State v. Garrett

555 P.3d 1116

• Court: Supreme Court of Kansas
• Decided: September 20, 2024
• Docket: 124329
• Status: Published

Opinion

IN THE SUPREME COURT OF THE STATE OF KANSAS

No. 124,329

STATE OF KANSAS, Appellant,

v.

PHILLIP JASON GARRETT, Appellee.

SYLLABUS BY THE COURT

1. The protections of the Fifth Amendment to the United States Constitution, made applicable to the states through the Due Process Clause of the Fourteenth Amendment, prohibit the State from relying on coerced or involuntary statements to establish a defendant's guilt. But these protections do not justify evidentiary suppression of confessions that are either unrelated to law enforcement tactics, or are connected to, but not causally related to, law enforcement tactics that constitute misconduct.

2. When determining whether a confession was obtained in violation of due process, a reviewing court must first consider the totality of the circumstances to determine whether any related law enforcement tactics constituted misconduct. If such law enforcement tactics do not constitute misconduct, a resulting confession cannot be rendered inadmissible because of those tactics.

3. If law enforcement committed misconduct related to a confession, a reviewing court must then assess whether, under the totality of the circumstances, the misconduct

1 caused the confession. In other words, the court must consider whether the misconduct caused the defendant's free will to be overborne, such that the resulting confession was not voluntary. If that happened, law enforcement has violated due process and the resulting confession must be suppressed.

Review of the judgment of the Court of Appeals in an unpublished opinion filed October 21, 2022. Appeal from Saline District Court; JARED B. JOHNSON, judge. Oral argument held May 15, 2023. Opinion filed September 20, 2024. Judgment of the Court of Appeals reversing the district court is affirmed. Judgment of the district court is reversed and remanded.

Kristafer R. Ailslieger, deputy solicitor general, argued the cause, and Derek Schmidt, attorney general, was with him on the briefs for appellant.

D. Justin Bravi, of Salina Regional Public Defender Office, argued the cause, and Mark J. Dinkel, of the same office, was with him on the briefs for appellee.

The opinion of the court was delivered by

WILSON, J.: Police interviewed Phillip Jason Garrett after he was accused of inappropriately touching a minor, L.A. Garrett confessed to some of the allegations during the interview. The district court suppressed his statements after concluding they were involuntary. A panel of the Court of Appeals reversed, holding the district court placed undue weight on the deceptive police practices while excluding "nearly all other relevant components of the inquiry." State v. Garrett, No. 124,329, 2022 WL 12129643, at *6 (Kan. App. 2022) (unpublished opinion). Garrett petitioned for review. We affirm the Court of Appeals' reversal of the district court.

2 FACTUAL AND PROCEDURAL BACKGROUND

In November 2018, L.A.'s biological father reported to Salina police that he found text messages on L.A.'s phone showing Phillip Garrett inappropriately touched her. L.A. was under 14 years-of-age at the time. Police interviewed L.A., and she confirmed Garrett had touched and penetrated her vagina and anus multiple times and rubbed her chest. Officers contacted Garrett and requested he come to the police station, and Garrett agreed.

Detective Tim Brown and Detective Gregory Jones interviewed Garrett in a locked room inside the police headquarters. The interview began with Detective Jones telling Garrett that Jones needed to "jump through some hoops" because the interview was taking place behind locked doors and Jones was a "cop and I ask questions." Jones then read Garrett his Miranda rights. The officers asked Garrett about the allegations, and Garrett denied them.

The officers then asked Garrett if he would submit to a computerized voice stress analysis (CVSA) to verify the truth of his statements. Garrett was hesitant, telling the officers he was very nervous and stressed and worried the stress would negatively impact the results of the test. The detectives offered to bring in Sergeant Sarah Cox, who administers the tests, to better explain the test and allay his fears. While they waited for Sergeant Cox, Detective Brown told Garrett the CVSA is more accurate than a polygraph.

Sergeant Cox entered the interview room and described the test to Garrett. She told him, "They're just a series of yes or no questions. If you're telling us the truth, then you should have no problem, okay? But if you're lying about these specific questions, the stress is going to pop up on those charts like nobody's business and we're gonna know. It is 100% effective." When Detective Brown let Sergeant Cox know Garrett was worried

3 his general stress was going to alter the results of the test, Cox told him "this test is 100% effective and what I mean by that [is] it doesn't matter if someone is drunk or high or sober, it's still going to measure that frequency and that stress in your vocal cords." Detective Brown told Garrett they appreciated his cooperation so far and thought he would want to continue to cooperate. Garrett eventually agreed to submit to the test.

Sergeant Cox led Garrett into a separate room to complete the CVSA. Before they began, Sergeant Cox handed Garrett a form to read and sign titled "Truth Verification Release Form." The form stated that Garrett was submitting to the test without any "threat, coercion, promise, reward or immunity," and that Garrett released all involved parties from any liability associated with the exam. The form also stated that Garrett understood all materials and recordings from the exam could be released for the purposes of testimony. The bottom of the form included Garrett's Miranda rights. Sergeant Cox read each right aloud to Garrett and had him initial beside each one.

After Garrett signed the form, Sergeant Cox offered more details about the exam:

"[T]he CVSA is actually a tool that's used all over the United States, even the military uses it, to verify whether someone is telling the truth. Okay? Instead of being called a lie detector test, the CVSA is considered a truth verification exam, and it is 100% effective. Okay? The CVSA works by analyzing the stress of one's voice when asked specific questions to determine whether the person being asked those questions is telling the truth or a lie. The test activates off of voice frequency alone so again it doesn't matter how high, drunk, sober a person is, it's going to be 100% accurate."

Sergeant Cox said, "by the time we're done in here, we're going to know what happened and what the truth is." She then administered the test and returned Garrett to the interview room.

4 After that, Sergeant Cox analyzed the results and concluded "there was stress present, and stress is an indication of deception." She then contacted another CVSA examiner to analyze the results, which is standard protocol. The two examiners then discussed the results and their separate analyses. Their discussion did not cause Cox to change her opinion about the test results.

The officers told Garrett the test had registered stress when Sergeant Cox asked if Garrett had touched L.A.'s anus. Garrett still denied the allegations. Detective Brown said he could tell Garrett loved L.A., and that his love had caused him to make a bad decision. The officers said they wanted to tell the prosecutor Garrett had been cooperative. Garrett eventually confessed to rubbing L.A.'s vagina four or five times.

The State charged Garrett with 11 counts of rape, 8 counts of aggravated indecent liberties with a child, 1 count of aggravated indecent solicitation of a child, and 2 counts of aggravated criminal sodomy.

Garrett moved to suppress the statements he made during the interrogation, arguing they had been coerced. The district court held a hearing on the motion, during which Detectives Jones and Brown and Sergeant Cox testified. The court admitted the recordings of the interview and the CVSA results. Gary Davis, a defense expert, testified about the CVSA and its accuracy. He stated that while he was not familiar with how the test is administered, the CVSA is a real test that has been used in official settings. But Davis also testified the test was "[n]o better than flipping a coin" to detect truthfulness. And he agreed literature shows it to be only 15 to 50 percent accurate in detecting truthfulness and that the CVSA cannot discriminate general stress from "case-specific" stress.

5 The district court initially concluded that Garrett's statement had been voluntary and denied the motion to suppress. We will discuss the district court's findings in more detail below.

Eighteen months later, before trial, the district court reversed its own judgment sua sponte. The court reconsidered the totality of the circumstances and—based largely on the officers' deceptive description of the accuracy of the CVSA and the postexam interview tactics—concluded Garrett's statements were involuntary. The court then suppressed Garrett's statements.

The State filed an interlocutory appeal, and the Court of Appeals reversed. Garrett, 2022 WL 12129643, at *6. The panel majority concluded the district court focused "almost entirely" on "its discontent with the CVSA and its attendant discussions, rather than adhering to its obligation to conduct a full and fair assessment based on the totality of the circumstances." 2022 WL 12129643, at *6. Judge Hurst concurred but wrote separately. 2022 WL 12129643, at *6-11 (Hurst, J., concurring). We granted Garrett's petition for review. ANALYSIS

Garrett argues the Court of Appeals erroneously reversed the district court's suppression by reweighing evidence and considering the officers' deceptive tactics in isolation rather than together with other coercive factors. He contends that, without these errors, the totality of the circumstances shows his statements were involuntary.

Standard of Review

When reviewing a district court's suppression order, an appellate court reviews the district court's "findings about historical facts regarding the circumstances of the confession as issues of fact"; thus, such findings "about these factors must be supported

6 by substantial competent evidence or, in other words, evidence that a reasonable person could accept as adequate to support a conclusion." State v. G.O., 318 Kan. 386, 407, 543 P.3d 1096 (2024). In making this determination, an appellate court "does not reweigh the evidence, assess witness credibility, or resolve evidentiary conflicts" and disregards "any conflicting evidence or other inferences that might be drawn from the evidence." G.O., 318 Kan. at 407. After assessing the evidentiary sufficiency of the district court's findings, an appellate court then reviews the district court's ultimate legal conclusion de novo. State v. Palacio, 309 Kan. 1075, 1081, 442 P.3d 466 (2019). As we recently explained, this involves our consideration of

"whether the state actor overreached, the determination of how the accused reacted to the external facts, and the legal significance of the reaction as issues of law. We examine the totality of circumstances and assess de novo the trial judge's legal conclusion based on those facts. This means we give no deference to the trial judge's legal conclusion that [the accused] did not voluntarily confess." G.O., 318 Kan. at 407.

Garrett argues the Court of Appeals erroneously reweighed the evidence and incorrectly assessed the legal effect of some of the district court's factual findings. Both arguments present legal questions subject to unlimited review. See State v. Neighbors, 299 Kan. 234, 240, 328 P.3d 1081 (2014); Palacio, 309 Kan. at 1081.

Discussion

The Fifth Amendment to the United States Constitution, made applicable to the states through the Fourteenth Amendment Due Process Clause, prohibits the State from compelling anyone "in any criminal case to be a witness against himself." This protection bars the State from relying on coerced or involuntary statements to establish a defendant's guilt. Palacio, 309 Kan. at 1087. A statement can be involuntary even if officers read a defendant their Miranda rights and the defendant waived a right to counsel. 309 Kan. at 1087.

7 Due process protects against involuntary confessions caused by coercive police tactics. These tactics fall into two broad categories: "(1) Those that are inherently coercive and a per se violation of the Due Process Clause and (2) those where a state actor uses interrogation techniques that because of the unique circumstances of the suspect are coercive." G.O., 318 Kan. at 397. The former group includes "interrogation techniques that in isolation are inherently offensive to a civilized system of justice" and usually involve "coercive techniques that included extreme psychological pressure or brutal beatings and other physical harm." 318 Kan. at 397-98. While Garrett claims the police's tactics were coercive, he does not argue they were of the sort that is "inherently offensive to a civilized system of justice."

When confronted, as here, with a confession alleged to have been caused by law enforcement tactics that were coercive (and thus misconduct) "because of the unique circumstances of the suspect," we consider the totality of the circumstances, including circumstances relevant to both law enforcement and the accused, to determine first whether the law enforcement tactics used in this instance constituted overreaching misconduct. In the absence of the State abusing its power, a confession does not violate due process. If such misconduct is found, appellate courts then must undertake a causal analysis to determine whether the misconduct resulted in the challenged confession. G.O., 318 Kan. at 398. That causal analysis is necessary if misconduct is found because the mere presence of police misconduct connected to a confession is not enough to require suppression. The misconduct must cause the defendant's free will to be overborne, such that the resulting confession is not voluntary. When that happens, law enforcement has violated due process and it is appropriate to suppress the confession for that violation. See G.O., 318 Kan. at 400. The State bears the burden of proving the voluntariness of a defendant's confession by a preponderance of the evidence. 318 Kan. 403-04 (citing State v. Brown, 286 Kan. 170, 172, 182 P.3d 1205 [2008]).

8 So what are the characteristics of law enforcement tactics and the accused to be considered here? Because we have described these characteristics, or factors, as nonexclusive, "any relevant factor may—and should—be considered." G.O., 318 Kan. at 402. Still, we have previously identified several such factors:

"Potential details of the interrogation that may be relevant include: the length of the interview; the accused's ability to communicate with the outside world; any delay in arraignment; the length of custody; the general conditions under which the statement took place; any physical or psychological pressure brought to bear on the accused; the officer's fairness in conducting the interview, including any promises of benefit, inducements, threats, methods, or strategies used to coerce or compel a response; whether an officer informed the accused of the right to counsel and right against self-incrimination through the Miranda advisory; and whether the officer negated or otherwise failed to honor the accused's Fifth Amendment rights.

"Potential characteristics of the accused that may be relevant when determining whether the officer's conduct resulted in an involuntary waiver of constitutional rights include the accused's age; maturity; intellect; education; fluency in English; physical, mental, and emotional condition; and experience, including experience with law enforcement." G.O., 318 Kan. at 403.

When evaluating misconduct, we consider the district court's findings about the interview itself and those findings about the defendant that would have been known to (or ascertainable by) law enforcement; only when evaluating voluntariness overall do we consider the factors which law enforcement would have had no way of knowing, such as a defendant's experience or subjective feelings.

Our task on review is thus akin to solving a jigsaw puzzle: first sorting out the district court's relevant findings of fact supported by sufficient evidence and then fitting those facts together to assess, as a matter of law, whether the final picture produced by those findings reveals police misconduct. If the answer is yes, we then proceed to

9 determining whether the facts demonstrate, as a matter of law, that the misconduct caused an accused person's will to be overborne, rendering the confession involuntary and inadmissible for violating due process.

Substantial competent evidence supports the district court's findings of fact.

On January 27, 2020, the district court held an evidentiary hearing on the motion to suppress. At the conclusion of the hearing, the court made the following findings:

• Garrett was brought in for questioning at the Salina Police Department, behind one set of locked doors. The interrogation began at around 1:20 p.m. • Detective Brown, Detective Jones, and other law enforcement were present throughout the interview. • Garrett was provided with his Miranda warning at the outset. There was some minimization of that process by the detective, who referred to the warning as "hoops that he needed to jump through" because Garrett was arguably in custody. • Soon it became apparent to Garrett that the interview was about his inappropriate touching and behavior toward L.A. • The officers were very fair as far as their tone and demeanor, and were not in any way coercive in an outright confrontational sense. • Garrett's mental condition was stressed; however, he was oriented to time, place, and circumstance. • Garrett was able to communicate with the outside world. • The duration of the interrogation was short. • Garrett is fluent in English. • Garrett is articulate, and of average or above average intellect. • At the time of the interview, Garrett's age was close to 40.

10 • The officer "oversold" the voice stress test. It was presented as 100 percent effective repeatedly, in the context of Garrett denying any inappropriate touching. The test was presented as a foolproof truth confirmation test. Nothing in the literature would indicate that it is 100 percent effective. To assert as much was a deceptive practice. • Sergeant Cox indicated the CVSA was 100 percent effective for truth confirmation and that the test could differentiate between the stress of the event and the stress of deception. The literature before the court indicated that was not accurate. • After being confronted with the result of the CVSA, Garrett changed his answers and started disclosing statements that incriminated him. • Garrett was not threatened. • No promises were made to Garrett concerning action that would be taken by a public official in reference to a deal or some favorable treatment.

On July 30, 2021, the district court held a hearing on its own motion to reconsider the motion to suppress. No additional evidence was presented. The court stated it did not intend to repeat every finding previously made, and incorporated its findings from January 27, 2020, but also supplemented those with additional findings. The additional findings were to replace any contrary findings from the earlier hearing. The supplemental findings were as follows:

• Garrett's interview was several hours in length, including a significant period the defendant spent alone or isolated. • The interview was conducted by three different officers at varying points. • Garrett was in his early 30s at the time and informed officers that he had not eaten or slept and that he was upset. At the time, Garrett worked at Dillons as a backup meat cutter and manager. • The questioning before the CVSA was nonconfrontational and less direct.

11 • Detectives told Garrett that nervousness was not part of the equation, that they thought it was "'a fantastic tool, an excellent tool[,]'" or words to that effect. • Sergeant Cox then arrived and explained the CVSA, and stated the CVSA is 100 percent effective, despite Garrett's stress. • Detectives asked Garrett if he would agree to take the test, and Garrett acquiesced. • The CVSA exam occurred in a separate room with Garrett and Cox present. Cox informed Garrett the exam would be recorded and he could stop it at any time. • Garrett was trying to read a form Cox gave him and he could not make out the word "coercion." Cox pronounced it for him and then he understood it. • Cox again administered the Miranda warnings to Garrett. Garrett was again told he could stop the exam at any time. • Cox explained they formerly used a polygraph exam but now use the CVSA. Cox said the CVSA is used throughout the United States, and even the military uses it. She explained it is a truth verification exam and it is 100 percent effective. Cox said by the time they were done they would know what happened and what the truth was. • Cox told Garrett she wanted him to pass the test and remember to be completely honest. Before administering the CVSA, she reviewed with Garrett the specific allegations of how the defendant touched L.A. Garrett denied the allegations. • Cox then administered the CVSA. • After the CVSA, law enforcement was not persuaded by Garrett's continued denials and told him they believed he touched L.A. as alleged. • Garrett was twice given Miranda warnings before testing and did not assert he wanted counsel or wished to remain silent. • Law enforcement's use of the "Reid Technique" to minimize and obstruct claims of innocence were egregious.

12 We conclude the district court's findings of fact are supported by substantial competent evidence. (While there is some discrepancy in whether Garrett was "in his early thirties" or "closer to forty[,]" both findings describe Garrett as a mature adult.)

The detectives did not overreach.

Having found that the district court's findings were supported by substantial competent evidence, we next turn to its legal conclusions: that under the totality of the circumstances there was overreaching law enforcement misconduct that caused Garrett's will to be overborne, such that his confession was involuntary. We review these aspects of the district court's decision de novo. In doing so, we note the district court, the Court of Appeals, and the parties did not have the benefit of our decision in G.O. clarifying several of the legal principles relevant to the proper legal analysis in this case.

Before the district court, Garrett focused on the officers' fairness in conducting the interrogation. He argued the officers unfairly coerced him into confessing by downplaying the significance of the interrogation and the Miranda warnings, misrepresenting the accuracy and admissibility of the CVSA exam, encouraging him to confess so the prosecutor would look favorably upon him, and utilizing an interrogation tactic called the Reid Technique by minimizing the serious nature of the crime and offering innocent explanations after asserting that the CVSA proved guilt.

After discussing the officers' unfairness in more detail at the second hearing, the district court explained its new ruling:

"The Court has reconsidered the totality of the circumstances. Clearly, the defendant was under stress. He reported not sleeping or eating. He had difficulty reading the word 'coercion,' had to be explained to him. Law enforcement minimized the need for Miranda. Delayed telling him the specific nature of the allegations. And what troubles

13 the Court most significantly is law enforcement deliberately misled the defendant regarding the effectiveness of the CVSA. As I mentioned previously, they oversold it as 100% effective and a way for the defendant to move past this.

"The Court understands that law enforcement are allowed to use misleading tactics at times during an investigation; however, the overselling and application of the CVSA process and the post CVSA interview tactics are a bridge too far in these circumstances.

....

"Considering the totality of all the circumstances, including the pre-CVSA tactics and approach, the process used during the CVSA and the post-CVSA tactics and approach, the Court finds the defendant's statements after the CVSA were not voluntary, they were not the product of his free and voluntary will, and the Court is suppressing those statements."

The Court of Appeals rejected what it perceived as the district court's "hard stop against deceptive interview techniques in general." Garrett, 2022 WL 12129643, at *5. It then appeared to disagree with the district court's finding that Garrett was stressed and tired and had not eaten: "As the testing period neared, Garrett repeatedly made his current level of anxiety known, explaining that he was 'hyped up' and felt nervous and stressed arguably in a thinly veiled attempt to offer an innocent explanation for any stress registered by the test." 2022 WL 12129643, at *5. The majority described Garrett's mental condition as "stable." 2022 WL 12129643, at *6.

The panel concluded:

"In our view, the district court entered its second ruling almost entirely as a product of its discontent with the CVSA and its attendant discussions, rather than adhering to its obligation to conduct a full and fair assessment based on the totality of the circumstances. Shining a light on those factors reveals that Garrett's mental condition was

14 stable and he was not subject to undue duress. While he admitted to experiencing stress, the evidence reflects that anxiety started to simmer the night before his interview after reviewing a text on his wife's phone from L.A.'s father about this matter. The manner and duration of Garrett's interview was also nonremarkable. The evidence reflects it started at roughly 1:30 in the afternoon and lasted only about 2 hours, which we find to be a reasonable length of time. While the record is absent any facts addressing whether Garrett sought to communicate with the outside world at the time, there is no evidence suggesting such a request was made and denied. Finally, turning to Garrett's age, intellect, and background. The evidence adduced demonstrated that Garrett was about 40 years old, married with a family, and worked in two capacities at a local store with one of those roles carrying managerial responsibilities. Accordingly, there is no evidence tending to show Garrett lacked the intellectual capabilities to appreciate his circumstances or what was being asked of him.

"The undue weight the district court afforded the deceptive techniques, to the exclusion of nearly all other relevant components of the inquiry, gave rise to a finding of involuntariness that is neither grounded in substantial competent evidence nor consistent with the longstanding law in this area. Accordingly, that decision cannot be permitted to stand. The district court's conclusion that Garrett's statements were involuntary and its suppression of the same is reversed." Garrett, 2022 WL 12129643, at *6.

Garrett argues the panel erred by reweighing the evidence and considering the legal effect of the relevant factors in isolation, rather than assessing the cumulative effect of all the relevant factors. As to the evidence considered by this court, we look to the findings as made by the district court. And we consider those findings both specifically and then as part of the overall circumstances to determine whether the law enforcement tactics constituted overreaching misconduct.

We believe the dissents' criticism that this review constitutes a "divide and conquer" approach is misplaced. Slip op. at 30. Consideration of the cumulative effect of the totality of the circumstances does not require that we neglect to examine each

15 circumstance. Rather, a close examination of the circumstances provides greater understanding of their total effect.

When considering all the circumstances related to both (a) law enforcement tactics and (b) what law enforcement knew about Garrett, we conclude that, as a matter of law, the police did not overreach. True, they exaggerated the CVSA's ability to detect truthfulness, which the district court found deceptive. And the district court found it "egregious" that law enforcement minimized and suggested justifications for Garrett's actions, using the Reid Technique.

Even so, law enforcement also understood Garrett was a grown man of apparently average intelligence who was fluent in English. The duration of the interrogation was not prolonged; there was no evidence Garrett was denied any request to communicate with the outside world. Garrett highlights his stress and tiredness, but he cites no caselaw indicating that either necessarily results in involuntary confessions. Appropriate law enforcement interrogation has never required a stress-free environment. Difficult allegations require difficult questions that cause stress. As we have previously explained, "A statement is not involuntary simply because a defendant was tired . . . the condition must have made the defendant seem confused, unable to understand, unable to remember what had occurred, or otherwise unable to knowingly and voluntarily waive the right to remain silent." State v. Galloway, 311 Kan. 238, 246, 459 P.3d 195 (2020). The district court made no such findings here, regardless of its observation that Garrett reported not eating or sleeping and was, understandably, under "stress."

Further, Garrett argues Detective Brown's statement that he wanted to tell the prosecutor that Garrett had cooperated also amplified the coercive nature of the interrogation. Admittedly, a promise of leniency can render a confession involuntary if it "concern[s] action to be taken by a public official[,] . . . would likely cause the accused to make a false statement to obtain the benefit of the promise[,]" and was made "by a person

16 whom the accused reasonably believed had the power or authority to execute it." State v. Garcia, 297 Kan. 182, 196, 301 P.3d 658 (2013). But an officer's statement that they would like to tell a prosecutor that a defendant cooperated is not a promise of leniency. State v. Johnson, 253 Kan. 75, 82, 84, 853 P.2d 34 (1993). Nor have we held such a statement to be coercive. See State v. Harris, 284 Kan. 560, 581, 162 P.3d 28 (2007) (resulting confession voluntary even though officer told defendant full cooperation would be viewed favorably); State v. Tillery, 227 Kan. 342, 344, 606 P.2d 1031 (1980) (confession voluntary when officers told defendant things would "go better" if they told the truth); State v. Harwick, 220 Kan. 572, 575, 552 P.2d 987 (1976) (confession voluntary even though officer told defendant that district attorney might be lenient). The officers' statements here were no different than those previously considered to be noncoercive. They did not promise leniency and thus were not improper.

Next, Garrett argues that the minimization of his Miranda rights at the outset of the interview contributed to the coercive environment. We agree that, at least in some circumstances, an officer's attempt at minimizing a defendant's rights can contribute to a coercive atmosphere that may lead to an involuntary statement. See, e.g., G.O., 318 Kan. at 407-09; Doody v. Ryan, 649 F.3d 986, 1002-06 (9th Cir. 2011); Ross v. State, 45 So. 3d 403, 434-35 (Fla. 2010), as revised on denial of reh'g (2010). But that did not happen here. Garrett was advised of his constitutional rights not once, but twice, before he made incriminating statements. While the reasons and importance of the first Miranda advisory were minimized, the second advisory repeated the ones already given and Garrett acknowledged each one. These rights included his right to stop the interview at any time (remain silent) and demand the assistance of counsel. This clearer and more forceful recitation of Garrett's rights alleviated any coercive effect that the initial reading caused.

Garrett also asserts the officers' deceptive exaggeration of the reliability of the CVSA to identify the "truth" was coercive. He insists this case is like State v. Stone, 291 Kan. 13, 29, 237 P.3d 1229 (2010). Despite certain similarities, Stone is distinguishable.

17 In Stone, the detectives falsely told the defendant they had found semen on the victim's pajama top and were sure it would match the defendant's DNA. They were also aggressive in their interrogation and implied the only thing that would keep the defendant out of jail or affect the length of the sentence was a confession. In addition, the defendant had a sore throat, an ankle injury, was suffering from exhaustion, and became confused to the point of offering garbled and disorganized responses throughout the interview and merely adopted the interrogator's suggested version of events. 291 Kan. at 22-23. In contrast, here the district court made no finding that officers were aggressive or indicated Garrett's confession would keep him out of jail or affect the length of his sentence, or that Garrett's mental state prohibited him from thinking clearly.

Sometimes deceptive practices by law enforcement constitute misconduct, but not always. The difference is a matter of degree, gauged by what the officers knew or could have ascertained about the defendant; a lie told to a child, after all, will have a far greater impact than a falsehood given to an adult. Here, nothing about Garrett himself or the other surrounding circumstances of the interrogation could have exacerbated the effect of the deception. While our threshold assessment of misconduct differs from the ultimate question of voluntariness, we note that many cases have found a voluntary confession even when presented with law enforcement's deceptive tactics. See Frazier v. Cupp, 394 U.S. 731, 737, 739, 89 S. Ct. 1420, 22 L. Ed. 2d 684 (1969) (statements voluntary even though officer falsely told defendant codefendant had confessed to his and the defendant's guilt); State v. Harris, 279 Kan. 163, 170, 105 P.3d 1258 (2005) (deceptive interrogation techniques alone do not establish coercion); State v. Swanigan, 279 Kan. 18, 32, 106 P.3d 39 (2005) (police free to lie about evidence that fingerprints were found and confirmed to be Swanigan's, but false information must be viewed as a circumstance in conjunction with others, e.g., additional police interrogation tactics); State v. Wakefield, 267 Kan. 116, 127-28, 977 P.2d 941 (1999) (questioning officer's false statement to defendant, when viewed as part of the totality of the circumstances, was insufficient to make the otherwise voluntary confession inadmissible).

18 The alleged deception here stems from the disparity between what officers represented the CVSA's accuracy to be—100 percent—and what a witness reported the scientific literature said: 15-50 percent accuracy. But we disagree with Justice Wall's dissent's claim that "[a]ll the while, police knew the testing was junk science, the results could not be used in court." Slip op. at 31. We simply do not know they knew that. No evidence supports the conclusion that the police in this case knew the CVSA test—a real test used by other law enforcement agencies—to be "junk science." Slip op. at 31. Nor do we know from the evidence that CVSA results could never be admissible in court for any reason. To the extent the discrepancy in the CVSA's real versus represented accuracy was deceptive, as the district court found it to be, we cannot conclude that it constituted misconduct as a matter of law. As Judge Hurst aptly summarized:

"[T]he deception here was not pervasive—while the interviewers extolled the accuracy of the CVSA exam, they did not heavily or repeatedly rely on its results. While the interviewers led Garrett to believe that some of his answers demonstrated stress, they did not say that the exam proved he lied or proved his guilt, and they did not belabor the exam results. They asked Garrett about the results just once or twice before changing tactics. Additionally, the interviewers did not lie about the existence of physical evidence, witnesses, or surveillance footage." 2022 WL 12129643 at *10 (Hurst, J., concurring).

The district court also criticized law enforcement's use of the Reid Technique as "egregious," so some explanation of the Reid Technique is needed. While the district court's use of the word "egregious" creates a negative connotation of these techniques, that conclusive connotation is not universally held. One secondary source more favorable to this law enforcement tactic describes it as follows:

"By virtue of its name, the Reid Technique of Interview and Interrogation may lend itself to the generalization that it teaches interrogators how to become better at eliciting confessions from suspects but no more than that. Moreover, critics . . . broadly assert that the technique is so powerful that interrogators use it to coerce suspects to

19 confess to crimes they haven't committed, yet so flawed that interrogators are unable to tell the difference between someone who is telling the truth and someone who is not. Reid's 'three-part process for solving crime,' however, is much more comprehensive than such generalizations would suggest.

"While behavior analysis and interrogation skills are the primary benefits derived from the textbook and seminars, the technique's overall structure has a system of checks and balances. To review, first there is 'factual analysis.' Prior to interviewing a suspect, interrogators are instructed to gather as much independent evidence as possible from the most reliable sources. Second is the 'behavior analysis' interview, in which investigators look for symptoms of deception, but under the admonition not to put too much weight in any one indicator. Finally, there is the 'nine-step interrogation method,' which is set up in a manner so that innocent people are likely to forcefully deny guilt as early as steps one, two, and three.

"By the time the interrogator reaches what is likely the most suggestive part of the interrogation, innocent suspects will have given many indications of truthfulness, thereby eliminating the need to move into this area. In such cases, Reid advises the interrogator to consider the process of 'stepping down.' Stepping down involves softening the intensity of the interview or terminating it completely. Which way to proceed here depends on whether the interrogator believes the person has some knowledge of the crime (as an accomplice, witness, etc.), or is completely uninvolved.

"Whether or not a confession is voluntary depends on an overall inquiry into the suspect's susceptibility to coercion as well as whether or not the police acted in a manner likely to overbear the suspects' desire not to speak." Goodman, Getting to the Truth: Analysis and Argument in Support of the Reid Technique of Interview and Interrogation, 21 Me. B.J. 20, 24-25 (2006).

Garrett submitted no evidence criticizing law enforcement's use of the Reid Technique as used here and cites no law prohibiting these techniques. Indeed, this court noted in Khalil-Alsalaami v. State, 313 Kan. 472, 507, 486 P.3d 1216 (2021), that "no

20 Kansas appellate decision had found" "'minimization'" techniques in interrogation "alone sufficient to render a defendant's confession involuntary."

But we pause to caution that the point is not whether a confession is truthful or false. The point is the process due the accused, regardless of the truth of the confession. As one federal district judge recently opined:

"While [Defendant] Monroe [criticizes] the Reid Technique . . ., there is nothing impermissible as a matter of law with this interrogation approach; it falls within the range of acceptable interrogation tactics sanctioned by the First Circuit. Monroe offers no authority, and the Court could not find any, for the contention that an agent's minimization of crimes, under these facts, renders a suspect's statements involuntary. Thus, Monroe's argument that the Reid Technique violated his Due Process rights must fail.

"The problem with this result, of course, is that it implicitly condones police interrogation tactics, such as lie detector tricks and the minimization and maximization of crimes, which, again, can lead to—or are at least present in—false confessions. Thus, the use of the Reid Technique on most competent adults is lawful until and unless it fails, and proving its failure is a herculean task to be sure. Generally, it would require overcoming a finding of guilt on a post-conviction claim of actual innocence. The solution to this problem is not to ban the Reid Technique by holding, as Defendant would have it, that its use constitutes a per se Fifth Amendment violation. But, at the same time, law enforcement agents need to consider carefully whether their tactics are appropriate in any given situation, and they should be fully trained, using real science (not company promotional propaganda), on the efficacy and frailties of various interrogation techniques.

"Indeed, all agents in the criminal justice system—prosecutors, defense attorneys, and judges—want a system that does not wrongfully convict innocent people. If law enforcement agents are led to believe incorrectly that the Reid Technique possesses a kind of special power to root out the truth—as the company's marketing material implies—they will be misled in certain cases, resulting in false confessions and

21 wrongful convictions. It is also particularly important to recognize the risk of false confessions in vulnerable populations." United States v. Monroe, 264 F. Supp. 3d 376, 392-94 (D.R.I. 2017), aff'd No. 19-1869, 2021 WL 8567708 (1st Cir. 2021) (unpublished opinion).

Here, law enforcement used both the Reid Technique and deception. We have previously held that neither is prohibited standing alone. Here, they did not stand alone, and our precedent requires that we consider their cumulative effect. Stone, 291 Kan. at 25. Even so, considering these and other tactics used by law enforcement, along with the factors relevant to Garrett as known by law enforcement, under the totality of the circumstances we conclude that law enforcement's actions did not go so far as to constitute misconduct in violation of due process. Since the tactics here were not misconduct, Garrett's resulting confession is not rendered inadmissible because of those tactics.

We affirm the Court of Appeals' reversal of the district court's suppression of Garrett's confession, albeit on different grounds. We remand the matter to the district court for further proceedings.

Judgment of the Court of Appeals reversing the district court is affirmed. Judgment of the district court is reversed and remanded.

***

ROSEN, J., dissenting: Today the majority sanctions law enforcement interrogation tactics that the district court described as "akin to a psychological rubber hose." While the majority may be unbothered, I believe that the deceptive tactics here went too far and functioned to defeat Garrett's free will. I would affirm the district court's suppression of Garrett's statements for the reasons I set out below. And I also join Justice Wall's dissent.

22 True to the majority's observation, the United States Supreme Court and this court have held that deceptive interrogation practices do not constitute a per se constitutional violation. But in each of those cases, the officers misrepresented existing physical evidence or witness' version of events. Frazier v. Cupp, 394 U.S. 731, 737, 739, 89 S. Ct. 1420, 22 L. Ed. 2d 684 (1969) (officer falsely told defendant codefendant implicated him in crime); State v. Harris, 279 Kan. 163, 170, 105 P.3d 1258 (2005) (officers lied about physical evidence and eyewitnesses); State v. Wakefield, 267 Kan. 116, 126, 977 P.2d 941 (1999) (officers lied about existence of incriminating fingerprints and witnesses).

In contrast, the officers in this case did not misrepresent the existence of physical evidence or witness observations or their cooperation with law enforcement. Their coercion was multi-leveled consisting of numerous acts of deception. The officers began the interrogation by minimizing the importance of the Miranda advisement, explaining they merely had to "jump through some hoops" before they began the interrogation— those "hoops" being Garrett's constitutional right to remain silent.

Before the officers relied on the results of the CVSA to wrench a confession out of Garrett, they had to coerce him into taking the exam. To do this, they told Garrett that the CVSA was a reliable tool used by the military that was 100% accurate and would verify that he was telling the truth when he denied the allegations against him. The implication is clear: if Garrett refused to take the CVSA and verify the truth of his statements, he was obviously lying about his innocence. In constitutional terms, Garrett's exercise of his right to remain silent would establish his guilt. This directly contradicts Fifth Amendment jurisprudence, which has long provided that a defendant's custodial silence is "insolubly ambiguous" and shall "carry no penalty." Doyle v. Ohio, 426 U.S. 610, 618, 96 S. Ct. 2240, 49 L. Ed. 2d 91 (1976). The officers also implied that his agreement to take the test would be communicated to the prosecutor as "cooperation." Garrett thus faced a reality in which his silence proved guilt—regardless of whether he was guilty—and an agreement

23 to take the CVSA might earn him better treatment from the prosecutor. Standing alone, this is highly coercive.

But the officers' coercion continued. After the exam's conclusion, the officers revealed to Garrett that he had failed the test. From Garrett's point of view, a scientific test he (and nearly everyone) has never heard of that law enforcement insists is 100% accurate had absolutely proved his guilt. Guilty or not, his only realistic option was to take the officers up on their offer to tell the prosecutor Garrett had been cooperative. Again, I agree with the district court's assessment of this situation as akin to a "psychological rubber hose."

Other courts have viewed deception regarding the results of lie detector tests to be highly coercive. In State v. Matsumoto, 145 Haw. 313, 327, 452 P.3d 310 (2019), results from the defendant's polygraph were inconclusive, but officers told the defendant he failed. The defendant confessed and the Supreme Court of Hawaii suppressed the confession. In doing so, the court explained why falsely telling a defendant that a scientific test had absolutely shown their guilt was so psychologically impactful:

"The polygraph is a scientific instrument that purports to accurately determine whether the subject of the test is telling the truth. . . . An examinee who has not lied does not expect to be given falsified polygraph test results from the police. It is thus not surprising that the presentation of falsified results may have serious and substantial effects on a suspect. '[E]xperiments have shown that . . . counterfeit test results . . . can substantially alter subjects' . . . beliefs, perceptions of other people, behaviors toward other people, emotional states, . . . self-assessments, [and] memories for observed and experienced events.' Saul M. Kassin et. al, Police-Induced Confessions: Risk Factors and Recommendations, 34 L. & Hum. Behav. 3, 17 (2010) (citing studies that have tracked the effects of counterfeit test results, along with other deceptive tactics) (internal citations omitted).

24 "Falsified polygraph results may pressure a suspect into changing the suspect's pre-test narrative. This pressure is intensified when an officer expresses confidence that the suspect is lying and is aggressive in pushing the suspect to confess on the basis of the officer's pre-formed belief of the suspect's guilt. Richard A. Leo & Richard J. Ofshe, The Truth About False Confessions and Advocacy Scholarship, 37 Crim. L. Bull. 293, 293- 370 (2001). Falsified polygraph results are geared towards making the suspect believe in one's own guilt or believing that the officer will not stop the interrogation until the suspect confesses guilt. See Klara Stephens, Misconduct and Bad Practices in False Confessions: Interrogations in the Context of Exonerations, 11 Ne. U. L. Rev. 593, 596 (2019) (finding that false polygraph results are 'bad practices' that produce both true and false confessions).

"Once a suspect believes that a confession of guilt is inevitable, the individual is cognitively geared to accept, comply with, and even approve of that outcome. Kassin et. al., supra, at 17, (citing Elliot Aronson, The Social Animal (1999)) (exploring how human beings cognitively respond once they view an outcome as inevitable). That is, false polygraph results may psychologically prime an innocent suspect to make a confession.

"Extensive scientific literature and numerous documented cases have demonstrated the coercive nature of falsified polygraph test results; they can change a suspect's beliefs, pressure a suspect to confess, and even cause the suspect to believe they committed the crime when they did not." Matsumoto, 145 Haw. at 326-27.

There is no allegation that Garrett was given falsified results from the CVSA. But, like the defendant in Matsumoto, he was falsely told that a lie detector test absolutely proved his guilt. In reality, the exam Garrett took was at best no more reliable than a coin flip.

This coercive effect of the CVSA was amplified by the highly dubious indication from the officers that the results could be used against Garrett in court. Before Garrett

25 began his test, he was given a document that informed him the exam would be recorded and could be released for purposes of testimony. I question this representation. Because polygraph results have proven to be scientifically unreliable, they are generally inadmissible in Kansas unless the parties agree to their admission. Wakefield, 267 Kan. at 133. I suspect CVSA results would fair similarly. Nonetheless, the form Garrett signed indicated his results could be used as testimony. This would have cemented Garrett's belief that he had no recourse but to give the officers the confession they wanted. At least one other court has held it was coercive for an officer to falsely tell a defendant his polygraph results would be admissible in court. See State v. Valero, 153 Idaho 910, 914, 916, 285 P.3d 1014 (2012) (confession involuntary in part because officers told defendant polygraph results were admissible in court, which is legally incorrect). And this court has hinted that it would agree. See State v. Sanders, 223 Kan. 273, 277-78, 574 P.2d 559 (1977) (use of polygraph did not render confession involuntary in part because officers did not discuss admissibility of polygraph results with defendant); see also State v. Morton, 286 Kan. 632, 652, 186 P.3d 785 (2008) ("While telling a suspect false information about the evidence against the suspect, standing alone, does not render a confession involuntary, giving the suspect false or misleading information about the law is more problematic.").

When the officers did not obtain a confession after convincing Garrett to sit for the CVSA and revealing the supposed 100% reliable results, they deployed another deceptive tactic. The officers began minimizing the nature of the alleged crimes and offering justification for their commission. Standing alone, I agree that this was not enough to render a confession involuntary, but it certainly piled on to the already highly coercive nature of the interrogation. Our Court of Appeals has described how this tactic can influence a person's free will:

"Although innocent, an individual may attribute the purported evidence against him or her to a horrible and likely uncorrectable mistake rather than to the interrogator's

26 deception. And the interrogator's categorical dismissal of each protest of innocence can cement that fear. The individual then considers the minimalized admission of guilt the interrogator has offered to be the best way out of an exceptionally bad predicament. See Kassin, 34 Law & Hum. Behav. at 14, 16-19; Gohara, A Lie for a Lie: False Confessions and the Case for Reconsidering the Legality of Deceptive Interrogation Techniques, 33 Fordham Urb. L.J. 791, 817-19 (2006); Ofshe & Leo, The Decision to Confess Falsely: Rational Choice and Irrational Action, 74 Denv. U. L. Rev. 979, 985-86 (1997)." State v. Fernandez-Torres, 50 Kan. App. 2d 1069, 1087, 337 P.3d 691 (2014).

Aggravating all this deception was Garrett's emotional state. The district court found Garrett had emotional turmoil, intense stress, and had not slept or eaten. It was noticeable because it interfered with his understanding of the CVSA consent form—he did not know the meaning of at least one term, and he attributed his confusion to his emotional and physical state.

The district court here thoroughly considered the interrogation and carefully analyzed the voluntariness of Garrett's statements. This is evident in the detailed description of the court's findings and legal conclusions and in the court's commendable decision to correct its earlier order after giving it more thought. The district court made findings supported by substantial competent evidence that: (1) Garrett was emotionally confused and volatile, to the extent it interfered with his comprehension of the CVSA consent form; (2) officers minimized the importance of his Miranda advisement to induce him into the interrogation; (3) multiple officers repeatedly misstated the reliability of CVSA testing; (4) the CVSA consent form indicated the test results could be used as evidence against Garrett; and (5) post-testing, the officers minimized the nature of the alleged crimes, offered justification for their commission, and suggested the prosecutor would view Garrett's confession favorably as a form of cooperation. Based on these findings and in light of the totality of the circumstances, like the district court, I conclude the officers' collective deceptive and coercive practices here fell too far over the line.

27 The majority acknowledges that the officers' misrepresentation of the accuracy of the CVSA was deceptive but concludes it was unproblematic because the officers' reliance on the CVSA results was not "'pervasive.'" Slip op. at 19 (quoting State v. Garrett, No. 124,329, 2022 WL 12129643, at *10 [unpublished opinion] [Hurst, J., concurring]). I do not understand this characterization or the reasoning. The officers began discussing the CVSA 30 minutes into the interview and did not abandon the topic until they had convinced Garrett to sit through the exam, administered the exam, and revealed the results over an hour later. After telling Garrett the exam showed he was being dishonest when he denied the allegations, the officers wrested a confession out of him in under 10 minutes. So, the officers may have mentioned the exam results only a few times after the exam was complete, but that was all it took for officers to get him to involuntarily waive his constitutional rights. I cannot see how this means use of the results was not pervasive or did not function to overpower Garrett's will.

I believe that the majority's rubber-stamp of the deception in this case paves the way for an onslaught of even more coercive trickery during police interrogations. Hyper- realistic digital impersonation that can be nearly impossible to debunk, or "deep fakes," as they have come to be known, are ever present. Chesney & Citron, Deep Fakes: A Looming Challenge for Privacy, Democracy, and National Security, 107 Cal. L. Rev. 1753, 1758 (2019). Advances in technology will continue to make these digital impersonations increasingly convincing. 107 Cal. L. Rev. at 1758. I fear it will not be long before law enforcement tests the limits of creating fabricated images of a detainee at the scene of the crime or artificially create other evidence in order to convince a suspect to forego their right to remain silent or cooperate with an investigation. The majority's blanket endorsement of deceptive police tactics, even in the face of a new and unfamiliar technology like the CVSA, signals this kind of highly concerning deceit is fair game.

This court has historically permitted the use of deceptive interrogation tactics. But I believe we should have drawn a line today and affirmed the district court's judgment to

28 suppress Garrett's statements. As Justice Wall adeptly points out in his separate dissent, when the circumstances are analyzed in their totality, they show Garrett's confession was involuntary. Also worth mention, it appears the Court of Appeals believed that Garrett's confession was truthful, and this may have influenced their voluntariness analysis. Garrett, 2022 WL 12129643, at *5 (opining "given the fact Garrett was able to independently provide details of the incidents, he knew participation in the test would require him to be dishonest"). To the extent this had any role, it was misguided. While coercive interview tactics have certainly resulted in false confessions—which demonstrates the psychological power of those tactics—false confessions are not the animating concern behind suppressing involuntary statements. Courts guard against coercive interrogative pressure to protect the individual's constitutional right to remain silent and to due process of law. Guilty or not, our Constitution guarantees every person these rights. The United States Supreme Court has emphasized this truth:

"Our decisions under [the Due Process Clause] have made clear that convictions following the admission into evidence of confessions which are involuntary, i.e., the product of coercion, either physical or psychological, cannot stand. This is so not because such confessions are unlikely to be true but because the methods used to extract them offend an underlying principle in the enforcement of our criminal law: that ours is an accusatorial and not an inquisitorial system—a system in which the State must establish guilt by evidence independently and freely secured and may not by coercion prove its charge against an accused out of his own mouth." Rogers v. Richmond, 365 U.S. 534, 540-41, 81 S. Ct. 735, 5 L. Ed. 2d 760 (1961).

I would remember this today, and for the reasons stated hold that Garrett's confession was involuntary. I would affirm the district court's judgment to suppress his statements.

WALL, J., joins the foregoing dissenting opinion.

29 ***

WALL, J., dissenting: I join Justice Rosen's dissent in its entirety. I write separately to critique the majority's application of the controlling legal standard for voluntariness.

Under that legal standard, Garrett's custodial statements must be suppressed unless the State proves by a preponderance of evidence that they were voluntary under the totality of the circumstances. State v. Spencer, 317 Kan. 295, 297, 527 P.3d 921 (2023). The majority opinion fails to analyze voluntariness under this standard.

Instead, the majority uses a clever analytical device—divide and conquer. It isolates each circumstance that contributed to the environment of coercion. Then, it points to caselaw suggesting each circumstance falls short of coercion on its own. State v. Garrett, 319 Kan. ___, slip op. at 13-22. The problem with this approach is that the Constitution requires us to consider the forest, not each tree. And when we do, the State's overreaching is apparent.

This critique is not groundbreaking. We recognized the illegitimacy of the divide- and-conquer approach in State v. Stone, 291 Kan. 13, 237 P.3d 1229 (2010)—cited by the majority. There, the district court reviewed each of the three alleged deceptive practices in isolation. It then cited caselaw to support its conclusion that each factor, alone, did not render the defendant's statements involuntary. 291 Kan. at 23. Stone held the district court erred by "failing to look at the circumstances of the interrogation in totality." 291 Kan. at 29. Even if each circumstance fell short on its own, the coercive environment became evident upon "a review of . . . all of these circumstances, as the law requires." 291 Kan. at 32-33.

30 Since Stone, Kansas appellate courts have "specifically rejected a divide-and- conquer approach to assessing the involuntariness of a confession." State v. Fernandez- Torres, 50 Kan. App. 2d 1069, 1092, 337 P.3d 691 (2014). Instead, the circumstances must be analyzed collectively. 50 Kan. App. 2d at 1092.

And when the court applies the legal standard correctly, it is often outcome determinative. Take State v. Swanigan, 279 Kan. 18, 106 P.3d 39 (2005), for instance. There, the defendant argued the totality of the circumstances rendered his statements involuntary. Those circumstances included detectives lying about the evidence and threatening to tell the prosecutor about the defendant's lack of cooperation. Swanigan held that "[a]lthough any one of these factors . . . may not be sufficient to show coercion, the combination of all of them" does. 279 Kan. at 39; see also Stone, 291 Kan. at 32-33. The same holds true here.

Garrett was sleep-deprived when he was summoned to police headquarters for custodial interrogation. From the start, he expressed confusion with the written advisement. But police diminished the importance of his constitutional right to silence and leaned on him to submit to their truth-verification technology. They repeatedly told Garrett this technology discerns truth from falsehood with 100% accuracy. And that the reliability of testing is not affected by other variables like intoxication or anxiety. The consent form for testing also suggested that the results could be used against him in court. All the while, police knew the testing was junk science, the results could not be used in court, and that Garrett's Fifth Amendment rights were not "hoops" to jump through. In fact, the district court was particularly troubled that "law enforcement deliberately misled the defendant regarding the effectiveness of the CVSA."

This conduct put Garrett in an untenable situation. He could assert his constitutional rights or roll the dice and submit to testing. If he chose the former, this would have been viewed as an admission of guilt by silence given law enforcement's

31 misrepresentations about the test's accuracy. So Garrett chose his only path to exoneration and submitted to testing.

Police later told Garrett he had "failed." And based on the misrepresentations in the consent form, he had reason to believe that evidence would come in at trial. Even so, the police overbore Garrett's will only after they continued to minimize the seriousness of the alleged conduct and imply that "cooperation" might encourage the prosecutor to be lenient.

This story is not conveyed through the majority's divide-and-conquer analysis. By focusing on parts of the story in isolation, the true nature of overreaching is skewed and diminished. And that is problematic here because the totality of the circumstances yields coercion greater than the sum of its parts. But the device serves the majority well. How else could the story of Garrett's interrogation be characterized as a proper exercise of State power in a civilized system of justice?

ROSEN, J., joins the foregoing dissenting opinion.