in violation of Miranda, an explanation as to how his daughter had been injured and an agree- ment to reenact that explanation the next morning while the officers videotaped him. 351 Or at 718. Our holding that the resulting videotape was the product of the earlier violations turned in large part on the fact that the defendant’s unwarned statements formed the script that he acted out the next morning while being videotaped. Id. We do not view the specific circumstances that we consid- ered in Jarnagin, as the dissent appears to do, as exhausting the totality of the circumstances that can bear on whether subsequently discovered evidence is the product of an earlier Miranda violation. Nor do we view Jarnagin as standing for the proposition that a defendant’s voluntary invitation to search can never atten- uate the taint of a Miranda violation. Cite as 357 Or 365 (2015) 381

remedy the taint of a Miranda violation. We begin with an argument that defendant alone advances. Defendant argues that Kennedy and Rodriguez are inapposite because, under Article I, section 12, his invita- tion to search his car will attenuate the taint of the Miranda violation only if the invitation was extended with knowl- edge of his right against self incrimination. As defendant notes, this court stated in Vondehn that Article I, section 12, requires Miranda warnings “to ensure that a person’s waiver [of his or her Article I, section 12, rights] is knowing as well as voluntary.” See Vondehn, 348 Or at 474. It follows from that proposition, defendant contends, that, because his invitation to search his car was made without knowledge of his Miranda rights, that invitation should have little or no weight in the attenuation analysis. One difficulty with that argument is that it fails to distinguish two separate issues. The issue in this case is not whether defendant’s response to Robeson’s question was knowing and thus admissible in his criminal trial. It may not have been.16 Rather, the issue in this case is whether defendant’s response was admissible in the hearing on his suppression motion to determine whether the physical evi- dence discovered in his car derived from the Miranda vio- lation. On the latter question, defendant offers no basis for saying that his invitation to search cannot be considered at a suppression hearing as evidence of attenuation. Cf. Wright, 315 Or at 131 (explaining that the fact that evidence is inad- missible under the evidence code at trial does not mean it is inadmissible in determining a motion to suppress). We customarily have considered a suspect’s responses to unwarned questioning in determining whether subsequently discovered evidence derived from or was a product of an earlier Miranda violation. See Jarnagin, 351 Or at 722-23 (considering the defendant’s responses to unwarned questioning in deciding attenuation); Vondehn, 16 In the trial court, defendant did not move to suppress his response to Robeson’s question. He moved to suppress the physical evidence found in his car and the warned statements that he later made. In any event, even if defen- dant had moved to suppress his response to Robeson’s question and even if that response should not have been admitted in his criminal trial, any error in admit- ting the response was harmless. 382 State v. Delong

348 Or at 485-86 (same). For example, among the factors that we considered in deciding attenuation in Jarnagin were “the use that the state has made of the unwarned state- ments” and whether “the unwarned interrogation left ‘little, if anything, of incriminating potential * * * unsaid.’ ” See 351 Or at 716, 722 (quoting Seibert, 542 US at 616-17 (plural- ity opinion)). Those factors necessarily entail considering a defendant’s responses to unwarned questioning in deciding whether subsequently discovered evidence was the product of an earlier Miranda violation.17 Defendant advances a second argument, which both dissents also raise. They reason that, even if defendant’s invitation would have been sufficient to attenuate the taint of an unconstitutional seizure, as this court held in Kennedy and Rodriguez, something more is required to attenuate the taint of a Miranda violation. They conclude that, because Miranda requires warnings following an arrest and because an arrest entails a greater level of restraint than a stop, an event that will be sufficient to attenuate the taint of an unlawful stop will be insufficient to attenuate the taint of a Miranda violation. Neither defendant nor the dissenting opinions, however, cite any case that stands for that cate- gorical proposition. If anything, the cases that address the issue have held that less is required to attenuate a Miranda violation than is required to attenuate an unconstitutional seizure. See Dickerson v. United States, 530 US 428, 440-41, 120 S Ct 2326, 147 L Ed 2d 405 (2000); Elstad, 470 US at 306; Patane, 542 US at 644-45 (Kennedy, J., concurring in the judgment). The United States Supreme Court explained in Elstad that “a procedural Miranda violation differs in sig- nificant respects from violations of the Fourth Amendment, which have traditionally mandated a broad application of the ‘fruits’ doctrine.” 470 US at 306. The Court accordingly held in Elstad that the defendant’s subsequent warned state- ments were admissible against him in his criminal trial 17 There is a suggestion in Justice Walters’ dissent that, because defendant did not know that he had a right to remain silent, his invitation to search his car cannot be considered in deciding attenuation. That reasoning proves too much. If that were correct, no invitation to search following a Miranda violation could be considered without belated Miranda warnings. Cite as 357 Or 365 (2015) 383

without regard to whether those statements were the fruit of his earlier admissions obtained in violation of Miranda. See id. at 316-17; see also Seibert, 542 US at 612 n 4 (plural- ity opinion) (same). And the Court rejected an argument in Elstad that the degree of attenuation required to purge the taint of coerced or compelled statements applies equally to statements obtained as a result of a “technica[l]” Miranda violation. 470 US at 318. To be sure, in Elstad, the Court justified a more lim- ited remedy for Miranda violations than Fourth Amendment violations on the ground that Miranda was a judge-made rule, not a constitutional right. See id. at 305-06. Since then, however, the Court has recognized that Miranda warnings are “constitutionally based,” but it has adhered to its con- clusion in Elstad that Miranda violations do not require as extensive a remedy as a Fourth Amendment violation and that the same degree of attenuation is not required. Dickerson, 530 US at 440-41; accord Seibert, 542 US at 612 n 4 (plurality opinion). As we read those decisions, they adhered to the conclusion in Elstad because of the prophylactic nature of the Miranda right. The purpose of Miranda warnings is “[t]o protect a person’s right against compelled testimony.” Jarnagin, 351 Or at 713. To ensure that that right is pro- tected, the United States and Oregon constitutions require officers to advise suspects who are in custody or comparable circumstances of their Miranda rights. However, this court has never equated the point at which the Miranda right attaches with the point at which a person’s statements are either actually compelled or coerced. See id. at 724 (distin- guishing statements obtained in violation of a defendant’s Article I, section 12, right to Miranda warnings from state- ments obtained in violation of Article I, section 12, as a result of actual coercion). Between those two points lies a range of circumstances that can affect whether subsequently dis- covered evidence derives from the failure to give required Miranda warnings. In this case, defendant was in custody or compara- ble circumstances, and his Miranda rights attached at that point. The dissents would give talismanic significance to 384 State v. Delong

that fact and hold that, as a result, defendant was disabled from inviting the officers to search his car. As we explained in Jarnagin, however, the question whether the circum- stances are sufficient to attenuate the taint of an officer’s failure to give Miranda warnings will turn on the facts of each case. 351 Or at 716. That entails a consideration of the extent to which the nature and extent of the custodial ques- tioning affected a suspect’s decision to invite the search. In this case, we are hard pressed to say that the failure to give required Miranda warnings disabled defen- dant from making an independent decision. It is true that defendant was in custody or comparable circumstances. However, that is true in every case in which an officer fails to give required Miranda warnings. Defendant was not detained in the stationhouse for an extended period of time, nor was he subjected to the sort of extended questioning that caused the Court to require Miranda warnings in the first instance. Rather, the detention was brief, only two to three minutes the trial court found. Robeson asked only one ques- tion that went beyond determining defendant’s identity, and defendant’s response to that question was not inculpatory.18 Given those circumstances, we conclude that our holdings in Kennedy and Rodriguez provide persuasive guidance for deciding this case. One other consideration cuts against the conclusion that defendant urges and that the dissents would reach. Defendant concedes in his brief on the merits that an officer lawfully may ask a suspect who is in custody or compelling circumstances for consent to search without first advising the suspect of his or her Miranda rights. It follows that, if the suspect consents and the officer finds incriminating evi- dence in the ensuing search, that evidence will be admissi- ble in the suspect’s criminal trial, even though the suspect 18 Justice Walters’ dissenting opinion states that the deputies “went on a fish- ing expedition—deliberately interrogating defendant and seeking incriminating evidence without first warning defendant of his right to remain silent and con- sult a lawyer.” 357 Or at 397 (Walters, J., dissenting). The trial court did not find, however, that the deputies “deliberately interrogat[ed]” defendant, nor did it find that they were on a “fishing expedition” or “seeking incriminating informa- tion.” What Robeson actually said to defendant (even the words that defendant recounted) seems far milder than might appear from the dissent’s description of the events. Cite as 357 Or 365 (2015) 385

was in custody or compelling circumstances when he or she consented. If, as defendant concedes, an officer need not advise a suspect in custody of his or her Miranda rights before ask- ing for consent to search, it is difficult to see why a suspect who is in custody cannot invite an officer to search. It may be, as defendant argues, that Robeson’s question prompted defendant’s invitation in this case. But, even if that is true, then Robeson’s question functioned implicitly the same way that an explicit request for consent would have, and defen- dant points to nothing in the first part of his answer to Robeson’s question (that there was nothing in his car of con- cern to the deputies) that would have affected or somehow tainted his decision to invite the deputies to search his car. Given Kennedy, Rodriguez, and defendant’s concession, we are not persuaded that we should automatically give less effect to invitations to search that follow a Miranda viola- tion than we give invitations to search that follow an unlaw- ful seizure. One final matter requires discussion. At trial and on appeal, defendant argued that the deputies exceeded the scope of his invitation when they opened a fanny pack they found in his car, which contained methamphetamine and drug paraphernalia. The Court of Appeals did not reach that issue because it held that defendant’s invitation to search his car and the resulting search were the fruit of the Miranda violation. Because we reach a different conclu- sion, it is necessary to resolve defendant’s argument that the officer’s search exceeded the scope of defendant’s invitation. The parties have not briefed that issue on review, and we conclude that the case should be remanded to the Court of Appeals so that it can decide that issue in the first instance. If the Court of Appeals finds that the deputy’s search did not exceed the scope of defendant’s invitation, then the remaining question is whether the statements that defen- dant made to the deputy after receiving belated Miranda warnings were admissible. As we understand defendant’s argument on that issue, it rests on the proposition that the deputies unlawfully discovered the physical evidence in his car and that, as a result, the belated Miranda warnings 386 State v. Delong

he received were not effective to render his statements voluntary. If the deputies lawfully discovered the physical evidence in defendant’s pack, then the physical evidence and defendant’s warned statements presumably would be admissible. Conversely, if the deputy’s search exceeded the scope of defendant’s consent, then the question will be, as it was in Vondehn, whether the belated Miranda warnings were effective. See Vondehn, 348 Or at 485-86 (holding that belated Miranda warnings were effective even though offi- cers unlawfully had discovered marijuana in the defendant’s backpack). The decision of the Court of Appeals is reversed, and the case is remanded to the Court of Appeals for further proceedings consistent with this decision. BREWER, J., concurring. I agree with the majority’s conclusion that the phys- ical fruits of the invited search in this case are not suppress- ible as a result of the admitted violation of defendant’s rights under Article I, section 12, of the Oregon Constitution. However, I distance myself from two aspects of the majori- ty’s discussion of the principles governing the suppression of evidence for violations of Article I, section 12. First, I agree with the majority that there is no per- suasive support for the proposition that something more is required to attenuate the taint of a Miranda violation than a violation of Article I, section 9. 357 Or at 382. However, I do not believe that less is required to attenuate a Miranda violation than is required to attenuate an unconstitutional search or seizure. Although this court has held that the two attenuation analyses are not identical, they are animated by similar concerns, and, in my view, the attenuation frame- works for violations of rights under Article I, section 12, and Article I, section 9, are structured—and ought to be applied—in comparable terms. To protect a person’s right against compelled self- incrimination under Article I, section 12, this court has held that, before questioning, law enforcement must give Miranda warnings to a person who is in “full custody” or Cite as 357 Or 365 (2015) 387

in circumstances that “create a setting which judges would and officers should recognize to be ‘compelling.’ ” State v. Jarnagin, 351 Or 703, 713, 277 P3d 535 (2012) (quoting State v. Smith, 310 Or 1, 7, 791 P2d 836 (1990)). When an officer fails to give the requisite warnings, a court must sup- press not only the statements that a suspect makes in direct response to unwarned questioning but also evidence that derives from that constitutional violation. State v. Vondehn, 348 Or 462, 476, 236 P3d 691 (2010).1 In Jarnagin, this court applied a totality of the cir- cumstances test in determining whether physical or testi- monial evidence derived from an earlier Miranda violation must be suppressed. Jarnagin, 351 Or at 716. In doing that, the court relied on Vondehn, 348 Or at 482, where it had directed courts to consider “all relevant circumstances” in deciding whether belated Miranda warnings were effec- tive in ensuring a knowing and voluntary waiver of rights. Among other factors, this court in Jarnagin stated that “the nature of the violation, the amount of time between the vio- lation and any subsequent statements, whether the suspect remained in custody before making any later statements, subsequent events that may have dissipated the taint of the earlier violation, and the use that the state has made of the unwarned statements” were proper considerations in a “fact-intensive” inquiry. Jarnagin, 351 Or at 716-17. The court in Jarnagin distinguished the attenua- tion inquiry under Article I, section 12, from the attenuation analysis under Article I, section 9:

1 In Vondehn, this court examined the basis for the requirement that law enforcement inform people in custody of their right against self-incrimination under Article I, section 12. The court explained that, “[b]ecause 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 vol- untary. 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.” 348 Or at 474. 388 State v. Delong

“Defendant argues that we should apply the particular methodology set out in State v. Hall, 339 Or 7, 24-25, 115 P3d 908 (2005), to determine whether evidence is the prod- uct of a Miranda violation. The Hall methodology applies to violations of Article I, section 9. It does not apply to viola- tions of Article I, section 12. Cf. Brown v. Illinois, 422 US 590, 602-03, 95 S Ct 2254, 45 L Ed 2d 416 (1975) (explain- ing that the question whether a statement is the product of a Fifth Amendment violation differs from the question whether it is a product of a Fourth Amendment violation). In Vondehn, we cited Hall once in describing the Court of Appeals’ reasoning, see 348 Or at 465, but we did not cite Hall afterwards or apply its methodology in determining whether the physical evidence in that case derived from the Miranda violation, see id. at 476. Similarly, we do not apply Hall’s methodology here.” Jarnagin, 351 Or at 717 n 9. Despite what the court said in Jarnagin, in reaching its decision in this case, the majority has borrowed liberally from the attenuation analyses of consent search cases under Article I, section 9. That, I submit, is because the analy- ses for both types of constitutional violations are aimed at similar concerns. Whenever the state has obtained evidence following the violation of a defendant’s Article I, section 9, rights, it is presumed that the evidence was tainted by the violation and must be suppressed. State v. Unger, 356 Or 59, 84, 333 P3d 1009 (2014). The state may rebut that pre- sumption by establishing that the disputed evidence “did not derive from the preceding illegality.” Hall, 339 Or at 25. When determining whether a defendant’s consent to search derived from police misconduct, courts are to consider the totality of the circumstances, including the temporal prox- imity between the misconduct and the consent; the existence of any intervening or mitigating circumstances; the nature of the misconduct, including its purpose and flagrancy and whether the police took advantage of it; and the volun- tariness of the consent. Unger, 356 Or at 89-93. Although expressed in different words, those factors closely track the factors that the court in Jarnagin indicated are pertinent in an Article I, section 12, attenuation analysis. In short, I do not perceive that the bar is set higher or lower for a violation of either provision. Cite as 357 Or 365 (2015) 389

Second, I agree with the majority that a knowing waiver of Miranda rights is not required for a court to con- clude that an ensuing consent or invitation to search is enti- tled to weight in the attenuation analysis. 357 Or at 381- 82. However, I would place more explicit emphasis on the absence of a knowing waiver as an aspect of the totality of circumstances. The reason is simple: a person’s lack of knowledge that he or she has a right not to self-incriminate should raise the same rebuttable presumption that evidence obtained in a consent search was tainted by the preceding illegality that arises in an Article I, section 9, attenuation analysis. In other words, a causal connection between a Miranda violation and the discovery of challenged evidence requires the state to establish the existence of circum- stances that either legally or factually break that connec- tion. See Vondehn, 348 Or at 476; see also id. at 490 (Linder, J., concurring). To be sure, the violation casts a meaningful shadow in the attenuation analysis even though it is not, by itself, dispositive.

Despite my differences with the majority’s analy- sis, I agree with the outcome that it reaches. This is a close case. Weighing in favor of suppression are the presumption of taint arising from the absence of warnings, the close tem- poral proximity between the violation and the invitation to search, the fact that defendant was handcuffed and in cus- tody when he invited the deputy to search, and the absence of any indication that the violation was inadvertent. For various reasons, people in custody sometimes invite law enforcement officers to search their persons or belongings, even when they know that contraband is likely to be found. We would be naive to assume that experienced law enforce- ment officers do not understand that.

However, Article I, section 12, does not set an insur- mountable bar to a custodial invitation to search, even when the invitation would not have been made in the absence of a constitutional violation. I agree with the majority that the physical fruits of the search in this case did not derive from the Miranda violation. The detention here was brief; only one impermissible question was asked; that question was not asked coercively; and, in asking the impermissible question, 390 State v. Delong

the deputy did not seek consent to search. Although there is no evidence that defendant knew that he had the right to remain silent, nothing prevented him—apart from his self-accountable volition—from simply answering the depu- ty’s question and leaving it at that. He did not need to invite the deputy to search his vehicle. In the end, that factor tips the scales for me. For the foregoing reasons, I concur. WALTERS, J., dissenting. This case begins with a conceded violation of the Oregon Constitution and once again, ends without legal consequence. See State v. Unger, 356 Or 59, 103, 333 P3d 1009 (2014) (Walters, J., dissenting). In this case, a deputy stopped defendant for not wearing his seat belt, handcuffed him, searched him, and placed him in the backseat of a patrol car. The deputy then committed a blatant and con- ceded constitutional violation when he interrogated defen- dant about illegal activity without informing defendant that he had a constitutional right to remain silent. Although the government is precluded from obtaining “a criminal con- viction through the use of evidence obtained in violation of [constitutional] rights,” State v. Davis, 313 Or 246, 253, 834 P2d 1008 (1992) (citing State v. Davis, 295 Or 227, 666 P2d 802 (1983)), and an individual has a right to suppression of illegally obtained evidence to preserve the individual’s rights “to the same extent as if the government’s officers had stayed within the law,” id., the majority holds otherwise. I respectfully dissent. In Unger, this court considered the appropriate consequences when officers violate the constitution. The defendant in that case was in his own home, not in com- pelling circumstances or subject to interrogation, when he consented to search. The court held that the evidence that officers obtained as a result need not be suppressed. The court reasoned that the fact that the officers had violated Article I, section 9, by entering the defendant’s backyard did not affect the defendant’s decision to consent to search. Unger, 356 Or at 92. Cite as 357 Or 365 (2015) 391

Unlike Article I, section 9, of the Oregon Constitution, Article I, section 12, does not prohibit an officer from enter- ing private property without a warrant. Rather, it applies when an officer holds an individual in compelling circum- stances and prohibits the officer from conducting a criminal interrogation without first warning the detained individual that he or she has the right to remain silent and to consult a lawyer. Despite those differences, the question that is pre- sented when an officer violates Article I, section 12, is the same as the question that is presented when an officer vio- lates Article I, section 9: Did the evidence “derive from” the constitutional violation? See Unger, 356 Or at 80 (“[O]ur task is to determine whether police ‘exploited’ or ‘took advantage of’ or ‘traded on’ their unlawful conduct to obtain consent, or—examined from the perspective of the consent—whether the consent was ‘tainted’ because it was ‘derived from’ or was a ‘product of’ the unlawful conduct.”); State v. Vondehn, 348 Or 462, 475-76, 236 P3d 691 (2010) (court’s task is to determine whether evidence “derived from” an Article I, sec- tion 12, violation). I understand and agree with the majority and con- currence that the principles that underlie the exclusionary rule and its attenuation exceptions apply equally to viola- tions of both Article I, section 9, and Article I, section 12, and that the same factors may be relevant in deciding whether evidence that officers obtain can be used to convict a defendant. Delong, 357 Or at 378; id. at 386 (Brewer, J., concurring). However, one of those factors is the nature of the police misconduct. Unger, 356 Or at 81. It seems to me that an officer’s unlawful entry onto an individual’s prop- erty when the individual is not in compelling circumstances may have a different effect on the individual’s consent to search than does an officer’s failure to advise an individual held in compelling circumstances that the individual has a right to remain silent. In this case, for example, the deputy’s violation of Article I, section 12, had a more direct causative effect on defendant’s consent to search than did the violation of Article I, section 9, that the court considered in Unger. Here, the deputy had handcuffed defendant and placed him in a 392 State v. Delong

patrol car, when, without telling defendant that he had the right to remain silent and to consult a lawyer, the deputy asked him for incriminating evidence. Thus, in this case, unlike in Unger, defendant was held in compelling circum- stances and was entitled to information that the defendant in Unger was not entitled to receive. And, unlike in Unger, the majority concludes that defendant’s consent to search derived from the deputy’s constitutional violation. As the majority acknowledges, it was the deputy’s question and con- stitutional violation that prompted defendant’s unknowing response—the consent to search—and that response must be suppressed. Delong, 357 Or at 376. Thus, although the principles underlying the exclusionary rule are the same, its application in this case is different from its application in Unger and compels a different result. Had the majority adhered to the “totality of the cir- cumstances” analysis that it has used in past Article I, sec- tion 12, cases and to its reasoning in Unger, it would have suppressed not only defendant’s response to the deputy’s question, but also the physical evidence that the deputies obtained as a consequence of that response. Just five years ago, this court discussed the basis for suppression of evi- dence obtained in violation of Article I, section 12, and flatly rejected the state’s argument that the “mere failure to pro- vide Miranda warnings” requires only suppression of state- ments made in response to the unwarned questions and not suppression of the physical evidence obtained. Vondehn, 348 Or at 475-76. In Vondehn, the officer asked the defen- dant three unwarned questions: (1) Is this your backpack? (2) Does it contain marijuana? (3) Can I search it? The defen- dant answered the first two questions affirmatively and, in response to the third, voluntarily consented to the search of his backpack. Nevertheless, the court concluded that the trial court had been required to suppress the defendant’s answers to all three questions and the marijuana that the officer had obtained as a result of the defendant’s consent. Id. at 476-77. The court held that, “[w]hen the police vio- late 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 Cite as 357 Or 365 (2015) 393

criminal conviction,” including the “physical evidence that is derived from that constitutional violation.” Id. at 475-76. The majority does not overrule Vondehn, but distin- guishes it on its facts, as does the concurrence. The majority points out that “Robeson’s unwarned question in this case was open-ended; defendant’s direct response to the question was exculpatory; and he invited the deputies to search his car without an express request for consent.” Delong, 357 Or at 374. The concurrence says that, in this case, the detention was brief, only one question was asked and not coercively, and the deputy did not seek consent to search. Id. at 389-90 (Brewer, J., concurring). Those factual differences exist, but they are not of consequence. An unwarned question is an unwarned ques- tion, no matter how open ended. Defendant’s response to the deputy’s unwarned question was of a piece, and it was inculpatory. And defendant’s “invitation to search” was not any less prompted by the deputy’s question than it would have been if the deputy had asked defendant for consent. In Unger, the court took pains to explain that there is little to distinguish “unprompted or volunteered consent,” like that in State v. Kennedy, 290 Or 493, 624 P2d 99 (1981), and State v. Rodriguez, 317 Or 27, 854 P2d 399 (1993), from consent that is given in response to a request for consent. Both types of consents, the court reasoned in Unger, are “prompted by the officer’s question about drugs and guns.” Unger, 356 Or at 79. The more salient inquiry, the court held in Unger, is not whether the officer sought consent, but “whether the consent was ‘tainted’ because it was ‘derived from’ or was a ‘product of’ the unlawful conduct.” Id. at 80-81. I agree. The question at hand is: Did defendant’s response, which included a consent to search, derive from the constitutional violation? “Yes,” the majority says, “it did.” Delong, 357 Or at 376. The majority is correct. A response to an unwarned question that is prompted by, and results from, the question is a direct link in the causal chain, not independent of it. Suppression of the response—the consent to search—should also result in suppression of the evidence that is a product of that response. 394 State v. Delong

That does not mean that Article I, section 12, sets, as the concurrence would have it, “an insurmountable bar” to the admission of evidence obtained pursuant to a custo- dial consent to search. Id. at 389 (Brewer, J., concurring). In State v. Jarnagin, 351 Or 703, 713, 277 P3d 535 (2012), the court considered the kinds of facts that can attenuate a Miranda violation, and I do not oppose consideration of a response to unwarned questions as one factor in the alterna- tive analysis. But I do think it important to look at how we have applied the “totality of the circumstances” test in the past. In Jarnagin, officers questioned the defendant about injuries to an eight-month-old victim at the police station and later at the hospital where the victim had been taken. The officers did not give the defendant the required Miranda warnings, and the defendant described his role in the victim’s injuries. The defendant also told the offi- cers that he would reenact those events. The next day, the defendant participated in a video reenactment at his home. The circumstances were not compelling—the officers did not challenge or confront the defendant during the reenact- ment—and the officers again did not administer Miranda warnings. At trial, the court granted the defendant’s motion to suppress not only the defendant’s statements at the sta- tion and the hospital, but also the videotape. On review in this court, the state argued that the officers’ Miranda violations at the station and the hospital did not require suppression of the videotape. The court dis- agreed. The court acknowledged that a change in time and circumstances can be sufficient to dissipate the effects of an earlier Miranda violation and that the Miranda viola- tions at the station and the hospital “were not flagrant.” Id. at 717. The officers had not physically restrained the defen- dant and had advised him that he was not under arrest. Id. “[A]ccording to the trial court’s unchallenged ruling,” the court explained, “the officers [had] failed to recognize that the circumstances had become sufficiently compelling to require Miranda warnings.” Id. Nevertheless, the court required suppression of the videotape, reasoning that the defendant had reenacted the same events that he had described earlier Cite as 357 Or 365 (2015) 395

and “[n]o advice of Miranda rights had intervened to break the causal chain.” Id. at 718. Significantly, the court did not view the defendant’s voluntary agreement or his voluntary participation in the reenactment as breaking the causal chain. See id. at 719. Neither the majority nor the concurrence rest their conclusions on the types of facts recognized in Jarnagin as attenuating the taint of a Miranda violation—a change in time or circumstances, a lifting of restraints, an officer’s failure to recognize that the circumstances were so compel- ling that Miranda warnings were required, or the fact that belated Miranda warnings were given. Instead, the primary fact that the majority and concurrence deem essential is a fact that was of no consequence in Vondehn or Jarnagin and that pertains in virtually every instance in which a defendant seeks to exclude evidence obtained as a result of a “mere” Miranda violation—the fact that defendant’s response was volitional and the deputy did not coerce defen- dant’s response. An officer’s coercion can, of course, make a defen- dant’s response involuntary. See State v. Foster, 288 Or 649, 656, 607 P2d 173 (1980) (defendant’s waiver of right to counsel not voluntary where police persisted in repeated efforts to persuade defendant to waive right); see also State v. Mendacino, 288 Or 231, 238, 603 P2d 1376 (1979) (later confession inadmissible where “coercive conditions which resulted in * * * [earlier] confessions were not effectively removed”). But, as the court made clear in Unger, volun- tariness alone does not necessarily make evidence obtained in violation of the constitution admissible. 356 Or at 79. Miranda warnings are required in circumstances that are compelling but not coercive to ensure that a defendant speaks not out of compulsion but as a result of a knowing, deliberate choice. The purpose of Miranda warnings is to ensure that a defendant knows that he or she has a right to remain silent and to consult a lawyer. Without that information, a waiver of rights, even a voluntary waiver, is invalid. Jarnagin, 351 Or at 716; Vondehn, 348 Or at 476; see State v. Joslin, 332 Or 373, 386, 29 P3d 1112 (2001) (holding that defendant’s waiver of Article I, section 12, rights, “although voluntary, 396 State v. Delong

was not knowingly made and, therefore, was invalid”). Thus, as the majority acknowledges, the failure to give Miranda warnings demands suppression of derivative evidence even when the consent to search is voluntary. Vondehn, 348 Or at 476-77. In this case, even if defendant’s response to the deputy’s unwarned interrogation, including his consent to search, can be described as volitional, it must be sup- pressed because, considering the totality of the circum- stances, it was the product of the deputy’s constitutional violation. That conclusion also compels the conclusion that the resulting physical evidence must be suppressed. Both the consent and the physical evidence are the products of the same constitutional violation and, on these facts, defen- dant’s consent to search cannot serve to attenuate the taint of the Miranda violation and permit the admission of the physical evidence any more than did the voluntary, but unknowing, consents of the defendants in Vondehn and Jarnagin. The only additional peg on which the majority hangs its hat is its characterization of the Miranda viola- tion in this case as “not egregious.” Delong, 357 Or at 379. But why? This is not a case in which the trial court could make an unchallenged ruling—as did the trial court in Jarnagin—that the deputy failed to recognize that defen- dant was in compelling circumstances. Here, defendant was handcuffed when the interrogation began, and the deputy had to know that Miranda warnings were required. The majority does not contend otherwise, but neverthe- less terms the deputy’s Miranda violation “not egregious,” implying that there may be different degrees of Miranda violations. Although there may be degrees of coercion, once coercion reaches the level at which Miranda warnings are required, an officer must provide the information that the constitution requires. When Miranda warnings are required, no adjectival description can render an officer’s constitutional violation meaningless. Finally, returning to Unger, it is important to note that, in that case, the court cautioned that “consent that fol- lows a random stop or seizure that lacks probable cause or Cite as 357 Or 365 (2015) 397

reasonable suspicion that a crime has been committed and that is nothing more than a fishing expedition for incrimi- nating evidence” may be consent that is voluntary but never- theless so tainted that suppression is required. 356 Or at 91. Here, although the deputy had probable cause to believe that defendant had been driving without a seatbelt, he did not have probable cause to believe that he was in possession of drugs. Nevertheless, the deputy handcuffed defendant and placed him in the back seat of the police car and went on a fishing expedition—deliberately interrogating defendant and seeking incriminating evidence without first warning defendant of his right to remain silent and to consult a law- yer. In Unger terms, defendant’s invitation to search may have been voluntary, but it was nevertheless so tainted that suppression is required. I fear that the majority advances such police prac- tices when it refuses to impose consequences for a depu- ty’s constitutional violation. And, on the other side of the coin, holding law enforcement officers accountable will not result in the long-term loss of evidence. If the majority is correct that, in circumstances like those in this case, detained individuals give consent to search not because they lack an understanding of their rights, but because they make a deliberate decision to waive those rights, then such individuals will give consent to search even if we require law enforcement officers to adhere to constitutional requirements. In Unger, the court declined to impose a judicial con- sequence for the officers’ constitutional violation, expressing its expectations that officers will act within constitutional limitations. Id. at 94. Most law enforcement officers do act within constitutional limits and they do so most of the time. But when they do not, it benefits neither the officers nor our system of justice to excuse their violations by saying that they were “not egregious.” Our system of justice depends on the public’s respect for law enforcement. When officers do not obey the law, the public loses respect for law enforcement and the law—posing a danger to both. When we counsel, but do not demand, the best from law enforcement, we imperil both law enforcement and our system of justice. 398 State v. Delong

Judicially expressed expectations are not enough to secure liberty. “Liberty comes not from officials by grace but from the Constitution by right.” Maryland v. Wilson, 519 US 408, 424, 117 S Ct 882, 137 L Ed 2d 41 (1997) (Kennedy, J., dissenting). Because the majority retreats from principles necessary to give effect to those rights and to protect our system of laws and those who enforce them, I respectfully dissent. Baldwin, J., joins in this dissenting opinion. BALDWIN, J., dissenting. I respectfully disagree with the majority’s conclu- sion that defendant’s consent to the search of his vehicle attenuated the taint of the Miranda violation by the depu- ties in this case. In particular, I disagree with the majori- ty’s conclusion that defendant’s consent—given in response to police questioning while he was in custody and in hand- cuffs—was volunteered. I disagree with the reasoning that the majority has used to reach its conclusion, and I am con- cerned about how that reasoning may affect other cases in which citizens are interrogated by law enforcement officers under compelling circumstances in violation of their right to remain silent under Article I, section 12, of the Oregon Constitution.1 I. I begin with a brief additional background. Sergeant Robeson stopped defendant because defendant was not wearing his seat belt while operating his automobile. After Robeson activated his overhead lights and pulled defendant over, defendant identified himself by name and date of birth but did not provide Robeson with a driver’s license. Robeson handcuffed defendant, searched him, and placed him in the backseat of his patrol car. Robeson testified that his purpose in taking defendant into custody was to establish defendant’s identity. While defendant was handcuffed in Robeson’s patrol car, Robeson questioned him about illegal activity unrelated to the stop without first warning him that 1 Article I, section 12, of the Oregon Constitution provides, in part: “No person shall * * * be compelled in any criminal prosecution to testify against himself.” Cite as 357 Or 365 (2015) 399

he had a right to remain silent. Robeson asked defendant if there was anything in defendant’s car that the deputies should be concerned about. According to Robeson, defendant “told me ‘no,’ and that if we wanted to search the vehicle, we could.” Robeson testified that the above encounter lasted “maybe two or three minutes.” Deputy Poe searched defen- dant’s vehicle immediately after Robeson’s conversation with defendant and found a fanny pack that contained drug paraphernalia and drug residue. Poe did not give defendant Miranda warnings until after his search had disclosed the physical evidence of drugs. At the suppression hearing, defendant agreed that he had told Robeson that Robeson would not find anything in his car, but he denied that he had consented to the search of his car. The trial court found that “during this period of time * * * is when Sergeant Robeson initiated the conversation with the defendant about does he have anything of concern in his vehicle, and the defendant responded no, and then ultimately gave consent to search. Deputy Robeson said he did not ask for consent. That it was volunteered in a way by the defendant and the Court finds that appears to be credible.

“* * * * *

“[I]t was after the search when Deputy Poe approached the defendant, he gave him his Miranda rights and after giv- ing those Miranda rights did ask extensive questions about what he found in the car and statements—incriminating statements were made by the defendant. The Court is find- ing that the prior conversation about identification while he was detained to pursue that investigation, that at that point Miranda wasn’t needed.”

On appeal, the state conceded that the trial court had erred and that Miranda warnings were required before Robeson could question defendant because defendant had been in custody and under compelling circumstances at the time. The state nevertheless argued that suppression of the physical evidence was not required, because, in its view, defendant had made a spontaneous offer of consent to search his car. Defendant argued that State v. Vondehn, 348 400 State v. Delong

Or 462, 475-76, 236 P3d 691 (2010), supported his contention that his consent (as found by the trial court), other state- ments he had made, and the physical evidence all derived from the Article I, section 12, violation and should have been suppressed. The Court of Appeals concluded that this case “is largely governed by the principles and reasoning that the Supreme Court set forth in Vondehn” and reversed the trial court’s denial of defendant’s motion to suppress. State v. Delong, 260 Or App 718, 722-23, 320 P3d 653 (2014). In Vondehn, the defendant was likewise handcuffed, placed in a patrol car, and briefly interrogated without the benefit of Miranda warnings, in violation of his rights under Article I, section 12. In response to a deputy’s question, the defendant admitted that he owned the backpack in the car and that it contained marijuana. The defendant consented to a search of the backpack, marijuana was found, and the officers then gave the defendant his Miranda warnings. This court observed that it had long held “that the Oregon Constitution requires suppression of statements made without the benefit of Miranda warnings.” Vondehn, 348 Or at 472. The court held that “the state is precluded from using evidence derived from the violation to obtain a criminal conviction,” including “physical evidence that is derived from that constitutional violation.” Id. at 475-76. This court also explained in Vondehn that the ratio- nale for the constitutional requirement that police warn cit- izens of their right against self-incrimination is the level of coercion inherent in custodial interrogations. Article I, sec- tion 12, protects citizens against the use of compelled state- ments, because such statements do not provide an acceptable basis for proving guilt of a crime in a civilized society. See State v. Mendacino, 288 Or 231, 236, 603 P2d 1376 (1979) (so stating). To give effect to that constitutional right, this court has prohibited the state from using physical evidence derived from an Article I, section 12, violation to prosecute a suspect. The principles of law that this court applied in Vondehn are highly pertinent to this case: “Since Magee, this court consistently has held that the Oregon Constitution requires suppression of statements Cite as 357 Or 365 (2015) 401

made without the benefit of Miranda warnings. See, e.g., State v. Roble-Baker, 340 Or 631, 643-44, 136 P3d 22 (2006) (suppressing unwarned statements made during custodial interrogation); State v. Smith, 310 Or 1, 7, 791 P2d 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, 386, 29 P3d 1112 (2001)); see also State v. Meade, 327 Or 335, 339, 963 P2d 656 (1998). * * * “* * * * * “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 interroga- tion that he or she has a right to remain silent and to con- sult with counsel and that any statements that the per- son 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 inter- rogation without first obtaining a knowing and voluntary waiver of the suspect’s rights, then they violate the sus- pect’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.”

348 Or at 472-74. In applying those principles to the facts in Vondehn, this court concluded: “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 con- ducted 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 402 State v. Delong

requires suppression of both the answers that defendant gave in response to, and the marijuana that the police iden- tified and seized as a result of, that interrogation.”

Id. at 476. II. With that background in mind, I now explain my disagreement with the majority’s conclusion that defen- dant’s consent to the search of his vehicle attenuated the taint of Robeson’s illegal questioning. The majority begins its attenuation analysis by “not[ing] that the amount of time that passed between the Miranda violation and the discov- ery of the physical evidence was brief. Additionally, defen- dant remained in custody during the encounter. In those respects, this case is similar to Vondehn.” 357 Or at 374. The majority does not, however, discuss the compelling nature of the circumstances confronting defendant or give those cir- cumstances any weight in its attenuation analysis. Not only did defendant “remain in custody” after he was stopped for a seat belt violation, he was searched, handcuffed, placed in the back of a patrol car, and then basically asked by Robeson if he had engaged in any illegal activity.2 The majority does not consider the effect of those circumstances on defen- dant at the time that he responded to Robeson’s unwarned question. The majority instead focuses on what it sees as “the primary factual difference” between Vondehn and this case—“defendant’s invitation to search his car.” 357 Or at 374. “When Sergeant Robeson asked defendant ‘if there was anything we should be concerned about’ in his car, defendant ‘told [him] “no,” and that if [the deputies] wanted to search 2 The majority acknowledges that Robeson’s question “went too far” and states that his question “was not limited to officer safety.” 357 Or at 379, 379 n 14. Because defendant had already been searched and handcuffed at the time that Robeson questioned him, however, officer safety would not appear to be an issue at all. Rather, based on the circumstances and the substance of Robeson’s question, it appears that Robeson’s sole purpose in asking the question was to elicit incriminating information from defendant. See Rhode Island v. Innis, 446 US 291, 300-01, 100 S Ct 1682, 64 L Ed 2d 297 (1980) (holding that “the Miranda safeguards come into play whenever a person in custody is subjected to * * * any words or actions on the part of the police * * * that the police should know are reasonably likely to elicit an incriminating response from the suspect”). Cite as 357 Or 365 (2015) 403

the vehicle [they] could.’ ” Id. Although defendant’s response to the unwarned question was immediate and directly per- tained to the deputy’s question, the majority sharply breaks the response into two parts, giving great weight to what it characterizes as defendant’s “invitation” to search his car. Notwithstanding the compelling circumstances confronting defendant, the majority views the “invitation” to search as spontaneous and in no way causally related to those circum- stances or to the deputy’s illegal conduct. The majority con- cludes that defendant’s consent was “volunteered.” 357 Or at 375.3 Thus, the majority ultimately views defendant’s con- sent as representing a complete break in the causal chain between the taint of the Miranda violation and the physical evidence obtained from the consent search. In my view, the majority’s conclusion that defen- dant’s consent to search was volunteered is inconsistent with the weight of pertinent case law. As the majority notes, Professor LaFave identifies two types of statements that properly may be considered “volunteered” in the Miranda context: (1) statements that are not prompted by police ques- tioning, and (2) statements that are nonresponsive to a police officer’s question. 357 Or at 375 (citing Wayne R. LaFave et al, 2 Criminal Procedure § 6.7(d) (3d ed 2007 and 2014 supp) (discussing volunteered statements)). The majority appears to implicitly concede that defendant’s statements did not fall into the category of unprompted statements. Rather, the majority posits that defendant’s statement that the dep- uties could search his car was nonresponsive because his answer to Robeson’s question “went beyond what Robeson had asked and included an offer for the officers to search his car.” 357 Or at 376. The cases that Professor LaFave cites for the prop- osition that a nonresponsive statement may be considered volunteered, however, do not support the majority’s position. 3 I have great difficulty viewing a consent to search given by a suspect in custody, in handcuffs, and in response to police questioning as a “volunteered” act. A “volunteer” is a “voluntary actor or agent in a transaction; * * * [s]omeone who gratuitously and freely confers a benefit on another.” Black’s Law Dictionary 1807 (10th ed 2014). For me, to characterize defendant’s consent, if we are to call it that, as “volunteered” under the compelling circumstances confronting defen- dant strains the meaning of that word to a breaking point. 404 State v. Delong

Those cases primarily involved suspects who had made incriminating statements that were wholly unrelated to the questions asked by the police or that were not a product of unwarned interrogation. See, e.g., United States v. Crisolis- Gonzalez, 742 F3d 830, 836-37 (8th Cir 2014) (the defen- dant’s statement that he had a gun under his mattress was volunteered, because that statement was “wholly unrelated” to law enforcement agent’s inquiry into his immigration sta- tus); United States v. Woods, 711 F3d 737, 741 (6th Cir 2013) (the defendant volunteered the “unexpected and unrespon- sive reply” that he had a weapon in his car when asked by officer what object was in his pocket); United States v. Fleck, 413 F3d 883, 893 (8th Cir 2005) (when the officer asked the defendants “how they liked the food” in the county jail, the officer’s question was not “calculated to elicit an incrim- inating response” from the defendants); United States v. Castro, 723 F2d 1527, 1530-31 (11th Cir 1984) (when law enforcement agent asked the defendant, “What in the world is going on here?”, the defendant’s subsequent offer to bribe the agent was a “spontaneously volunteered” statement that was “totally unresponsive” to the agent’s question). In contrast to the above cases, the state has con- ceded that Robeson’s question whether “there was anything [the deputies] should be concerned about” in defendant’s car constituted a Miranda violation. Moreover, defendant’s statements that nothing in his car should concern the depu- ties and that they could search his car were made in direct response to that unwarned question. Accordingly, because defendant’s statements were both prompted by and made in response to Robeson’s question, I would conclude that defen- dant’s consent to search in this case may not properly be considered “volunteered.” In determining that defendant’s consent attenuated the taint of the Miranda violation, the majority relies on State v. Unger, 356 Or 59, 333 P3d 1009 (2014), a case recently decided by this court in which police officers trespassed into the defendant’s backyard in violation of his privacy rights protected by Article I, section 9, of the Oregon Constitution. In that case, the officers knocked on the back door of the defendant’s residence, and the defendant consented to the entry into and the search of his residence. The defendant Cite as 357 Or 365 (2015) 405

was not physically restrained by handcuffs or otherwise subjected to compelling circumstances. Thus, no Miranda warnings were required or at issue in Unger. The majority cites Unger as “reaffirm[ing] that, when a defendant’s con- sent to a search either was not affected or was only tenuously connected to a prior illegality, the defendant’s voluntary con- sent can be sufficient to break the causal chain [between the illegality and the evidence obtained in the consent search].” 357 Or at 378. The majority does not explain, however, why the attenuation analysis in Unger—an Article I, section 9, case—should apply in an Article I, section 12, case in which a suspect was not given his Miranda warnings before he consented to a search under compelling circumstances. Moreover, in applying Unger, the majority again disregards the effect of the compelling circumstances confronting defen- dant as a causal factor in producing defendant’s consent and the physical evidence obtained in the consent search. In my view, Unger does not provide a readymade answer to how this court should conduct an attenuation analysis under the circumstances of this case. In relying on Unger, the majority applies three fac- tors identified in that case that it thinks bear on when a consent to search will attenuate the illegality of a Miranda violation: “That legal determination—whether, in the circumstances of a particular case, consent has so attenuated the con- nection between the prior illegal conduct and the evidence obtained in the consent search—requires a court to con- sider the illegal conduct that comprised the stop or search, the character of the consent, and the causal relationship between the two.” Unger, 356 Or at 78. First, the majority considers the illegal conduct of law enforcement. Again, no consideration is given to the compelling circumstances confronting defendant when he consented to the search. Instead, the majority observes that the violation of defendant’s Miranda rights “can hardly be characterized as egregious * * *. Robeson did not engage in repeated efforts to wear down [defendant’s] resistance. * * * [O]ther than the initial background questions he asked, Robeson asked defendant only one question[.]” 357 Or at 378-79 (internal quotation marks omitted). Of course, 406 State v. Delong

there was no reason for Robeson to ask defendant repeated or additional questions. Under compelling circumstances, defendant provided incriminating information to the dep- uties based on a single question. From Robeson’s question and the circumstances, it is apparent that the purpose of the question was to elicit potentially incriminating infor- mation from defendant. Here, that purpose was achieved by Robeson asking a single question. Second, the majority considers “the character of defendant’s consent.” 357 Or at 379. Again, without any refer- ence to the compelling circumstances confronting defendant, the majority concludes that, in response to an accusation by a sheriff’s deputy who had searched and handcuffed him, defendant simply “invited the deputies to search his car if they wanted to do so. Defendant’s invitation to search his car in this case is virtually identical to the invitations in Kennedy and Rodriguez, which this court held attenuated the taint of the unlawful seizures in those cases.” 357 Or at 379 (citing State v. Kennedy, 290 Or 493, 624 P2d 99 (1981), and State v. Rodriguez, 317 Or 27, 854 P2d 399 (1993)). However, as with its reliance on Unger, the majority fails to recognize critical differences between Kennedy and Rodriguez and this case. In Kennedy, the defendant was not confronting compelling circumstances when he consented to a search of his luggage after police officers had stopped him at an airport without reasonable suspicion. The defendant was not in custody or improperly questioned as part of any Miranda violation. In Rodriguez, the defendant consented to a search of his apart- ment after he “had been read his Miranda rights and stated that he understood them. He was under no compulsion to answer the agent’s question.” 317 Or at 41 n 15. Both cases involved alleged violations of the defendants’ privacy rights under Article I, section 9—not a defendant’s right to remain silent protected by Article I, section 12. Finally, the majority purports to consider “the causal connection between the violation and defendant’s invitation.” 357 Or at 380. As previously noted, defendant’s response to Robeson’s question immediately followed that question. The conversation was brief, with no intervening circumstances occurring between the question and the response. Notwithstanding the direct and obvious causal Cite as 357 Or 365 (2015) 407

connection between the question and the consent, the major- ity observes that “[t]his is not a case in which Robeson’s unwarned questioning left ‘little, if anything, of incrimi- nating potential * * * unsaid.’ ” 357 Or at 380 (quoting State v. Jarnagin, 351 Or 703, 722, 277 P3d 535 (2012) (internal quotation marks omitted)). However, the quoted portion of Jarnagin pertains to the efficacy of belated Miranda warn- ings. The quoted discussion from Jarnagin appears to have little bearing on the causal connection between the Miranda violation and defendant’s consent to search in this case.4 In short, I do not find the majority’s attenuation analysis in this case persuasive. As noted, pertinent case law does not support the majority’s conclusion that defen- dant’s consent was volunteered. Moreover, the majority does not recognize that the compelling circumstances to which defendant was subjected bears on whether defendant’s con- sent attenuates the police illegality in this case. Although the majority purports to consider the causal connection between the illegality and defendant’s consent, it declines to actually look at those compelling circumstances as a causal factor. In my view, that oversight represents a major flaw in the majority’s attenuation analysis. The majority then com- pounds that error by giving inordinate weight to defendant’s consent as a factor that—by itself, and in a highly fictional- ized manner5 —attenuates the taint of the police illegality. 4 The complete quote in Jarnagin is as follows: “With that background in mind, we turn to the facts of this case. We note, as an initial matter, that this is not a case, as in [Missouri v. Seibert, 542 US 600, 124 S Ct 2601, 159 L Ed 2d 643 (2004)], where the unwarned interroga- tion left ‘little, if anything, of incriminating potential * * * unsaid,’ making it ‘unnatural’ not to ‘repeat at the second stage [of the interrogation] what had been said before.’ See Seibert, 542 US at 616-17 (plurality opinion).” 351 Or at 722. 5 I understand that legal fictions are commonly used in the analysis of legal principles and in their application to particular facts. See Louise Harmon, Falling Off the Vine: Legal Fictions and the Doctrine of Substituted Judgment, 100 Yale LJ 1, 2-16 (1990) (discussing historical debate on use of legal fictions). Professor Lon Fuller defined a legal fiction as “either (1) a statement propounded with a complete or partial consciousness of its falsity, or (2) a false statement recognized as having utility.” Lon L. Fuller, Legal Fictions 9 (1967). Fuller distinguished a fiction from a lie “by the fact that it is not intended to deceive.” Id. at 6. He distin- guished a fiction from an erroneous conclusion “by the fact that it is adopted by its author with knowledge of its falsity.” Id. at 7. Thus, for Fuller, a legal fiction was problematic only if it was used without recognition of its falsity: “In practice, it is precisely those false statements that are realized as being false that have utility. 408 State v. Delong

III. As noted in Vondehn, 348 Or at 476 n 8, other state courts have decided under their state constitutions that physical evidence obtained in violation of Miranda rights must be excluded at trial as “fruit of the poisonous tree.” See, e.g., State v. Peterson, 181 Vt 436, 446-47, 923 A2d 585 (2007) (holding, under Vermont Constitution, that “[p]hysi- cal evidence gained from statements obtained under circum- stances that violate Miranda is inadmissible in criminal proceedings as fruit of the poisonous tree”); Commonwealth v. Martin, 444 Mass 213, 215, 827 NE2d 198 (2005) (adopt- ing common-law rule under Massachusetts Constitution that physical evidence, “if derived from unwarned state- ments where Miranda warnings would have been required by Federal law in order for them to be admissible, is pre- sumptively excludable from evidence at trial as ‘fruit’ of the improper failure to provide such warnings”); State v. Knapp, 285 Wis 2d 86, 123, 700 NW2d 899 (2005) (noting that “the goals of the exclusionary rule and fruit of the poisonous tree doctrines are to curb ‘illegal governmental activity,’ ” and concluding that “it is appropriate that the exclusion- ary rule bars physical fruits obtained from a deliberate Miranda violation under Article I, Section 8” of Wisconsin Constitution); but see State v. Sole, 185 Vt 504, 514, 974 A2d 587 (2009) (holding that physical evidence not tainted by prior Miranda violation where the defendant consented to search in response to officer’s request because “a consent request is not designed to elicit an incriminating response” (internal quotation marks omitted)). The physical evidence that defendant has challenged in this case is derivative in nature, because it was obtained A fiction taken seriously, e.g., ‘believed,’ becomes dangerous and loses its utility. It ceases to be a fiction under either alternative of the definition given above.” Id. at 9-10. Other commentators have agreed with Fuller that the dangerousness of a fiction derives from the failure to acknowledge its falsity and have some- times criticized the fiction of “consent” in criminal procedure on that basis. See, e.g., David A. Sklansky, Traffic Stops, Minority Motorists, and the Future of the Fourth Amendment, 1997 Sup Ct Rev 271, 322 (1997) (“Unfortunately, the fiction of consent in criminal procedure is used by the Supreme Court with something far short of ‘a complete consciousness of its falsity.’ ”) (Quoting Fuller.). In my view, using a legal fiction without recognizing its falsity is particularly danger- ous when the scope of protection afforded a constitutional right is determined based on the use of that fiction. Cite as 357 Or 365 (2015) 409

as the result of an unwarned question directed by Robeson to defendant when defendant was in custody. Defendant is therefore entitled to a determination as to whether that derivative evidence is “tainted” by the constitutional viola- tion or, as famously stated by Justice Frankfurter, whether the evidence is the “fruit of the poisonous tree.” Nardone v. United States, 308 US 338, 341, 60 S Ct 266, 84 L Ed 307 (1939). As pointed out by Professor LaFave, Nardone “established the doctrine of ‘attenuation’ by authoritatively recognizing that the challenged evidence might sometimes be admissible even if it did not have an ‘independent source’ because the ‘causal connection * * * may have become so attenuated as to dissipate the taint.” LaFave, 3 Criminal Procedure § 9.3(a) at 419-20; see also Wong Sun v. United States, 371 US 471, 83 S Ct 407, 9 L Ed 2d 441 (1963) (sus- pect’s confession to crime untainted by his illegal arrest the day before, because it was given after his release and his voluntary return to police station). In this case, there was an immediate, direct causal relationship between the compelling circumstances con- fronting defendant and the consent to search that defen- dant gave in response to Robeson’s unwarned question. As observed in Vondehn, this court has long recognized the inherent level of coercion that exists when suspects in custody are questioned by police officers. Vondehn, 348 Or at 472; see Joslin, 332 Or at 380 (noting that protection against compelled self-incrimination under Article I, sec- tion 12, “extends to custodial interrogations, because of the inherent level of coercion that exists in such interroga- tions”); Meade, 327 Or at 339 (court “has recognized that a level of coercion is inherent in any custodial setting”); State v. Brewton, 247 Or 241, 244, 422 P2d 581, cert den, 387 US 943 (1967) (recognizing “the inherently coercive char- acter of police interrogation of a suspect in custody who has not been advised of his rights”). Indeed, the United States Supreme Court identified the compulsion inher- ent in custodial interrogations as the primary reason for requiring police officers to warn suspects of their rights. See Miranda v. Arizona, 384 US 436, 467, 86 S Ct 1602, 16 L Ed 2d 694 (1966) (concluding that, without providing such warnings, “the process of in-custody interrogation of 410 State v. Delong

persons suspected or accused of crime contains inherently compelling pressures which work to undermine the indi- vidual’s will to resist and to compel him to speak where he would not otherwise do so freely”). In recent years, several courts have recognized the compelling effect of a police officer’s presence on an indi- vidual’s consent and have considered empirical studies in fashioning appropriate tests for determining the validity of consent searches. For example, the Supreme Court of New Jersey cited various psychological studies regarding the compulsion inherent in police-citizen encounters in State v. Carty, 170 NJ 632, 790 A2d 903 (2002). The court noted, “In the context of motor vehicle stops, where the individual is at the side of the road and confronted by a uniformed officer seeking to search his or her vehicle, it is not a stretch of the imagination to assume that the individual feels compelled to consent.” Id. at 644 (citing psychological studies that have shown that “there is an almost reflexive impulse to obey an authority figure”). The court also cited data from the New Jersey State Police Independent Monitors’ reports that indi- cated that nearly 95 percent of detained motorists granted a law enforcement officer’s request for consent to search. Id. at 644-45. Based in part on that social science, the court altered its test for determining the validity of a motorist’s consent to search, holding that law enforcement personnel must have a reasonable and articulable suspicion of crimi- nal wrongdoing, beyond the initial valid motor vehicle stop, before seeking consent to search. Id. at 647. Similarly, in Brown v. State, 182 P3d 624 (Alaska Ct App 2008), the Court of Appeals of Alaska relied on var- ious studies on the inherently compelling nature of police- citizen interactions to conclude that federal law does not adequately protect motorists. The court noted that consent searches are nearly always held to be valid under the Fourth Amendment: “The federal law in this area is premised on the assumption that, all things being equal, a motorist who does not wish to be subjected to a search will refuse consent when the officer seeks permission to conduct a search. But experience has shown that this assumption is wrong.” Id. at 630. After citing various studies that have undercut the Cite as 357 Or 365 (2015) 411

notion that a person truly consents to be searched, the court concluded: “Whatever the exact reasons for motorists’ willingness to accede to the requests of law enforcement officers, it is clear that large numbers of motorists are consenting to be searched each year—indeed, each month, and each week. Motorists are giving consent in such large numbers that it is no longer reasonable to believe that they are making the kind of independent decision that lawyers and judges typically have in mind when they use the phrase ‘consent search.’ ” Id. at 630-31. Ultimately, the court held, under the search- and-seizure provision of the Alaska Constitution, that an offi- cer who had stopped the defendant for a vehicle-equipment violation was prohibited from requesting the defendant’s consent to search her person and vehicle for drugs. Id. at 634. Several other courts have acknowledged the wide- spread academic criticism of case law pertaining to con- sent searches. The Supreme Court of Kansas, for example, observed, “Commentators, in addition to noting the difficulty in applying the case law relating to consensual searches to specific fact situations, argue that the [United States Supreme] Court’s analysis utilizes an ill-crafted paradigm [for interpreting and applying the Fourth Amendment].” State v. Thompson, 284 Kan 763, 777-79, 166 P3d 1015 (2007), as modified (Oct 17, 2007) (noting numerous scholars’ cri- tiques that consent searches after routine traffic stops are inherently coercive). Similarly, the Supreme Court of Iowa described the abundant academic commentary on consent searches pursuant to traffic stops and acknowledged that a common criticism with the consensual search doctrine is that “a traffic stop gives rise to an element of compulsion.” State v. Pals, 805 NW2d 767, 780-82 (Iowa 2011). In particular, scholars have criticized the wide gap between the fiction that ordinary citizens consent to a police officer’s requests and the reality that police-citizen encoun- ters involve such inherently compelling circumstances as to vitiate any true choice on the part of the citizen. According to a study performed in Maryland and Ohio, which examined 412 State v. Delong

motorists’ compliance with police requests for consent to search their vehicles, approximately 90 percent of the motorists studied consented to have their vehicles searched. Daniel J. Steinbock, The Wrong Line Between Freedom and Restraint: The Unreality, Obscurity, and Incivility of the Fourth Amendment Consensual Encounter Doctrine, 38 San Diego L Rev 507, 534-35 (2001). Similar studies on police-citizen encounters have shown that “people tend to underestimate the strength of situational constraints and overestimate the voluntariness of others[’] actions.” Josephine Ross, Can Social Science Defeat a Legal Fiction? Challenging Unlawful Stops Under the Fourth Amendment, 18 Wash & Lee J Civil Rts & Soc Just 315, 332 (2012) (advocating for defeat of legal fiction of consensual nature of police-citizen encounters through use of social science). Indeed, one commentator has concluded, “The truth is that people consent so often that it undermines both the mean- ingfulness of the consent and the believability that the police are really respecting the doctrine.” Oren Bar-Gill & Barry Friedman, Taking Warrants Seriously, 106 Nw U L Rev 1609, 1662 (2012).

In State v. Jenkins, 298 Conn 209, 3 A3d 806 (2010) (Palmer, J., dissenting), Justice Palmer recently surveyed much of the academic commentary and social science regard- ing the coercive effect that an officer’s request for consent is likely to have on a motorist who has been detained in con- nection with a traffic stop. Id. at 325-34. He summarized that literature as follows:

“[E]mpirical studies over the last several decades on the social psychology of compliance, conformity, social influ- ence, and politeness have all converged on a single conclu- sion: the extent to which people feel free to refuse to comply is extremely limited under situationally induced pressures. * * * It therefore has been argued that the United States Supreme Court should incorporate the empirical find- ings on compliance and social influence into * * * consent [search] jurisprudence * * * to dispel the air of unreality that characterizes the current doctrine.”

Id. at 326 (internal quotation marks omitted). Cite as 357 Or 365 (2015) 413

The foregoing authorities have all recognized that compulsion is an inherent feature of a police encounter when a motorist is detained in connection with a routine traffic stop. In this case, we have the additional circumstances that defendant was personally searched, handcuffed, and placed in the back of a patrol car; that strong show of authority subjected defendant to more compulsion than is ordinarily inherent in a routine traffic stop. Based on its prior prece- dents, this court should give those compelling circumstances appropriate weight in considering the causal connection between Robeson’s Miranda violation and defendant’s con- sent to the search of his car. Vondehn, 348 Or at 472; Joslin, 332 Or at 380; Meade, 327 Or at 339; Brewton, 247 Or at 244.

This court has generally looked to the totality of the circumstances and applied a fact-intensive inquiry to deter- mine whether physical or testimonial evidence derives from or is a product of a Miranda violation. Jarnagin, 351 Or at 716-17; see Vondehn, 348 Or at 482 (considering “all relevant circumstances” in deciding whether belated Miranda warn- ings were effective in ensuring valid waiver of rights). Here, Robeson illegally asked defendant an unwarned question in violation of Article I, section 12, for the purpose of elicit- ing incriminating information from defendant. Defendant’s direct and immediate response to that question, given under compelling circumstances, included defendant’s consent to search his car. Those circumstances involved a strong show of police authority, including personally searching defendant, handcuffing him, and placing him in the back of Robeson’s patrol car. There were no intervening events between Robeson’s unwarned question and defendant’s con- sent. Robeson exploited that unwarned question to obtain incriminating information from defendant. And that infor- mation, in turn, was offered by the state to convict defen- dant of possession of a controlled substance. Under those circumstances, defendant’s consent cannot properly be viewed as a complete break in the causal chain between the Miranda violation and the physical evidence obtained by the police. Put differently, defendant’s consent was more than tenuously related to the Miranda violation. Jarnagin, 351 Or at 716-17; see also State v. Ayles, 348 Or 622, 636-39, 414 State v. Delong

237 P3d 805 (2010) (Miranda warnings alone not sufficient to “ensure that the unlawful police conduct did not affect, or had only a tenuous connection to, [the] defendant’s responses”); LaFave, 3 Criminal Procedure § 9.3(c) at 423-24 (discussing relevant criteria for determining “when there is only an ‘attenuated connection’ between a violation and cer- tain derivative evidence”); Comment, Fruit of the Poisonous Tree—A Plea for Relevant Criteria, 115 U Pa L Rev 1136, 1148-49 (1967) (source of “relevant criteria” relied upon by Professor LaFave). For the foregoing reasons, I would hold that defen- dant’s consent to the search of his car did not attenuate the taint of the Miranda violation by Robeson. Where, as here, a suspect consents to the search of his car under compel- ling circumstances and in direct response to an unwarned question by a law enforcement officer seeking to elicit incriminating information from that suspect, the physical evidence obtained from that search must be excluded as the “fruit of the poisonous tree” to give effect to the consti- tutional protection against self-incrimination provided for by Article I, section 12, of the Oregon Constitution. In my view, the majority’s contrary conclusion unduly diminishes the vital constitutional protection against self-incrimination provided for by Article I, section 12. I therefore respectfully dissent. Walters, J., joins this dissenting opinion.